January 2012
Access high quality clinical content written by international experts
January 2012 volume 9 no. 1
January 2012 volume 8 no. 1
How best to train postgraduates
and residents in robotic techniques?
January 2012 volume 8 no. 1
Molecular biomarkers
in bladder cancer
Stratifying response to targeted therapies
nrurol_OFC_JAN12.indd 1
Mucosal pathogenesis and new concepts
in the role of mucosal inflammation
Clinical applications of MRI
in Crohn’s disease
January 2012 volume 8 no. 1
13/12/2011 10:31
Evaluating the extent of intestinal damage
nrgastro_OFC_JAN12.indd 1
07/12/2011 16:34
Influence on cardiovascular risk
parameters in RA
Metabolic regulation of puberty
and fertility
Cellular and molecular mechanisms
of fibrosis
nrrheum_OFC_JAN12.indd 1
08/12/2011 11:47
January 2012 volume 8 no. 1
An official publication of
Neuroendocrine neoplasms
of the gut and pancreas
Knowledge, advances and controversies
Relevance and potential
applications in kidney disease
Diagnostic and treatment options
nrneph_OFC_JAN12.indd 1
28/11/2011 15:31
January 2012 volume 9 no. 1
Using breast cancer gene
signatures in the clinic
Systemic sclerosis
nrendo_OFC_JAN12.indd 1
nrclinonc_OFC_JAN12.indd 1
January 2012 volume 9 no. 1
January 2012 volume 9 no. 1
New recommendations for MRI-based
monitoring of immunomodulation
Time for a rethink to avoid overtreatment
of older men and those at low risk
07/12/2011 15:55
nrneurol_OFC_JAN12.indd 1
Role of the autonomic
nervous system
nrcardio_OFC_JAN12.indd 1
Pathogenesis and diagnosis
02/12/2011 12:49
An official publication of
Acute kidney injury
in multiple myeloma
Postural tachycardia syndrome
Characteristics, neuroepidemiology
and pathophysiological mechanisms
13/12/2011 11:42
Transcatheter aortic
valve implantation
Valves and approaches
24/11/2011 15:32
• Junior doctors keep up-to-date with Research Highlights
• Busy clinicians appreciate “Key point” summaries
• Medical Researchers value in-depth Reviews
The January issues are freely available on our website:
Follow us.
Download the free app for your
phone at
Get the free mobile app for your phone
http:/ /
design: annette fenner & nicola hawes
Cover image: Hippocrates of Cos
reflected in the style of the Roman
god, Janus. Based on an engraving
by Peter Paul Rubens (1638) from website.
Key Advances in Medicine
he articles included in Nature Reviews Key Advances in
Medicine were originally published in the February 2012
issues of the eight clinical Nature Reviews journals. The journals’
editors commissioned international experts to write a short
essay highlighting up to five key papers that made the biggest
contribution to their field in 2011. Between them, the eight clinical
Nature Reviews journals published 43 articles, which are collated
in this eBook; if you choose to cite an article, please use the
original journal citation rather than citing the eBook.
We hope you enjoy reading Nature Reviews Key Advances in
Medicine. If you would like to find out more about the Nature
Reviews series, please visit:
Don’t miss out on next year’s Key Advances in Medicine
Sign up to free e-ALERTS and be notified when next year’s Key Advances in Medicine eBook
is published. You won’t miss out on Nature Reviews collections and Focus issues either!
COPYRIGHT © 2012 Nature Publishing Group.
No part of this publication may be
reproduced, stored in a retrieval system,
or transmitted in any form (electronic or
otherwise) without prior permission from the
copyright holder.
for internal or personal use, or internal or
personal use of specific clients, is granted
by Nature Publishing Group to libraries
and others registered with the Copyright
Clearance Center (CCC) Transactional
Reporting Service, provided the relevant
copyright fee is paid to CCC, 222 Rosewood
Drive, Danvers, MA 01923, USA.
DISCLAIMER: Although every effort is made
by the publishers to see that no inaccurate
or misleading data, opinions or statements
appear in this collection, they wish to make
it clear that the data and opinions appearing
in articles and advertisements herein are the
responsibility of the contributor or advertiser
concerned. The journal does include the
personal opinions of the authors; therefore,
it is not intended to be relied on solely as
a guide to good practice or safe treatment.
Accordingly, the publishers, employees,
offices and agents accept no liability
whatsoever for the consequences of any such
inaccurate or misleading data, opinion or
statement. Although every effort is made to
ensure that drug doses and other quantities
are presented accurately, readers are advised
that the new methods and techniques
involving drug usage and described within
this journal should only be followed in
conjunction with the drug manufacturer’s own
published literature.
Nature Reviews Cardiology
S1 acute coronary syndromes | Walking the tightrope between efficacy and bleeding
Payal Kohli and Christopher P. Cannon
S3 atrial fibrillation | Stroke prevention in AF
Gregory Y. H. Lip
S5 heart failure | Heart failure therapy—technology to the fore
John J. V. McMurray
S7 hypertension | New insights—from risk factors to treatment implications
George L. Bakris
S9 valvular disease | Breakthrough for intervention?
Volkmar Falk
Nature Reviews Clinical Oncology
S11 ovarian cancer | Mutations and non-inferiority analyses show a way forward
Maurie Markman
S12 prostate cancer | Hitting old targets better and identifying new targets
Yu Chen and Howard I. Scher
S14 hematological cancer | New therapeutic targets and treatment strategies
Paula Cramer and Michael Hallek
S16 melanoma | A new paradigm tumor for drug development
Alexander M. M. Eggermont and Caroline Robert
S18 bone cancer | Prevention and treatment of bone metastases
Robert E. Coleman
Nature Reviews Endocrinology
Nature Reviews Neurology
S21 thyroid disease in pregnancy | Thyroid function—effects on
mother and baby unraveled
Anthony P. Weetman
S55 stroke | Major advances across the spectrum of stroke care
Lee H. Schwamm
S22 primary aldosteronism | Towards a better understanding
of causation and consequences
Michael Stowasser
S24 polycystic ovary syndrome | Genes, aging and sleep apnea
in polycystic ovary syndrome
Andrea Dunaif
S26 epigenetics and metabolism | Epigenetics, the life-course
and metabolic disease
Peter D. Gluckman
S28 osteoporosis | Osteoporosis therapy—dawn of the postbisphosphonate era
Roland Baron
S57 movement disorders | Translating new research findings into
clinical practice
Christine Klein and Dimitri Krainc
S58 multiple sclerosis | Advances in therapy, imaging and risk factors
in MS
Bianca Weinstock-Guttman and Murali Ramanathan
S60 dementia | Microbleeds in dementia—singing a different ARIA
Philip Scheltens and Jeroen D. C. Goos
S62 epilepsy | Insights into epilepsy treatments and biomarkers
Fernando Cendes
S30 type 1 diabetes mellitus | Heterogeneity of T1DM raises questions
for therapy
Paolo Pozzilli
Nature Reviews Gastroenterology & Hepatology
Nature Reviews Rheumatology
S33 hepatitis c | A new standard of care and the race towards
IFN-free therapy
Wolf Peter Hofmann and Stefan Zeuzem
S65 rheumatoid arthritis | Advances in diagnosis, treatment
and definition of remission
Gerd R. Burmester
S35 hepatocellular carcinoma | Genomics in hepatocellular
carcinoma—a big step forward
Ryosuke Tateishi and Masao Omata
S67 juvenile idiopathic arthritis | New takes on categorization
and treatment
Alberto Martini
S36 ibd | Advances in IBD management—towards a tailored approach
Guillame P. Pineton de Chambrun and William J. Sandborn
S68 systemic lupus erythematosus | Deciphering the role of NETs
and networks in SLE
Thomas Dörner
S38 the gut microbiota | Translating the microbiota to medicine
Fergus Shanahan
S40 neurogastroenterology | Emerging concepts in
neurogastroenterology and motility
Keith A. Sharkey and Gary M. Mawe
S70 osteoarthritis | Age-related OA—a concept emerging from infancy?
Thomas Aigner and Wiltrud Richter
S72 systemic sclerosis | From mechanisms to medicines
Luc Mouthon
S74 vasculitis | The renaissance of granulomatous inflammation in AAV
Stephan D. Gadola and Wolfgang L. Gross
Nature Reviews Nephrology
Nature Reviews Urology
S43 glomerular disease | New clues to environmental influences
in glomerular disease
Peter J. Nelson and Charles E. Alpers
S77 prostate cancer | Redefining the therapeutic landscape
for CRPC
Carmel Pezaro and Gerhardt Attard
S44 polycystic kidney disease | Connecting the dots toward a
polycystic kidney disease therapy
Vicente E. Torres and Peter C. Harris
S79 bladder cancer | The dawn of personalized medicine
Thomas W. Flaig and Dan Theodorescu
S46 acute kidney injury | Biomarkers are transforming our
understanding of AKI
Lakhmir S. Chawla and John A. Kellum
S48 nondialysis chronic kidney disease | Progression, prediction,
populations and possibilities
Adeera Levin
S50 dialysis | Can cardiovascular risk in dialysis patients be decreased?
Peter Stenvinkel and Peter Bárány
S52 transplantation | New agents, new ideas and new hope
Titte R. Srinivas and Bruce Kaplan
S80 sexual dysfunction | Advances in epidemiology, pathophysiology
and treatment
Eric Chung and Gerald B. Brock
S82 male factor infertility | Semen quality, sperm selection
and hematospermia
Amichai Kilchevsky and Stanton Honig
S84 kidney cancer | Objectifying risk for localized renal masses
Marc C. Smaldone and Robert G. Uzzo
Walking the tightrope between efficacy
and bleeding
Payal Kohli and Christopher P. Cannon
Major advances in the diagnosis of acute coronary syndromes (ACS) have occurred in 2011, but physicians
treating ACS still walk the tightrope between efficacy and bleeding. Key publications have shed light on this
delicate balance and heralded a new era of novel oral anticoagulants for the treatment of ACS.
Kohli, P. & Cannon, C. P. Nat. Rev. Cardiol. 9, 69–71 (2012); published online 20 December 2011; doi:10.1038/nrcardio.2011.206
The year 2011 has witnessed an evolution in
all aspects of the management of acute coronary syndromes (ACS). The intro­duction of
new-generation troponin assays has compli­
cated the diagnostic dilemma in ACS by
increasing the number false positive diagnoses. One of the most-interesting studies
of 2011 addressed whether the reduced
threshold for detection of myocardial injury
translated into improved clinical outcomes.
Mills et al. undertook a prospective study of
>1,000 patients who presented with possible
ACS before and after the introduction of a
lower diagnostic threshold (from 0.20 ng/ml
to 0.05 ng/ml) for myocardial infarction (MI)
using a high-sensitivity troponin assay with
a low coefficient of variability (<10%).1 The
study demonstrated that the lower threshold resulted in a 29% increase in detection
of MI. Among patients with small increases
in troponin (0.05–0.19 ng/ml), this diagnostic reclassification was also associated with
improved risk stratification and better use of
evidence-based therapies. Most importantly,
the rate of death or recurrent MI was 52%
lower (P = 0.01) during the period in which
the lower troponin thres­hold was used.1
Therefore, we can surmise that changing the
diagnostic threshold for myocardial injury
has identified a new, previously misclassified
high-risk patient population that was and has
led to improvements in clinical outcomes.
Individualized medicine and tailored
treatments have continued to be a dominant theme of this decade. The ELEVATETIMI 562 and GRAVITAS3 studies tested the
effects of high-dose clopidogrel on platelet
reactivity and clinical outcomes, respectively. In ELEVATE-TIMI 56,2 hetero­zygote
carriers of the CYP2C19*2 allele needed
225 mg of clopidogrel to achieve the same
KEY ADVANCES IN MEDICINE degree of platelet inhibition as noncarriers
receiving 75 mg, suggesting that genotype
could be important for choosing not only
the type of antiplatelet agent, but also the
dose. Homozygote carriers of CYP2C19*2
in ELEVATE-TIMI 56 were highly resistant
to clopidogrel and maintained high degrees
of platelet reactivity, even with 300 mg of the
drug. Whereas ELEVATE-TIMI 56 confirmed
the importance of pharmaco­genomic interactions in platelet reactivity, the GRAVITAS
trial3 bridged the gap between laboratory
findings and clinical outcomes and answered
the question of whether increasing doses
of clopidogrel would improve clinical outcomes. The investi­gators randomly assigned
2,214 patients with high on-­treatment platelet reactivity to high-dose (600 mg loading,
150 mg maintenance) or standard-dose
(75 mg maintenance alone) clopidogrel.
Key advances
■■ Use of a sensitive troponin assay with
a lower threshold for diagnosis of MI
reduced death and recurrent MI in high-risk
■■ Increased doses of clopidogrel are
necessary to achieve platelet inhibition in
CYP2C19*2 heterozygotes 2
■■ Apixaban 5 mg twice daily increases
bleeding without decreasing the risk of
cardiovascular death, MI, and stroke in
patients with ACS5
■■ Adding very low dose rivaroxaban to dual
antiplatelet therapy reduces cardiovascular
death, MI, and stroke with no increase
in fatal hemorrhage, despite a dosedependent increase in major bleeding6
■■ A bleeding risk score derived and validated
on the basis of baseline characteristics
demonstrated effective risk stratification
for major bleeding in patients with ACS8
Despite a 22% absolute reduction in high
on-treatment platelet reactivity at 30 days
and 6 months with the high-dose strategy
(P <0.001), no difference was observed
between the two groups in the incidence of
cardiovascular death, MI, and stent thrombo­
sis or in the rate of severe or moderate bleeding at 6 months (although the overall number
of events in the trial was small).3 These results
emphasize that, despite variation in drug
metabolism and activity, no clinical impact
of altering the dosing regimen of clopidogrel
has yet been demon­strated. Furthermore, the
reduction in platelet reactivity with second
generation thienopyridines appears marginal as the dose is increased, and the effects
in homozygote carriers of CYP2C19*2 are
severely limited. With the introduction of
more-potent agents, including prasugrel
and ticagrelor, many of these pitfalls can
potentially be avoided.
With the emergence of novel oral anti­
coagulants that require minimal monitoring, a new paradigm for treating ACS was
ushered in—combining a low dose anticoagulant with dual antiplatelet therapy.
With novel direct factor Xa inhibitors, such
as apixaban, rivaroxaban, and darexaban,
straightforward outpatient anticoagulation
finally became a realistic option for patients
with ACS. However, the road was not
without its stumbling blocks. In a phase II
trial of darexa­b an, increases in bleeding
were demon­strated without an improvement
in efficacy.4 The phase III APPRAISE‑2 trial5
of 7,392 patients with ACS was stopped early
after an increase in major bleeding (HR 2.59,
P = 0.001) emerged for apixaban (5 mg twice
daily) without an improvement in efficacy.5
Against these headwinds, the results
of the ATLAS ACS 2‑TIMI 51 trial6 were
JANUARY 2012 | S1
15 –
Events* per 1,000 patients
P = 0.001
10 –
P = NS
P <0.001
P <0.001
P <0.001
P = 0.03
P = 0.04
P = NS
–5 –
P = NS
P = 0.002
–10 –
P = 0.02
–15 –
CVD, MI, stroke
CVD, MI, stroke, RI, UR
P = NS
TIMI major bleeding
Figure 1 | Comparison of the results of the APPRAISE‑2,5 ATLAS ACS 2‑TIMI 56,6 and TRACER7
trials. The data are for the study drug versus placebo. Bleeding was increased in all studies; only
the ATLAS ACS 2‑TIMI 56 trial demonstrated efficacy and decreased mortality (data for rivaroxaban
2.5 mg shown with modified intention-to-treat analysis). Follow-up times are as follows: APPRAISE‑2
median 8 months; ATLAS ACS 2‑TIMI 56 mean 13 months; TRACER median 16.5 months. *Events
are reported on the basis of individual end points. Abbreviations: CVD, cardiovascular death;
ICH, intracranial hemorrhage; MI, myocardial infarction; RI, recurrent ischemia with
rehospitalization; UR, urgent coronary revascularization.
particu­larly favorably received. The investi­
gators randomly assigned 15,526 patients
with ACS to rivaroxaban 2.5 mg or 5.0 mg
twice daily, or placebo on a background of
aspirin alone (stratum 1) or aspirin plus a
thienopyridine (stratum 2; 93% of patients).
Notably, the doses of rivaroxaban used were
much lower than that used for full anticoagulation, such as in atrial fibrillation (20 mg
once daily). In the pooled cohort, a highly
significant 16% relative risk reduction (8.9%
vs 10.7%, P = 0.008) in cardiovascular death,
MI, and stroke was observed for rivaroxaban, with a similar reduction in all-cause
mortality, MI, and stroke (9.2% vs 11.0%,
P = 0.006) and a 31% (2.3% vs 2.9%, P = 0.02)
reduction in stent thrombosis over a 2‑year
period. A dose-dependent increase in TIMI
major bleeding occurred (placebo: 0.6% vs
rivaroxa­ban 2.5 mg: 1.8% and rivaroxaban
5 mg: 2.4%, P <0.001), but no increase in
fatal bleeding was noted. Most importantly,
a highly significant 34% reduction in cardiovascular mortality and a 32% reduction in
total mortality was observed for rivaroxaban
2.5 mg twice daily.6 Despite its overwhelming efficacy, this trial reminded us that
bleeding continues to be a limiting factor
in studies of anticoagulant plus antiplatelet
agents and that the window of efficacy needs
to be carefully defined.
S2 | JANUARY 2012
Another major trial involved the oral
thrombin receptor antagonist vorapaxar.
The TRACER trial7 faced issues with bleeding and was terminated early. In this study,
12,944 patients with ACS were randomly
assigned to receive vorapaxar or placebo
in addition to standard care. Vorapaxar
failed to reduce cardio­vascular death, MI,
stroke, recurrent ischemia, and urgent
revasculariza­tion. This drug led to a 3.4-fold
increased risk for intracranial hemorrhage
(P <0.001) and a 35% increase in bleeding.7 The results of this trial highlighted the
ever-present hazards of increased bleeding
when adding new antithrombotic agents to
standard therapy, especially in patients who
are at increased risk for intracranial hemorrhage. A second trial (TRA 2°P-TIMI 50) of
vorapaxar in patients with stable coronary
artery disease will be reported next year; thus
the final word on this drug is still to come.
Side-by-side comparison of the three major
oral anticoagulant trials shows that only the
ATLAS ACS 2‑TIMI 56 trial met its primary
efficacy end point, reduced mortality, but all
studies demon­strated significant increases
in bleeding (Figure 1).
Given the delicate balance between efficacy
and bleeding, strategies to identify patients
at high risk, and then triaging therapies
accordingly, have generated much interest.
Mathews et al. included 90,273 patients
with ST-segment elevation or ST-segment
elevation MI from the ACTION registry–
GWTG™ to derive (72,813 patients) and
valid­ate (17,960 patients) a bleeding risk
score using only baseline characteristics.8
Twelve patient characteristics and presenting
factors emerged as predictors of major bleeding in the model. Notably, history of stroke or
transient ischemic attack, which was present
in only 8% of this cohort, did not emerge as
risk factors for major bleeding in this model,
although they have previously been identified
as risk factors for intra­cranial hemorrhage.8
The risk stratification model by Mathews
et al. could guide clinical practice with respect
to patient selection for antiplatelet and anticoagulant agents, but the extent to which it
can be generalized remains unknown. Ideally,
the score should be validated in the setting
of randomized trials, various agents, and in
high-risk and low-risk patient populations
before it is applied widely.
The year 2011 was a historic landmark
in the treatment of ACS. Major strides were
made in diagnostic technology, tailored
medical therapy, and the safe use of a very
low-dose anticoagulant. Without doubt, the
years to come will see many more trials to
better educate physicians on how to walk the
tightrope between efficacy and bleeding.
TIMI Study Group, Cardiovascular Division,
Department of Medicine, Brigham and
Women’s Hospital and Harvard Medical School,
350 Longwood Avenue, 1st floor office, Boston,
MA 02115, USA (P. Kohli, C. P. Cannon).
Correspondence to: C. P. Cannon
[email protected]
Competing interests
C. P. Cannon declares associations with the following
companies: Accumetrics, Alnylam Pharmaceuticals,
AstraZeneca, Automedics Medical Systems, BristolMyers Squibb/Sanofi, GlaxoSmithKline, Intekrin
Therapeutics, Merck, Novartis, Pfizer, and Takeda.
P. Kohli declares no competing interests.
Mills, N. L. et al. Implementation of a sensitive
troponin I assay and risk of recurrent
myocardial infarction and death in patients
with suspected acute coronary syndrome.
JAMA 305, 1210–1216 (2011).
Mega, J. L. et al. Dosing clopidogrel based on
CYP2C19 genotype and the effect on platelet
reactivity in patients with stable cardiovascular
disease. JAMA 306, 2221–2228 (2011).
Price, M. J. et al. Standard- vs high-dose
clopidogrel based on platelet function testing
after percutaneous coronary intervention:
the GRAVITAS randomized trial. JAMA 305,
1097–1105 (2011).
Steg, P. G. et al. RUBY‑1: a randomized, doubleblind, placebo-controlled trial of the safety and
tolerability of the novel oral factor Xa inhibitor
darexaban (YM150) following acute coronary
syndrome. Eur. Heart J. 32, 2541–2554 (2011).
Alexander, J. H. et al. Apixaban with antiplatelet
therapy after acute coronary syndrome. N. Engl.
J. Med. 365, 699–708 (2011).
Mega, J. L. et al. Rivaroxaban in patients with
a recent acute coronary syndrome. N. Engl.
J. Med.
Tricoci, P. et al. Thrombin-receptor antagonist
vorapaxar in acute coronary syndromes.
N. Engl. J. Med.
Mathews, R. et al. In-hospital major
bleeding during ST‑elevation and
non‑ST‑elevation myocardial infarction care:
derivation and validation of a model from
the ACTION Registry®-GWTG™.
Am. J. Cardiol. 107, 1136–1143
Stroke prevention in AF
Gregory Y. H. Lip
In 2011, key trials with oral factor Xa inhibitors in patients with atrial
fibrillation highlighted promising data on these novel anticoagulants.
Patients with ≥1 stroke risk factors can be considered for oral
anticoagulation. These novel, fixed-dose drugs are given without
monitoring, so clinicians must learn to balance stroke and bleeding risks.
Lip, G. Y. H. Nat. Rev. Cardiol. 9, 71–73 (2012); published online 20 December 2011;
The year 2011 saw the publication of three
pivotal phase III trials for two oral direct
factor Xa inhibitors, apixaban and rivaroxa­
ban,1–3 as well as important articles on bleeding risk assessment 4 and the net clinical
benefit of thromboprophylaxis.5 Novel oral
anticoagulants that are viable alternatives to
warfarin have clearly changed the landscape
of stroke prevention in patients with atrial
fibrillation (AF).6 Until recently, the recommended approach was artificially to stratify
patients with AF into low, intermediate, and
high risk strata—despite stroke risk being a
continuum—so that those classed as being
at high risk could be targeted for an inconvenient (and potentially dangerous) drug,
warfarin. Many studies, however, have shown
that the categorization of patients into low,
intermediate, or high risk strata has poor
correlation with actual warfarin prescribing, and that the predictive value of risk
schemes such as the CHADS2 score (Box 1)
to identify high-risk patients is suboptimal.7
Consequently, guidelines now recommended
the use of the CHA2DS2‑VASc score (Box 1)
to complement the older (but simpler)
CHADS2 score.
Indeed, emphasis is now directed towards
identification of ‘truly low-risk’ patients with
AF by being more inclusive (rather than
exclusive) of common risk factors for stroke,
because these patients might not need any
antithrombotic therapy. Meanwhile, patients
with ≥1 risk factor for stroke should be considered for effective stroke prevention with
oral anticoagulation, whether with (very)
well-controlled warfarin or one of the
novel agents, either an oral direct thrombin inhibitor (such as dabigatran) or an
oral direct factor Xa inhibitor (for example,
rivaroxaban or apixaban). 1 Indeed, the
CHA2DS2‑VASc score has consistently been
shown to outperform the CHADS 2 score
in identification of truly low-risk patients
and is as good as—and possibly better
than—the CHADS2 score in the identification of high-risk patients who subsequently
suffer thromboembolism.7,8
Key advances
■■ In the AVERROES trial, apixaban was
superior to aspirin for stroke prevention
in patients with atrial fibrillation (AF),
with similar rates of major bleeding and
improved tolerability1
■■ In the ARISTOTLE trial, apixaban was
superior to warfarin for prevention of
stroke and systemic embolism in patients
with AF, with significantly less major
bleeding and improved survival2
■■ In the ROCKET‑AF trial, rivaroxaban was
noninferior to warfarin for stroke prevention
in a high-risk population of patients with
AF, with similar rates of major bleeding3
■■ The HAS‑BLED score is well validated
to predict major-bleeding events in
patients receiving anticoagulation therapy,
and outperforms other bleeding risk
assessment schemes4
■■ ‘Truly low-risk’ patients with a
CHA2DS2‑VASc score of 0 do not
require thromboprophylaxis; net clinical
benefit is greatest in patients with a high
HAS‑BLED score, where reduced ischemicstroke risk outweighs the increased
intracranial-bleeding risk5
Attention has also been directed to assessment of bleeding risk. Common risk factors
for bleeding (as well as potentially correctable
risk factors, such as uncontrolled blood pressure and concomitant aspirin use in patients
receiving anticoagulation therapy) can
inform clinical decision-­making, especially
with the novel oral anticoagulants that can
come in high-dose and low-dose regimens.9
Investigators in the AVERROES trial1
studied 5,599 patients with AF and ≥1 risk
factor for stroke, and who had refused war­
farin or been deemed unsuitable for war­farin
by the investigators on the basis of the inclusion criteria. The trial was stopped early
because of the clear superiority of apixaban
over aspirin, with a 55% reduction in the
primary end point of stroke and systemic
embolism (HR 0.45, 95% CI 0.32–0.62,
P <0.001), but with no significant difference
between apixaban and aspirin for major
bleeding (HR 1.13, 95% CI 0.74–1.75, P = 0.57)
or intracranial hemorrhage. Furthermore,
fewer drug discontinuations occurred with
patients taking apixaban, which indicates that
this drug is better tolerated than aspirin. Until
now, patients who refused or were unsuitable
for warfarin were treated with aspirin; therefore, this trial will certainly change clinical
practice, especially given that the evidence
for a beneficial (and safe) role for aspirin in
patients with AF is weak.10
Researchers in the ARISTOTLE trial 2
randomly allocated 18,201 patients with AF
to receive either warfarin or apixaban, and
reported a 21% reduction in the primary
end point of stroke and systemic embolism with apixaban (HR 0.79, 95% CI 0.66–
0.95, P <0.001 for noninferiority, P = 0.01
for superi­ority), with a 31% reduction in
major bleeding (HR 0.69, 95% CI 0.60–0.80,
P <0.001) and an 11% reduction in mortality
(HR 0.89, 95% CI 0.80–0.99, P = 0.047). The
primary end point was driven by the reduction in hemorrhagic stroke (HR 0.51, 95% CI
0.35–0.75, P <0.001), with no difference in
ischemic stroke (HR 0.92, 95% CI 0.74–1.13,
P = 0.42) between apixaban and warfarin.
Another oral direct factor Xa inhibitor,
rivaroxaban, was compared with warfarin in
the ROCKET‑AF trial.3 A higher-risk cate­
gory of patients (n = 14,264, mean CHADS2
score = 3.5) was enrolled in ROCKET‑AF
compared with the ARISTOTLE and
AV E R ROE S t r i a l s ( m e an C HA D S 2
score = 2.1 in both). Also, 55% of the
ROCKET‑AF population was a secondary
prevention cohort (by contrast with 20% and
trials, respectively). Rivaroxaban was clearly
JANUARY 2012 | S3
Box 1 | Abbreviations and definitions
Congestive heart failure (1 point)
Hypertension (1 point)
Age ≥75 years (1 point)
Diabetes mellitus (1 point)
Stroke or thromboembolism (2 points)
Congestive heart failure (1 point)
Hypertension (1 point)
Age ≥75 years (2 points)
Diabetes mellitus (1 point)
Sex, female (1 point)
Vascular disease (1 point)
Age 65–74 years (1 point)
Stroke or thromboembolism (2 points)
Hypertension (uncontrolled blood pressure)
Abnormal renal or liver function
Bleeding history or predisposition
Labile international normalized ratio (INR)
Elderly (age >65 years)
Drugs concomitantly (for example,
concomitant aspirin or nonsteroidal antiinflammatory drugs, or alcohol abuse)
Net clinical benefit
[ISR with no treatment – ISR on treatment]
– 1.5*[IHR on treatment – IHR with
no treatment]
Abbreviations: IHR, intracranial-hemorrhage rate;
ISR, ischemic-stroke rate.
noninferior to warfarin for the primary end
point of stroke and systemic embolism
(HR 0.79, 95% CI 0.66–0.96, P <0.001 for
non­inferiority). In the more-conventional
and conservative intention-to-treat analysis,
rivaroxa­ban failed to achieve superiority
(HR 0.88, 95% CI 0.74–1.03, P <0.001 for
noninferiority, P = 0.12 for superiority),
although a per-protocol, on-treatment
analysis reported superiority for rivaro­
xaban (HR 0.79, 95% CI 0.65–0.95, P = 0.02).
The primary safety end point, major and
non­m ajor clinically relevant bleeding,
was not different (HR 1.03, 95% CI 0.96–
1.11, P = 0.44), although hemorrhagic
stroke, intracranial hemorrhage, and fatalbleeding events were lower with rivaroxaban
than warfarin.
Until the advent of novel oral anti­
coagulants, patients with AF who were
receiving oral anticoagulation often came
under the care of a warfarin clinic, which
was responsible for anticoagulation monitoring and doses, rather than cardiologists
per se. Now, clinicians will have to make
informed decisions on doses of these novel
drugs. Of the novel direct thrombin inhibitors, only dabigatran is approved for use for
S4 | JANUARY 2012
stroke prevention in patients with AF, and
this drug is licensed at doses of 110 mg or
150 mg twice daily (or 75 mg twice daily for
patients with severe renal failure in the USA).
Apart from the criteria of age (>80 years)
and concomitant drug use (such as verapamil), how should the dose of dabigatran be
decided on the basis of bleeding risk? To help
clinical decision-making, a simple acronym
of common risk factors for bleeding (the
HAS‑BLED score; Box 1) has been recommended in European and Canadian guidelines. Various validations of the HAS‑BLED
score were published in 2011.4,5
In a large, contemporary, anti­coagulated
population of patients with AF, the HAS‑
BLED score performed best of all the tested
bleeding-risk schemes, and more-accurately discriminated patients on the basis
of their bleeding risk (whether assessed
by the c‑­s tatistic, or estimates of Net
Reclassification Improvement and Integrated
Discrimination Improvement), with a stepwise increase in rates of major bleeding with
increasing HAS‑BLED score (P <0.0001 for
trend).4 Multivariate analyses identified concurrent aspirin use (HR 2.10), renal impairment (HR 1.98), age ≥75 years (HR 1.63),
diabetes mellitus (HR 1.47), and heart failure
or left ventricular dysfunction (HR 1.32) as
significant predictors of bleeding.
What does this mean for clinicians in
everyday practice, who have to balance
the risks of stroke and bleeding? In a large,
nationwide, ‘real-world’ cohort study from
Denmark (n = 132,372), Olesen and colleagues balanced stroke risk (assessed by
the CHADS2 and CHA2DS2‑VASc scores)
with bleeding risk (assessed using the
HAS‑BLED score), and the net clinical
benefit (Box 1) of ischemic-stroke prevention versus intracranial-­hemorrhage risk
was calculated.5 The net clinical benefit was
not positive for aspirin at any stroke-risk or
bleeding-risk strata, which led the investi­
gators to conclude that “aspirin should not
be used for thromboprophylaxis in any
patient with AF”.5 Unsurprisingly, the combination of warfarin plus aspirin did not
confer any additional benefit in preventing thromboembolism, but substantially
increased bleeding risk.
In patients treated with warfarin, the only
category where the net clinical benefit was
negative was patients with a CHA2DS2‑VASc
score of 0, which reflects their truly low-risk
status. The net clinical benefit was non-­
negative (either neutral or positive) at a
CHADS2 score ≥0 and a CHA2DS2‑VASc
score ≥1. In addition, the net clinical benefit
was clearly positive, in favor of warfarin,
in patients with increased risk of stroke or
thromboembolism, and was even greater at
high HAS‑BLED scores (≥3), which indicates
that patients at high risk of bleeding are also
at high risk of stroke or thrombo­embolism
—and that the absolute benefit of warfarin
in reducing the risk of ischemic stroke far
outweighs the smaller absolute increase in
the risk of serious bleeding. The practical
use of the HAS‑BLED score also encourages clinicians to consider bleeding risk
factors that can be corrected, such as uncontrolled blood pressure, concomitant use of
aspirin or nonsteroidal anti-­inflammatory
drugs, and time in therapeutic range if the
patient is on warfarin (the H, D, and L in
HAS‑BLED, respectively).
The year 2011 has seen great advances in
assessment of stroke and bleeding risks
in patients with AF, and exciting new data
on the novel oral anticoagulants. Prospects
are much improved for patients with AF.
University of Birmingham Centre for
Cardiovascular Sciences, City Hospital,
Dudley Road, Birmingham B18 7QH, UK
[email protected]
Competing interests
The author declares associations with the following
companies and organizations: Astellas, AstraZeneca,
Bayer, Biotronik, Boehringer Ingelheim, Bristol-Myers
Squibb/Pfizer, Daiichi‑Sankyo, focusAF, Merck,
Portola and Sanofi Aventis. See the article online for
full details of the relationships.
Connolly, S. J. et al. Apixaban in patients with
atrial fibrillation. N. Engl. J. Med. 364, 806–817
Granger, C. B. et al. Apixaban versus warfarin
in patients with atrial fibrillation. N. Engl. J. Med.
365, 981–992 (2011).
Patel, M. R. et al. Rivaroxaban versus warfarin
in nonvalvular atrial fibrillation. N. Engl. J. Med.
365, 883–891 (2011).
Lip, G. Y., Frison, L., Halperin, J. L. & Lane, D. A.
Comparative validation of a novel risk score for
predicting bleeding risk in anticoagulated
patients with atrial fibrillation: the HAS‑BLED
(Hypertension, Abnormal Renal/Liver Function,
Stroke, Bleeding History or Predisposition,
Labile INR, Elderly, Drugs/Alcohol
Concomitantly) score. J. Am. Coll. Cardiol.
57, 173–180 (2011).
Olesen, J. B. et al. Risks of thromboembolism
and bleeding with thromboprophylaxis in
patients with atrial fibrillation: a net clinical
benefit analysis using a ‘real world’ nationwide
cohort study. Thromb. Haemost. 106, 739–749
Ahrens, I., Lip, G. Y. & Peter, K. New oral
anticoagulant drugs in cardiovascular disease.
Thromb. Haemost. 104, 49–60 (2010).
Karthikeyan, G. & Eikelboom, J. W. The CHADS2
score for stroke risk stratification in atrial
fibrillation—friend or foe? Thromb. Haemost.
104, 45–48 (2010).
Lip, G. Y. Stroke in atrial fibrillation: epidemiology
and thromboprophylaxis. J. Thromb. Haemost.
9 (Suppl. 1), 344–351 (2011).
[ESC] Working Group on Thrombosis. Thromb.
Haemostat. 106, 997–1011 (2011).
10. Lip, G. Y. The role of aspirin for stroke
prevention in atrial fibrillation. Nat. Rev. Cardiol.
8, 602–606 (2011).
Lip, G. Y. H. et al. Bleeding risk assessment and
management in atrial fibrillation patients:
executive summary of a position document from
the European Heart Rhythm Association [EHRA],
endorsed by the European Society of Cardiology
Heart failure therapy—technology
to the fore
John J. V. McMurray
Studies published in 2011 in the field of heart failure have reinforced
the benefit of cardiac resynchronization therapy in patients with mild
symptoms and confirmed the value of left ventricular assist devices and
CABG surgery in selected patients. Conversely, the efficacy of nesiritide
in acute heart failure has been questioned.
McMurray, J. J. V. Nat. Rev. Cardiol. 9, 73–74 (2012); doi:10.1038/nrcardio.2011.212
trials, in which recurrent events that are frequent and of great clinical significance than
first HF events are ignored.
A report from INTERMACS 2 also re-­
inforced the impressive technological progress made with left ventricular assist devices
(LVADs). In a post-approval study required
by the FDA, outcomes in the first 169 consecutive patients receiving a HeartMate II®
(Thoratec Corporation, Pleasanton, CA,
USA) device as a bridge-to-transplant were
compared with those in 169 patients receiving different LVADs for the same indication
over the same time period. The proportion of
patients transplanted, recovered, or receiving
continuing LVAD support at 6 months was
91% in the HeartMate II® group and 80%
in the comparator group (85% and 70%,
© Newlight |
Although many interesting studies on
epidemio­logy, pathophysiology, and biomarkers in heart failure (HF) were published in 2011, my focus in this article is on
treatments that are available for use and can
make an impact in everyday practice. Five
important studies, covering device and drug
therapy, and surgery in patients with HF will
be discussed.
In August 2011, a follow-up report from
MADIT-CRT1 reinforced the impressive
effect of cardiac resynchronization therapy
(CRT) on outcomes among patients with
HF, even in those with mild symptoms. The
investigators showed that CRT plus a defibrillator, compared with a defibrillator alone,
reduced the risk of a first episode of worsening HF by 46% (P <0.001) and, importantly,
also reduced the risk of subsequent episodes by 38% (P = 0.003). Because patients
experiencing a first episode of worsen­ing
HF were at a substantially elevated risk of
a further event, the absolute reduction in
risk of subsequent episodes with CRT was
large (from 72 to 44 events per 100 patient
years of follow-up). The benefit was greatest in patients with left bundle branch block.
In addition, a 19-fold increase in mortality
risk was reported in patients experiencing a
second episode of worsening HF.1
These findings are important for
several reasons. Clearly, worsening of HF
is extremely unpleasant for patients and
usually leads to hospital admission with
major consequences (economic and otherwise) for health-care systems, and is associated with an increased risk of death.1 These
findings also highlight the limitations of the
conventional ‘time to first-event’ analysis of
respectively, at 1 year). These excellent outcomes from ‘real-world’ practice at 77 centers
in the USA are consistent with the findings
of the pivotal, randomized HeartMate II®
trial,3 which showed that this continuousflow device was superior to the comparator
pulsatile-flow device.
Monitoring devices are currently of
immense interest in HF research and the
findings of the CHAMPION study 4 were
notable. This moderate-sized trial showed
that a small, implantable hemodynamic
monitor improved outcomes in patients
with NYHA class III HF who had been
hospitali­zed for HF in the previous year
and were optimally treated. The monitor
was placed transvenously in a pulmonary
artery branch and allowed wireless non­
invasive measurement of pulmonary artery
pressure. Physicians in the control group
and patients in both groups did not have
knowledge of pressure measurements;
however, site investi­gators, the trial principal
investi­gators, and sponsor (CardioMEMS,
Atlanta, GA, USA) employees had access
to pressure measurements and these were
used to adjust therapy. Addition of nitrates
and hydralazine and adjustment of diuretic
therapy, including use of intravenous diure­
tics, were more frequent in the inter­vention
group. The primary outcome was the
rate of HF hospitaliza­tion over 6 months;
83 hospitaliza­tions occurred among 55 of
the 270 patients in the intervention group,
compared with 120 among 80 of the 280
patients in the control group (rate 0.31 vs
0.44; HR 0.70, 95% CI 0.60–0.84, P <0.0001).
In absolute terms, 12.5 fewer hospitaliza­
tion events per 100 patients occurred over
JANUARY 2012 | S5
Key advances
■■ Cardiac resynchronization therapy
reduced not only the first, but also
subsequent, episodes of worsening
heart failure (HF) in patients with systolic
dysfunction and mild symptoms1
■■ Use of the HeartMate II®(Thoratec
Corporation, Pleasanton, CA, USA)
continuous-flow left ventricular assist
device was associated with excellent
6-month and 12-month survival in a
’real-world’ registry study2
■■ Pharmacological intervention prompted
by noninvasive pulmonary artery
monitoring may reduce the risk of
HF hospitalization in patients with
moderately severe symptoms and
previous admission for HF4
■■ In selected patients with ischemic
systolic HF and a life expectancy of more
than 2 years, CABG surgery may reduce
cardiovascular mortality and morbidity6
■■ Nesiritide has a small and clinically
insignificant effect on dyspnea and no
effect on mortality or worsening HF in
patients with acute HF7
6 months in the intervention group, meaning
that eight patients needed to be treated to
prevent one hospitalization. 4 Although
these results are impressive, the trial has
been criticized because of the active involvement of the sponsor in encouraging treatment changes in the intervention group, and
whether the benefit would be as substantial
when used in ordinary practice has been
questioned. Nevertheless, the positive findings of the CHAMPION trial4 contrasted
strikingly with those of other studies, such
as the negative study DOT-HF5 in which
an implantable intrathoracic impedance
detection device was used.
Another study that raised as many questions as answers was the STICH trial.6 In this
study, 1,212 patients with a left ventric­ular
ejection fraction <35%, coronary artery
disease amenable to surgery, and no indica­
tion for revascularization (that is, left main
disease or Canadian Cardiovascular Class III
angina or greater) were randomly assigned
to receive medical therapy or CABG
surgery. The primary end point was death
from any cause. Over a median follow-­up of
56 months, 218 of the 610 patients assigned
to surgery and 244 of the 602 patients
alloca­ted to medical therapy died (36% vs
41%; HR 0.86, 95% CI 0.72–1.04, P = 0.12).
However, occurrence of the key secondary end points (cardiovascular death and
death or cardiovascular hospitalization) was
signifi­cantly less common in the surgical
S6 | JANUARY 2012
group, as was all-cause mortality in an
adjusted intention-to-treat analysis and in
a per protocol analysis. As in prior trials of
CABG surgery, there was an initial mortality
excess in the surgery group and benefit from
surgery was not apparent until after approximately 2 years of follow-up.6 On balance, the
STICH trial probably was a positive study,
although the patients were highly selected
and atypically young (mean age 60 years)
for the majority with HF; two-thirds had
symptoms of angina. Unexpectedly, pre­
operative evaluation of myocardial viability
did not help to identify patients who would
benefit from revascularization. The takehome message seems to be that, in carefully selected patients with systolic HF and
coronary artery disease likely to survive for
more than 2 years, CABG surgery leads to
reduced long-term mortality and morbidity
compared with medical therapy alone.
The list of treatments that are ineffective
in acute HF grew longer in 2011 with the
publication of the results of ASCEND-HF,7
which showed that nesiritide (recombinant
human B‑type natriuretic peptide) had only
a small effect on dyspnea at 6 h and 24 h and
no effect on mortality or worsening HF at
30 days. This trial ended a long-running
controversy about the safety and efficacy of
nesiritide and highlighted the difficulty
of demonstrating a benefit for novel therapies over and above that of conventional
treatment with diuretics and nitrates, which
is rapid and substantial.8 Despite the lack
of success with treatments for acute HF
to date, interesting new drugs continue to
be developed. One of these, with intriguing proof-of-concept data in patients, is
a ‘first-in-class’ cardiac myosin activator,
omecamtiv mercarbil, which is thought to
improve cardiac performance by increasing the rate of transition from the weak
myosin-acting binding state to the strongly
bound (that is, force-generating) state.
This agent increases the duration of systole
without increasing heart rate or myocardial
oxygen consumption.9
Finally, I would like to mention an important study relevant to the holistic care of
patients with HF. Peterson and colleagues10
looked at health literacy in 1,494 patients
with an emergency room visit or hospitaliza­
tion with HF using three brief screening
questions—how often do you have someone
help you read hospital materials; how often
do you have problems learning about your
medical condition because of difficulty
reading hospital materials; and how confident are you filling out forms by yourself?
More than one in six patients showed a
low level of health literacy, which was an
indepen­dent predictor of all-cause mortality, after taking account of other prognostic
variables.10 These findings argue for testing
of the effect of interventions thought to
improve health literacy on outcomes in HF.
In summary, 2011 has been largely a year
of consolidation for HF therapeutics. We
have an array of effective drugs, devices, and
surgical procedures for patients with systolic
HF, but substantial progress has yet to be
made in the treatment of patients with preserved ejection fraction and those with
acute decompensation.
British Heart Foundation Cardiovascular
Research Centre, University of Glasgow,
126 University Place, Glasgow G12 8TA, UK.
[email protected]
Competing interests
The author declares associations with the following
companies: Amgen and Johnson & Johnson/Scios.
See the article online for full details of the
Goldenberg, I. et al. Reduction of the risk
of recurring heart failure events with cardiac
resynchronization therapy: MADIT-CRT
(Multicenter Automatic Defibrillator
Implantation Trial With Cardiac
Resynchronization Therapy). J. Am. Coll.
Cardiol. 58, 729–737 (2011).
2. Starling, R. C. et al. Results of the post‑US
Food and Drug Administration-approval study
with a continuous flow left ventricular assist
device as a bridge to heart transplantation:
a prospective study using the INTERMACS
(Interagency Registry for Mechanically Assisted
Circulatory Support). J. Am. Coll. Cardiol. 57,
1890–1898 (2011).
3. Slaughter, M. S. et al. Advanced heart failure
treated with continuous-flow left ventricular
assist device. N. Engl. J. Med. 361, 2241–2251
4. Abraham, W. T. et al. Wireless pulmonary artery
haemodynamic monitoring in chronic heart
failure: a randomised controlled trial. Lancet
377, 658–666 (2011).
5. van Veldhuisen, D. J. et al. Intrathoracic
impedance monitoring, audible patient alerts,
and outcome in patients with heart failure.
Circulation 124, 1719–1726 (2011).
6. Velazquez, E. J. et al. Coronary-artery bypass
surgery in patients with left ventricular
dysfunction. N. Engl. J. Med. 364, 1607–1616
7. O’Connor, C. M. et al. Effect of nesiritide in
patients with acute decompensated heart
failure. N. Engl. J. Med. 365, 32–43 (2011).
8. Felker, G. M. et al. Diuretic strategies in
patients with acute decompensated heart
failure. N. Engl. J. Med. 364, 797–805 (2011).
9. Cleland, J. G. et al. The effects of the cardiac
myosin activator, omecamtiv mecarbil,
on cardiac function in systolic heart failure:
a double-blind, placebo-controlled, crossover,
dose-ranging phase 2 trial. Lancet 378,
676–683 (2011).
10. Peterson, P. N. et al. Health literacy and
outcomes among patients with heart failure.
JAMA 305, 1695–1701 (2011).
New insights—from risk factors
to treatment implications
George L. Bakris
The results of several hypertension studies published in 2011 have
contributed to our knowledge on the risks of and treatment for this
condition, including the effects of slow-wave sleep, nocturnal dosing
of medication, variability in post-stroke blood-pressure reduction,
and the impacts of a low-sodium diet.
Bakris, G. L. Nat. Rev. Cardiol. 9, 75–77 (2012); published online 13 December 2011;
Four key studies, completed and published
in 2011, should affect the evaluation of
patients with hypertension and practice patterns in this population. In one such trial,
conducted in Spain, medications were dosed
at night in order to improve the dipping
status of blood pressure (BP). In this trial
661 patients with hypertension who were
at various stages of chronic kidney disease
(CKD) were randomly assigned to take all
prescribed hypertension medications upon
awakening or at least one drug before going
to bed.1 At the end of follow-up (median
5.4 years), patients who took the nocturnal
dose had an adjusted risk for total cardiovascular events that was approximately onethird that of patients who took medication
upon awakening. 1 Patients who received
nocturnal treatment also had a significantly
lower mean BP during sleep and a greater
proportion of these indivi­duals demonstrated controlled ambulatory BP compared
with those taking medication upon awakening.1 These data are consistent with those
from a shorter-term study of patients with
stage 2–3 CKD and no signifi­cant proteinuria.2 However, a pilot study in the African
American Study of Kidney Disease (AASK)
cohort of ~100 patients with stage 4 CKD
and proteinuria using similar methodo­logy
as the Spanish group failed to show this
benefit on BP dipping status or BP control
in the morning after 1 year of dosing antihypertensive medications at night.3 One of
the major differences between the studies
was that 100% of the patients in the AASK
cohort had proteinuria at various levels and
stage 4 nephropathy, whereas only about
10% of the patients in the Spanish study fit
this description. A nocturnal dosing strategy
was also tested in the large CONVINCE
trial, 4 but failed to show a reduction in
cardio­vascular risk. However, the CKD and
dipping status of the individuals enrolled in
the trial was unknown. Given the increased
cardiovascular risk among patients with
stage 3 CKD, ambulatory BP monitoring
(ABPM) is required to assess dipping status
and indivi­dualize dosing regimens in people
who meet the criteria as nondippers. As
yet no ABPM data are available from large
cohorts with stage 4 CKD and so the recom­
mendation cannot be extended to this group
of patients.
The second major advance in 2011 was
the finding of an association between poor
sleep quality and the development of hyper­
tension.5 This relationship is unrelated to
sleep apnea and has more to do with the
amount and type of sleep. Fung et al. evaluated whether incident hypertension is
associ­ated with polysomnography measures
of sleep-disordered breathing, sleep duration, and sleep architecture in older men
(mean age 75 years).5 Participants included
784 community-dwelling, ambulatory men
from the Outcomes of Sleep Disorders
in Older Men Study who did not have
Key advances
■■ Among patients with moderate-stage
nephropathy, dosing medications at night
has the advantage of improving blood
pressure (BP) goal attainment in the
early morning when cardiovascular risk
is highest1
■■ Failure to recognize insufficient quality of
sleep secondary to frequent disturbances
of awakening is a major contributing
risk for hypertension development and
poor BP control5
■■ Lowering BP by more than 15–20% in
patients with hypertension early in the
post-stroke period can result in long-term
cognitive consequences6
■■ Low-sodium diets clearly reduce BP, but
tend only to reduce cardiovascular events
in trials; in populations, low-sodium diets
reduce mortality8
hypertension at the time of their in-home
sleep studies and who returned for followup after 2–4 years. By the end of follow-up
(mean 3.4 years), 243 men met the criteria for incident hypertension. After adjustment for age, nonwhite race, study site, and
BMI, the only sleep index to remain signifi­
cantly associated with incident hyper­tension
was decreased stage N3 (or slow-wave)
sleep. No attenuation of this association was
seen after accounting for sleep duration,
sleep fragmentation, and indices of sleepdisordered breathing. Thus, failure to get
an appropriate amount of slow-wave sleep
each night is a risk factor for development
of hyper­tension. Reduced amounts of slowwave sleep are seen commonly in elderly
patients and could be related to increased
frequency of waking to urinate. Physicians
should be aware of decreased deep sleep as a
risk factor for hypertension and ensure that
patients get an extended rest period at night
by decreasing stimuli, such as caffeine and
liquid intake. This strategy is used in my
practice with a fair amount of success.
A third important trial published in 2011
provides new information on the consequences of aggressive BP management
in the setting of acute stroke. In SCAST,6
2,029 patients from northern Europe with
acute ischemic or hemorrhagic stroke and
systolic BP >140 mmHg were randomly
assigned within 30 h of symptom onset to
cande­sartan or placebo for 7 days. The dose
was increased from 4 mg on day 1 to 16 mg
on days 3–7. During the 7‑day treatment
period, BP was 5/2 mmHg lower in the candesartan group than in the placebo group
(mean BP 147/82 mmHg vs 152/84 mmHg).
At the end of the 6 month follow-up, the risk
of the primary composite vascular end point
did not differ between treatment groups.
The co-primary end point of functional
outcome (modified Rankin scale), however,
was associ­ated with a higher risk of poor
outcome in the candesartan group. The
results were interpreted by the investigators
to suggest some harm and no benefit for BP
lowering after acute stroke.6
To understand this finding, clinicians
should be aware that current BP guidelines
for the post-stroke period differ according to whether the stroke is hemorrhagic
or ischemic. 7 In addition, the approach
to BP management for the prevention of
stroke is totally different from its management post stroke. In hemorrhagic stroke,
current guidelines suggest cautious lowering of BP by no more that 20% in the
first 24 h.7 Agents with rapid, short-term
JANUARY 2012 | S7
© Elena Babushkina |
action such as labetalol, nicardipine, and
fenoldapam are preferred as nitroglycerin
and nitroprusside can increase intra­cranial
pressure. For patients with ischemic stroke
who are not receiving thrombolytic therapy,
BP should be lowered if markedly elevated
(>220/120 mmHg). 7 In SCAST, baseline
BP was 171/90 mmHg,6 so a reduction to
~140 mmHg is excessive given the recommendation and, therefore, the lack of benefit
is not surprising.
The fourth notable study published in
2011 is a meta-analysis in which the investigators aimed to assess the long-term effects
of dietary salt reduction on mortality and
cardiovascular morbidity and whether BP
reduction is an explanatory factor in any
effect of such dietary interventions.8 A total
of 7 studies were included, 3 in normo­
tensive individuals, two in patients with
hyper­tension, one in a mixed population,
and one in patients with heart failure with
end of trial follow-up of 7–36 months and
longest observational follow up to 12.7 years.
The data failed to show strong evidence
for a reduced risk of all-cause mortality or
cardio­vascular morbidity in normo­tensive
indivi­duals and patients with hyper­tension
on low-salt diets. The investigators suggest
their review had insufficient power to
exclude clinically important effects, since
only 665 deaths occurred in 6,250 participants. They further state that their estimates
of benefits from dietary salt restriction are
consistent with the predicted small effects
on clinical events attributable to the small
BP reduction achieved.
This information puts into perspective the
limitations of clinical trials. The evidence
that salt reduction lowers BP is clear and
consistent. Since BP reduction is critical for
reducing the risk of stroke and progression of
CKD, low-salt diets should also be of benefit.
S8 | JANUARY 2012
However, most people do not consistently
maintain a low sodium intake leading to
fluctu­ation in BP, which is a major risk factor
for stroke. The real impact of low sodium
intake on mortality has been demonstrated in
Japan and Finland where sodium reduction
policies have been instituted for all foods sold
in stores. In the USA, reducing dietary salt by
3 g per day is projected to reduce the annual
number of all-cause deaths from 44,000 to
92,000.9 Hence, large numbers of indivi­
duals and years of follow-up are required to
see the effect of consistent reduced sodium
intake. Not surprisingly, therefore, trial data
and even meta-­analyses cannot adequately
demon­strate these effects. Clinicians should
focus on the established fact that BP reduction is consistently associ­ated with reductions
in cardio­vascular event rates and continue to
stress the importance of a low-sodium diet
to patients with hypertension.
Finally, the multicenter, prospective, randomized Symplicity HTN‑2 trial, which
was published in late 2010, 10 provided
ground-breaking results for a new concept
of treatment for refractory hypertension.
Participants with systolic BP ≥160 mmHg (or
≥150 mmHg for those with type 2 diabetes
mellitus) were randomly allocated to undergo
renal denervation together with continuation
of previous treatment or maintain previous
treatment alone. Office-based BP measurements were reduced by 32/12 mmHg (baseline 178/96 mmHg) in the renal denervation
group, control group BP values did not differ
from the baseline of 178/97 mmHg, and
between-group differences in BP at 6 months
were 33/11 mmHg. No serious procedure or
device-related complications occurred and
adverse events did not differ between groups.
The investi­gators concluded that catheterbased renal denervation is a safe and viable
treatment for patients who cannot achieve
BP goals with pharmaco­therapy.10 In 2011,
recruitment for the Symplicity HTN‑3 trial
started in the USA; with a protocol that is
more stringent than Symplicity HTN‑2 and
involves more frequent ABPM. This new
trial will include between-group differences
in ABPM as a prespecified secondary end
point. Hypertension management prior to
study enrollment will be more intense and
will include spironolactone. If the results of
the Symplicity HTN‑3 trial are as gratifying
as that of Symplicity HTN‑2, a true alternative to medications will exist for refractory
hypertension. Many other ongoing clinical
trials have been designed to address and,
hopefully, answer important questions
about specific approaches to reducing cardio­
vascular risk and the progression of CKD in
patients with hypertension.
Hypertensive Diseases Unit, Section of
Endocrinology, Diabetes and Metabolism,
Department of Medicine, University of Chicago
Pritzker School of Medicine, 5841 South
Maryland Avenue, MC 1027, Chicago,
IL 60637, USA.
[email protected]
Competing interests
The author declares associations with the following
companies: Abbott, CVRx, Eli Lilly, FDA, Forest
Laboratories, Johnson & Johnson, Medtronic,
Novartis, Relypsa, Servier, and Takeda. See the
article online for full details of the relationships.
Hermida, R. C., Ayala, D. E., Mojon, A.
& Fernandez, J. R. Bedtime dosing of
antihypertensive medications reduces
cardiovascular risk in CKD. J. Am. Soc. Nephrol.
Minutolo, R. et al. Changing the timing of
antihypertensive therapy to reduce nocturnal
blood pressure in CKD: an 8‑week uncontrolled
trial. Am. J. Kidney Dis. 50, 908–917 (2007).
Rahman, M. & Appel, L. J. Should reducing
nocturnal blood pressure be a therapeutic
target in CKD? The time is ripe for a clinical
outcomes trial. Am. J. Kidney Dis. 50, 901–903
Black, H. R. et al. Principal results of the
Controlled Onset Verapamil Investigation
of Cardiovascular End Points (CONVINCE) trial.
JAMA 289, 2073–2082 (2003).
Fung, M. M. et al. Decreased slow wave sleep
increases risk of developing hypertension in
elderly men. Hypertension 58, 596–603 (2011).
Sandset, E. C. et al. The angiotensin-receptor
blocker candesartan for treatment of acute
stroke (SCAST): a randomised, placebocontrolled, double-blind trial. Lancet 377,
741–750 (2011).
Aiyagari, V. & Gorelick, P. B. Management of
blood pressure for acute and recurrent stroke.
Stroke 40, 2251–2256 (2009).
Taylor, R. S., Ashton, K. E., Moxham, T.,
Hooper, L. & Ebrahim, S. Reduced dietary salt
for the prevention of cardiovascular disease:
a meta-analysis of randomized controlled trials
(Cochrane review). Am. J. Hypertens. 24,
843–853 (2011).
Bibbins-Domingo, K. et al. Projected effect of
dietary salt reductions on future cardiovascular
disease. N. Engl. J. Med. 362, 590–599
10. Esler, M. D. et al. Renal sympathetic denervation
in patients with treatment-resistant hypertension
(the Symplicity HTN‑2 trial): a randomised
controlled trial. Lancet 376, 1903–1909 (2010).
Breakthrough for intervention?
Volkmar Falk
In 2011, both the PARTNER‑A trial, in high-risk patients with severe aortic
stenosis, and EVEREST II, in patients with mitral insufficiency, showed
noninferiority of transcatheter interventions compared with surgery for
the chosen primary end points. However, both of the trials, and important
registry data, identified limitations of transcatheter valve interventions.
Falk, V. Nat. Rev. Cardiol. 9, 77–78 (2012); published online 20 December 2011;
Data from two major randomized, controlled trials in which interventional
therapy was compared with conventional
surgery for the treatment of aortic stenosis in high-risk patients1 and for selected
patients with mitral insufficiency 2,3 were
published in 2011. Both trials showed noninferiority for their respective primary end
points, but (together with data from national
registries4,5) also highlighted limitations of
interventional therapy.
The most-important trial in the field of
valvular heart disease that was published in
2011 was the PARTNER‑A trial,1 in which
transcatheter aortic-valve implantation
(TAVI) was compared with surgical aorticvalve replacement (SAVR) in a high-risk
population of patients with severe aortic
stenosis. From a total of 3,105 screened
patients, 699 individuals (23%) were randomly allocated to either SAVR (n = 351) or
TAVI (n = 348, transfemoral or trans­apical)
using the Edwards SAPIEN ® (Edwards
Lifesciences Corporation, Irvine, CA, USA)
balloon-expandable valve. Patients were
considered high risk on the basis of coexisting conditions associated with a risk of death
≥15% at 30 days. In the intention-to-treat
analysis, death from any cause (the primary
end point) was not different between the
TAVI and SAVR groups at 30 days (3.4% vs
6.5%, respectively) or 1 year (24.2% vs 26.8%,
respectively), which led to the conclusion that
TAVI was non­inferior to SAVR. Patients in
the TAVI group had a shorter length of stay
in the intensive-care unit and a shorter index
hospitalization compared with patients
undergoing SAVR.
Bleeding and new onset of atrial fibrillation were more frequent in the surgical group, whereas rates of vascular and
neuro­logical complications were higher in
the TAVI group. The rates of major stroke
were 3.8% and 2.1% at 30 days, and 5.1%
and 2.4% at 1 year, in the TAVI and surgical groups, respectively. At 1 year, TAVI
was associated with a lower mean aorticvalve gradient compared with the surgical
group, but with a higher rate of moderate or
severe paravalvular regurgitation at 30 days
(12.2% vs 0.9%) and 1 year (6.8% vs 1.9%).
Functional improvement was significant
for both groups after 1 year and not different between the groups. The results
of the PARTNER‑A trial 1 demonstrate
that TAVI is not only superior to medical
therapy in non­surgical candidates, but also a
true alternative to surgical replacement for
a selected, high-risk subgroup of patients
with aortic stenosis.
Data from a number of national TAVI registries were published in 2011. The FRANCE
registry 6 included data on 244 patients who
received an Edwards SAPIEN® (68%) or
CoreValve® (Medtronic CV Luxembourg
S.a.r.l., Luxembourg; 32%) in 2009. Device
success rate was 98.3% and 30‑day mortal­ity
Key advances
■■ In high-risk patients with severe aortic
stenosis, transcatheter aortic-valve
implantation (TAVI) can be performed
with similar 1‑year survival and functional
outcomes as conventional surgery1
■■ Moderate and severe paravalvular
leakage after TAVI is an independent
predictor of late mortality4,5
■■ Percutaneous edge-to-edge mitral-valve
repair can be safely performed with low
periprocedural complications2
■■ Residual moderate mitral regurgitation
is more frequent after percutaneous
edge-to-edge repair than surgery and
might impair long-term results3
was 12.7%. Stroke and vascular complications occurred in 3.6% and 7.3% of patients,
respectively. Data from the FRANCE 2 registry,7 which included 2,419 patients treated
between January 2010 and July 2011, were
presented at the ESC Congress in 2011.
Device success rate was 97.1%, and 30‑day
and 6‑month mortality were 9.9% and
17.2%, respectively. The stroke rate of 4.0%
was similar to the rates reported in the
PARTNER trials.1 The high incidence of
stroke after TAVI is concerning, and might
result from a substantial embolic load during
the TAVI procedure, as has been demonstrated using transcranial Doppler imaging
during implantation and cerebral MRI after
the procedure.
The UK TAVI registry 8 included 890
patients. Procedural success (97.2%) and
stroke rate (4.1%) were similar to the French
experience. Mortality at 30 days was 7.1%, and
1‑year and 2‑year survival were 78.6% and
73.7%, respectively. A degree of paravalvular
aortic regurgitation (AR) regarded as suboptimal or unacceptable occurred in 61% of
patients. The presence of moderate or severe
AR (13% of patients) was an independent
predictor of mortality at 1 year.
In the German TAVI registry, 4 which
comprised 690 patients (84% Medtronic
CoreValve ® system), the rate of AR ≥2
was 17.2% and its presence was a strong,
in­dependent predictor of in-hospital death
(adjusted OR 2.43). In the Italian, multicenter
registry 5 of 663 patients who underwent
transfemoral TAVI with the third-generation
18 Fr CoreValve® device, the rate of paravalvular leak ≥2 after implantation was 21%,
and independently associated with mortality between 30 days and 1 year (HR 3.79).
Therefore, data from three national registries
have shown that paravalvular leakage is an
independent risk factor for mortality after
TAVI.4,5,8 The development of improved prosthetic implants and implantation techniques
is required.
In the absence of long-term follow-up data
after TAVI, the results of the PARTNER‑A
trial1 and the data from the national registries4–8 must be interpreted with caution.
Recommendations to individual patients
must balance the obvious advantages of
a less-invasive transcatheter approach
compared with SAVR, with the increased
risk of stroke and unfavorable longterm consequences of paravalvular leakage
after TAVI.
The landmark mitral-valve trial published in 2011 was EVEREST II.2 In total,
279 patients with severe mitral regurgitation
JANUARY 2012 | S9
Figure 1 | Images from a patient in NYHA class IV because of ischemic cardiomyopathy with
poor LV ejection fraction (<20%) and mitral-valve insufficiency III–IV, who underwent a MitraClip®
(Abbott Vascular, Abbott Park, IL, USA) procedure. a | The preprocedural LV end-diastolic volume
was assessed as 550 ml. b | At 1 year follow-up after the procedure, LV end-diastolic volume was
reduced to 480 ml, the patient had improved to NYHA class II, and residual mitral-valve
regurgitation was <1+. Abbreviation: LV, left ventricular.
(MR; grade 3+ or 4+) were randomly allocated to either surgery or percutaneous edgeto-edge clipping using the MitraClip® device
(Abbott Vascular, Abbott Park, IL, USA).
Acute procedural success was not achieved
with the MitraClip® in 23% of patients, the
majority of whom underwent subsequent
mitral-valve surgery. The transvenous–­
transseptal approach was associated with
fewer adverse events at 30 days compared
with surgery. The composite end point of
major adverse events was reached in 9.6%
and 57% of patients in the MitraClip® and
surgical groups, respectively. This striking
difference was almost exclusively driven by
the need for blood transfusion (8% vs 53%,
respectively). Noninferiority for the clinicaleffectiveness 1‑year end point (freedom from
death, mitral-valve surgery, mitral-valve dysfunction >90 days after the index procedure,
and MR >2+ at 12 months) was met, despite
the fact that 18.5% and only 3% of patients
had residual mitral insufficiency >2+ at
1 year in the MitraClip® and surgical groups,
respectively. Another 33% of patients in the
MitraClip® group had residual mitral insufficiency = 2+, and more than 15% of patients
in the device arm subsequently underwent
mitral-valve surgery because of persistent,
severe MR.
The arbitrary end point of freedom from
mitral insufficiency >2+ has triggered lively
discussion. Residual mitral insufficiency
>1+ after mitral repair is clearly unacceptable, and is an independent predictor of
late mortality. If the EVEREST II investi­
gators had chosen presence of mitral insufficiency <2+ instead of absence of mitral
S10 | JANUARY 2012
insufficiency >2+ as part of the effectiveness end point, fewer than half of the
patients (47.9%) treated with the MitraClip®
would have fulfilled the combined criteria.
Statistical noninferiority for effectiveness
at 1 year was reached by definition, but the
debate whether this result can be regarded
as noninferior from a clinical perspective is
likely to continue—particularly because the
intention-to-treat analysis revealed that
the primary composite end points at 2 years
were still significantly better with surgery
than percutaneous intervention.3 Longterm data are needed before the indication
for this therapy can be extended to patients
with degenerative mitral-valve disease, in
whom the success rate of endoscopic surgical mitral-valve repair is >95%, with excellent
short-term and long-term results.9 Mitralvalve repair after previous clip placement
is often not possible because the device is
encapsulated over time by fibrous tissue;
the use of the MitraClip® should, therefore,
be restricted to patients who are at high
surgical risk.
The unconvincing long-term results of
downsizing annuloplasty in patients with
severe leaflet tethering make the MitraClip®
an attractive option in patients with functional MR. In the PERMIT trial,10 51 severely
symptomatic nonresponders to cardiac
resynchronization therapy with severe left
ventricular dysfunction and functional MR
underwent treatment with the MitraClip®.
Residual MR ≥2+ was present in <20% of
patients at discharge, and in only about 10%
at 1 year follow-up. Periprocedural mortality was 5.8%, but was not less than surgery
in similar groups of patients. The NYHA
functional class had already improved at discharge in 73% of patients. Reverse remodel­
ing was observed to continue at 6 months
and 1 year (Figure 1). A randomized trial
with larger patient numbers than in the
PERMIT trial10 will be needed to confirm
these findings and determine the role of the
MitraClip® in patients with functional MR.
The treatment of valvular heart disease
continues to progress with the evolution
of percutaneous approaches as alternatives
to surgical therapy in selected patients. To
provide patients with the best-possible
treatment, short-term effectiveness must be
balanced with long-term outcomes.
Division of Cardiovascular Surgery,
University of Zurich, Rämistrasse 100,
Zurich CH‑8091, Switzerland.
[email protected]
Competing interests
The author declares associations with following
companies: Edwards Lifesciences, Medtronic,
Philips, St Jude Medical, and Valtech. See the article
online for full details of the relationships.
Smith, C. R. et al. Transcatheter versus surgical
aortic-valve replacement in high-risk patients.
N. Engl. J. Med. 364, 2187–2198 (2011).
2. Feldman, T. et al. Percutaneous repair or
surgery for mitral regurgitation. N. Engl. J. Med.
364, 1395–1406 (2011).
3. George, J. C., Varghese, V., Dangas, G. &
Feldman, T. E. Percutaneous mitral valve repair:
lessons from the EVEREST II (Endovascular
Valve Edge-to-Edge REpair Study) and beyond.
JACC Cardiovasc. Interv. 4, 825–827 (2011).
4. Abdel‑Wahab, M. et al. Aortic regurgitation
after transcatheter aortic valve implantation:
incidence and early outcome. Results from the
German transcatheter aortic valve interventions
registry. Heart 97, 899–906 (2011).
5. Tamburino, C. et al. Incidence and predictors
of early and late mortality after transcatheter
aortic valve implantation in 663 patients with
severe aortic stenosis. Circulation 123,
299–308 (2011).
6. Eltchaninoff, H. et al. Transcatheter aortic
valve implantation: early results of the FRANCE
(FRench Aortic National CoreValve and
Edwards) registry. Eur. Heart J. 32, 191–197
7. Gilard, M. et al. FRANCE 2: FRench Aortic
National CoreValve and Edwards Registry
view.aspx?eevtid=48&fp=3126 (2011).
8. Moat, N. E. et al. Long-term outcomes after
transcatheter aortic valve implantation in
high-risk patients with severe aortic stenosis:
The U.K. TAVI (United Kingdom Transcatheter
Aortic Valve Implantation) Registry. J. Am. Coll.
Cardiol. 58, 2130–2138 (2011).
9. Mihaljevic, T. et al. Robotic repair of posterior
mitral valve prolapse versus conventional
approaches: potential realized. J. Thorac.
Cardiovasc. Surg. 141, 72–80 (2011).
10. Auricchio, A. et al. Correction of mitral
regurgitation in nonresponders to cardiac
resynchronization therapy by MitraClip improves
symptoms and promotes reverse remodeling.
J. Am. Coll. Cardiol. 58, 2183–2189 (2011).
Mutations and non-inferiority analyses show
a way forward
Maurie Markman
Highly clinically relevant ovarian cancer clinical research in 2011 focused on an increased understanding of the
biology of the malignancy, limitations of strategies for early detection and screening, and the provocative reports of
alternative primary and second-line management strategies.
Markman, M. Nat. Rev. Clin. Oncol. 9, 69–70 (2012); published online 20 December 2011; doi:10.1038/nrclinonc.2011.200
Although there were a number of very interesting preliminary reports of thera­peutic
advances in ovarian cancer in 2011 (for
example, bevacizumab in the first-line and
second-line management of the malignancy,
and olaparib as maintenance therapy for
high-grade serous cancers), as of the writing
of this commentary these studies have not
appeared in the peer-reviewed oncology literature. Despite the absence of major advances
in the realm of treatment, several papers
published in 2011 provide highly clinically
relevant insight into the management and
unique biology of ovarian cancer.
Perhaps the most important paper in
2011 was the long-awaited final report from
the Prostate, Lung, Colorectal and Ovarian
(PLCO) cancer screening randomized controlled trial dealing specifically with ovarian
cancer.1 The study, involving 78,000 women
with ages ranging from 55 to 74 years, randomly assigned participants to what was
classified as ‘usual care’ or a rather-intensive
screening strategy that included annual
serum CA125 determinations and trans­
vaginal ultrasounds. It should be noted that
this study did not specifically target women
identified as being at ‘high risk’ for the
develop­ment of ovarian cancer (for example,
those with a family history of ovarian cancer).
The screening protocol, undertaken from
November 1993 to July 2001, was performed
at one of 10 centers in the USA and the
median follow-up period for the population
was 12.4 years, with patients being followed
until death or for a maximum of 13 years.
The primary study end point was the rate
of mortality from ovarian cancer (including
primary peritoneal or fallopian tube cancer),
with secondary end points of disease incidence and complications associ­ated with
Table 1 | Influence of BRCA1 or BRCA2 on outcome in high-grade serous ovarian cancer6
Mutational status
PFS (%)
PFS (%)
HR compared
with WT patients
OS (%)
OS (%)
HR compared
with WT patients
BRCA1 mutation
0.81 (P = 0.44)
0.76 (P = 0.35)
BRCA2 mutation
0.40 (P = 0.004)
0.33 (P = 0.003)
Abbreviations: HR, hazard ratio; OS, overall survival; PFS, progression-free survival; WT, wild type.
the screening strategy. The study revealed
no difference in deaths from ovarian cancer
or other causes between the screening and
usual-care groups. Specifically, there were
212 cases of ovarian cancer and 118 deaths
from the malignancy in the screening group
versus 176 cases and 100 deaths in the routine-care population. Furthermore, and of
considerable relevance, 1,080 surgeries were
performed among the 3,285 women with
false-positive screening tests, and 163 of
these individuals developed at least a single
‘serious complication’.1
The important data from this screening
trial provide no support for the routine use
of annual CA125 determinations or vaginal
ultrasounds for completely asymptomatic
women as a screening strategy to detect
ovarian cancer. However, it is again rele­vant
to acknowledge that this trial did not specifically address the issue of screening a more
high-risk population, nor did it attempt to
define the utility of these tests in the detection of ovarian cancer in indivi­duals presenting with symptoms (for example, several
weeks of persistent mild abdominal pain).
A study that is somewhat related to the
PLCO trial attempted to address the important question of whether an earlier diagnosis of ovarian cancer in a sympto­matic
individual might be associated with an
improved outcome.2 Australian investigators
retrospectively examined the survival of 1,300
patients with ovarian cancer seen by a physician within varying time intervals (55%, 70%
and 92% of the women presented within 1,
2 and 6 months, respectively) from the onset
of their initial symptoms. The investigators
were unable to find any difference in survival
based on the duration of symptoms before
the time of diagnosis. Importantly, these data
do not support the hypothesis that a somewhat earlier diagnosis (timeline measured
in ‘months’) will be associ­ated with superior
ovarian cancer-specific survival,3 although
the more-timely recognition of the correct
diagnosis will likely result in more-rapid initiation of a management plan that will hopefully favorably impact serious symptoms and
the individual’s overall quality of life.
It is well established that mutations in
BRCA1 and BRCA2 are associated with an
increased lifetime risk for the development
of ovarian cancer. 4 Furthermore, highly
provocative data from a number of ovarian
cancer investigators have suggested that
within the population of women with docu­
mented advanced-stage ovarian cancer,
patients with BRCA mutations experience
an overall superior survival (compared with
patients with wild-type BRCA), a difference
that is possibly related to increased sensiti­
vity to platinum-based chemotherapy.5 In
a most-provocative report, investigators
JANUARY 2012 | S11
Key advances
■■ There are currently no evidence-based
data supporting the clinical utility of any
ovarian cancer screening strategy in
non‑high-risk populations1
■■ Provocative data suggest there may be a
clinically meaningful difference between
the presence of a BRCA1 or a BRCA2
mutation in influencing outcome in ovarian
■■ Under specific circumstances (for example,
neuropathy) it might be reasonable to
substitute pegylated liposomal doxorubicin
for paclitaxel in the front-line chemotherapy
management of ovarian cancer7
conducted an observational study involving
316 women with high-grade serous ovarian
cancer whose BRCA mutational status was
established, to evaluate the impact of such
mutations on outcome.6 Although the total
sample size was limited (37 and 29 patients
with BRCA1 and BRCA2 mutations, respectively), there was a rather striking difference
in both the inherent chemosensitivity and
survival between the two mutation groups,
and compared with women without mutations (Table 1). Patients with BRCA2 mutations experienced a statistically significant
improvement in 5‑year survival compared
with BRCA1 deficient and BRCA ‘wild-type’
cases. Although these results will need to be
confirmed by other investigative groups with
larger sample sizes, they suggest the potential
that these two genetic abnormalities might
exert quite different influences on outcome
in high-grade serous ovarian cancer. At the
present time, there are no known thera­peutic
implications associated with these findings,
but it is possible that future research may
result in recommendations for different
management strategies for patients with
BRCA1 or BRCA2 mutations.
Finally, it is relevant to note an interesting
(although unfortunately flawed) phase III
randomized trial that directly compared
a regimen of carboplatin plus paclitaxel to
carboplatin plus pegylated liposomal doxorubicin as primary treatment for epithelial
ovarian cancer.7 Although this potentially
paradigm changing study included a total
of 820 patients, it was specifically designed
to demonstrate the therapeutic superiority
(rather than non-inferiority with possibility
an improved toxicity profile) of the investi­
gative pegylated liposomal doxorubicincontaining regimen. In fact, although the
study demonstrated similar median progression-free survival times (19.0 months
and 16.8 months) and median overall survival times (61.6 months and 53.2 months)
S12 | JANUARY 2012
for the carboplatin–pegylated liposomal
doxorubicin and carboplatin–paclitaxel
study arms, respectively, the trial failed to
achieve its primary end point (the documentation of ‘superiority’ for the experimental
arm over the established therapy arm). As a
result of this failure, the investigators appropriately concluded that the investigative
regimen “was not superior to carboplatin–
paclitaxel which remains the standard firstline chemo­t herapy for advanced ovarian
cancer.”7 However, it is reasonable to suggest
that for patients unable to tolerate the taxane
(for example, patients who develop neuro­
pathy early in the treatment course or severe
paclitaxel-associated hypersensitivity reaction) the substitution of pegylated liposomal
doxorubicin is not an unreasonable option.
Thus, it is reasonable to characterize the
ovarian cancer peer-reviewed literature in
2011 as providing modestly useful information regarding disease management and
including important discussions of the limitations in our ability to modify the relatively
early natural history of the malignancy. It
is rather striking that in a clinical research
world where unique targets have been discovered in multiple cancers (for example,
HER2 overexpression in breast cancer, EGFR
mutations in lung cancer, and BRAF mutations in melanoma) leading to exciting new
treatment strategies, there remain no such
molecular targets in ovarian cancer, the
presence of which would lead to specific
management paradigms to favorably impact
outcome. It is clear that much work needs
to be done to improve our understanding of
the fundamental biology of ovarian cancer
to change this current state-of-affairs.
Cancer Treatment Centers of America, Eastern
Regional Medical Center, 1331 East Wyoming
Avenue, Philadelphia, PA 19124, USA.
[email protected]
Competing interests
The author declares an association with the
following company: Genentech. See the article
online for full details of the relationship.
Buys, S. S. et al. Effect of screening on ovarian
cancer mortality: the Prostate, Lung, Colorectal
and Ovarian (PLCO) cancer screening
randomized controlled trial. JAMA 305,
2295–2303 (2011).
Nagle, C. M. et al. Reducing time to diagnosis
does not improve outcomes for women with
symptomatic ovarian cancer: a report from the
Australian Ovarian Cancer Study Group. J. Clin.
Oncol. 29, 2253–2258 (2011).
Anderson, M. R. et al. Combining a symptoms
index with CA125 to improve detection of
ovarian cancer. Cancer 113, 484–489 (2008).
King, M. C. et al. Breast and ovarian cancer
risks due to inherited mutations in BRCA1 and
BRCA2. Science 302, 643–646 (2003).
Tan, D. S. et al. “BRCA-ness” syndrome in
ovarian cancer: a case-control study describing
the clinical features and outcome of patients
with epithelial ovarian cancer associated with
BRCA1 and BRCA2 mutations. J. Clin. Oncol.
26, 5530–5536 (2008).
Yang, D. et al. Associations of BRCA1 and
BRCA2 mutations with survival, chemotherapy
sensitivity, and gene mutator phenotype in
patients with ovarian cancer. JAMA 306,
1557–1565 (2011).
Pignata, S. et al. Carboplatin plus paclitaxel
versus carboplatin plus pegylated liposomal
doxorubicin as first-line treatment for patients
with ovarian cancer: The MITO‑2 randomized
phase III trial. J. Clin. Oncol. 29, 3628–3635
Hitting old targets better
and identifying new targets
Yu Chen and Howard I. Scher
Options to treat late-stage castration-resistant prostate cancer continued
to increase in 2011, as three agents with different mechanisms of
action prolonged life and a fourth reduced the morbidity of skeletal
metastases. These outcomes contrasted with the heightened controversy
generated by the recommendation against PSA screening and other early
detection strategies.
Chen, Y. & Scher, H. I. Nat. Rev. Clin. Oncol. 9, 70–72 (2012); published online 10 January 2012;
In the Western world, one in six men will
be diagnosed with prostate cancer and,
of these, one in six will die of metastatic
disease. Improving outcomes for men with
prostate cancer depends on the one hand
on developing more-effective systemic
therapies and, on the other hand, on early
diagnosis and treatment, before the disease
has metastasized.
2011 is the 70 th anniversary of when
Charles Huggins established that prostate cancer is an androgen-dependent
malignancy and, until recently, docetaxel was
the only non-hormonal therapy to prolong
life. In 2010, sipuleucel‑T and cabazitaxel
were shown to confer a survival benefit and
were subsequently approved by the FDA.
This trend continued in 2011, as three therapies, CYP17 inhibitor abiraterone acetate,1
bone-targeting agent radium‑223, 2 and
androgen-signaling inhibitor MDV3100,3
were shown to prolong life in definitive
phase III clinical trials (Figure 1), and the
RANKL inhibitor, denosumab, was shown
to be superior to zoledronic acid in reducing the morbidity associated with bone
metastases.4 The demonstration that agents
targeting unique aspects of the malignant
process associated with tumor cell growth
and survival could provide meaningful clinical benefits has highlighted the importance
of understanding the biology of castrationresistant prostate cancer (CRPC). The trials
established important principles, which will
be discussed.
The first principle is that CRPCs are not
hormone refractory. CRPCs acquire diverse
mechanisms to reactivate the androgen
receptor signaling pathway in the environment of castrate levels of androgens, including an increase in the androgen biosynthetic
machinery and overexpression of androgen
receptor. Thus, further decreasing androgen levels by inhibiting steroid synthesis
enzymes in the adrenal glands and tumor
may be of benefit. CYP17 is a cytochrome
P450 enzyme that catalyzes the rate-limiting
step of androgen synthesis; abiraterone
acetate is a structural analog of the CYP17
substrate pregnenolone that is rationally
designed to be a specific and irreversible
inhibitor of CYP17. A large international
phase III trial (Cougar AA‑301) that compared abiraterone acetate plus prednisone
(to prevent mineralocorticoid excess) and
placebo plus prednisone in patients with
CRPC who had received docetaxel showed
an increase in median overall survival from
10.9 to 14.8 months (hazard ratio = 0.65;
P <0.001).1 All major subgroups of patients
benefitted and treatment was well tolerated. A separate trial assessing abiraterone
acetate in chemotherapy-naive patients
has completed accrual and the results are
awaited (NCT00887198).
Another drug supporting the principle
that CRPCs are not hormone refractory is
MDV3100, a next-generation anti-androgen
that was rationally designed to overcome
several deficiencies of available agents
including modest receptor binding and
agonist properties that can promote tumor
Rising PSA
Rising PSA
Non-castrate prostate cancer
Castrate prostate cancer
Trial drug shown to improve
survival in phase III trials
Figure 1 | Clinical states model of prostate cancer progression and systemic therapies shown to
improve survival for castrate disease. Blue text represents FDA approved therapies.
growth. The drug displays no androgen
receptor activation, blocks nuclear trans­
location of the receptor and completely
inhibits the ability of androgen receptor
to bind to DNA. 3 Promising activity of
MDV3100 was demonstrated in a phase I–
II trial in 140 patients, which led to the
phase III AFFIRM trial in patients with
chemotherapy-treated CRPC that compared
MDV3100 with placebo. 3 This trial was
stopped by the Data Safety and Monitoring
Board in November 2011 when the first
planned interim analysis showed a 37%
reduction in mortality, and superior overall
survival (median 18.4 versus 13.6 months)
in favor of MDV3100.3
The second principle was that targeting
the bone environment can confer benefits,
independent of an effect on prostate-specific
antigen (PSA). Bone is the most common
site of prostate cancer spread and is responsible for the highest morbidity burden from
the disease. Therapies directed at the bone
microenvironment can be divided into three
classes: osteoclast inhibitors of bone resorption, radiopharmaceuticals that target bone,
and kinase inhibitors. Despite the radiographic appearance of osteoblastic lesions,
there is heightened osteoclastic activity and
bone turnover in metastatic prostate cancer
lesions. Osteoclast inhibitors, including zoledronic acid, are used as an adjuvant therapy
to maintain bone density and to decrease
skeletal-related events (SRE) in patients with
CRPC. Denosumab is a humanized monoclonal antibody that blocks RANKL, which
is required for osteoclast activation and
survival. In 1,904 patients with CRPC that
had meta­stasized to the bone, denosumab
delayed the median time to first SRE from
17.1 to 20.7 months (P = 0.008) compared to
zoledronic acid, leading to FDA approval.4
The beta-emitting bone-seeking radioisotopes strontium‑89 and samarium‑153
lexidronam are approved for the palliation of
pain, but neither has been shown to prolong
life. Radium‑223 is a unique bone-directed
radioisotope that emits high-energy alpha
particles that penetrate only several cell
layers in tissue, significantly improving
the radiation delivery to areas of sclero­
tic bone where the tumor resides while
minimizing radiation to hematopoietic
tissue. In a phase III trial in 922 patients
with CRPC and bone pain, 6-monthly
doses of radium‑223 conferred a significant overall survival benefit over placebo
(14.0 versus 11.2 months; P = 0.00185);2 it
is the first bone-directed agent to prolong
overall survival.
Multiple signaling cascades are critical
for bone homeostasis and turnover. Two
promising investigational agents targeting this axis deserve mention. Dasatinib
inhibits kinases including Src family tyrosine kinases that are important for bone
turnover; a single-agent phase II trial in
patients with CRPC showed a marked
decrease in bone turnover, disease stabilization5 and additive effects with docetaxel.
A phase III trial comparing dasatinib with
placebo in combination with docetaxel has
completed accrual. Cabozantinib inhibits
kinases including Ret, Met, and VEGFR2;
in a randomized phase III discontinuation
trial in patients with CRPC, treatment with
cabozantinib was associated with normalization of radionuclide bone scans in 86% of
patients, improvement of bone pain in 64%
of patients, decline in bone marker levels,
improvement of anemia, and a marked
reduction in bone pain independent of an
effect on PSA.6 Two phase III trials will be
opened for recruitment soon: the first for
an indication for the palliation of pain and
a second for an improvement in overall
survival in patients with CRPC.
Limiting the use of agents shown to
prolong life to men with advanced metastatic CRPC will not considerably decrease
the number of men who die from prostate
cancer. To do so requires the identification
and treatment of those cancers that are destined to metastasize, produce symptoms
and ultimately shorten a patient’s anticipated life expectancy when tumor burdens
JANUARY 2012 | S13
Key advances
■■ Abiraterone acetate and MDV3100
prolong overall survival in patients with
docetaxel-treated castration-resistant
prostate cancer (CRPC)1,3
■■ Radium‑223 prolongs overall survival
in patients with CRPC, bone pain from
disease, and who are docetaxel-treated
or ineligible for docetaxel treatment2
■■ Denosumab delays skeletal-related
events compared with zoledronic acid in
patients with CRPC and bone disease4
are minimal. With this goal but without
firm evidence, the advent of the PSA test
led to widespread screening that has been
accompanied by a marked increase in
men diagnosed with localized disease
who undergo radical prostatectomy and
definitive radiotherapy with the associated
morbidities. Over the past 2 years, the US
PLCO trial showed no reduction in prostate
cancer-specific mortality (PCSM) at 7‑years
after the initiation of PSA screening; the
European Randomized Study of Screening
for Prostate Cancer showed a 20% decrease
in PCSM at 9 years after the initiation of
PSA screening; and the Göteborg randomized screening trial showed a 44% decrease
in PCSM at 14 years after the initiation of
screening.7–9 These results, along with an
analysis of other early detection studies led
the USPSTF to recommend against PSA
screening because of “moderate or high certainty that the service has no net benefit or
that the harms outweigh the bene­fits.”10 The
recommendation triggered a firestorm of
debate, proponents of screening highlighting flaws in methodology and interpretation
and opponents the limited benefits, anxiety
and morbidity a diagnosis produces, and the
high societal costs.
There is strong evidence prostatectomy
and radiotherapy can decrease PCSM by
approximately 50% in clinically localized
disease. However, in low-risk disease where
the 15-year PCSM is <2%, these benefits
become tiny, especially in those with limited
longevity. Despite this, more than half of the
men treated with prostatectomy and radiotherapy in the USA have low-risk disease.
It is important to note that the harms of
screening come not from the PSA test, but
from the chain of actions following PSA
measurement. It has been ingrained into
the psyche of physicians and patients that
all cancers are rapidly lethal unless treated
aggressively. The USPSTF recommendation
is an appropriate pushback against nondiscriminate screening and subsequent
S14 | JANUARY 2012
treatment and argues that early detection
strategies focus on diagnosing cancers that
need treatment to reduce morbidity and
mortality, and avoid finding highly prevalent indolent cancers for which any treatment is overtreatment. The two overarching
objectives in the field are to decrease death
and suffering but, at the same time, decrease
those who are subjected to unnecessary
treatment that can produce significant morbidity. It is clear that neither our current
practice pattern nor the blanket discontinuation of PSA screening can adequately
address these goals. The benefit-to-harm
ratio can be shifted if both screening and
treatment are targeted to high-risk populations and surveillance becomes more widely
adopted for non-aggressive localized tumors.
Human Oncology and Pathogenesis Program
and Department of Medicine, Memorial Sloan–
Kettering Cancer Center, 1275 York Avenue,
New York, NY 10065, USA (Y. Chen, H. I. Scher).
Correspondence to: H. I. Scher
[email protected]
Competing interests
H. I. Scher declares associations with the following
companies: Amgen, Dendreon, Exelixis, Medivation,
Ortho Biotech Oncology Research & Development,
Sanofi-Aventis. See the article online for full details
of the relationships. Y. Chen declares no competing
de Bono, J. S. et al. Abiraterone and increased
survival in metastatic prostate cancer. N. Engl. J.
Med. 364, 1995–2005 (2011).
Parker, C. et al. Overall survival benefit of
radium‑223 chloride (Alpharadin™) in the
treatment of patients with symptomatic bone
metastases in castration-resistant prostate
cancer (CRPC): a phase III randomized trial
(ALSYMPCA) [abstract]. Eur. J. Cancer 47,
a7003 (2011).
3.Medivation. Medivation and Astellas announce
positive survival data from interim analysis of
phase 3 AFFIRM trial of MDV3100 in men with
advanced prostate cancer [online], http://
releaseid=620500 (2011).
4. Fizazi, K. et al. Denosumab versus zoledronic
acid for treatment of bone metastases in men
with castration-resistant prostate cancer: a
randomised, double-blind study. Lancet 377,
813–822 (2011).
5. Yu, E. Y. et al. Phase II study of dasatinib in
patients with metastatic castration-resistant
prostate cancer. Clin. Cancer Res. 15,
7421–7428 (2009).
6. Hussain, M. et al. Cabozantinib (XL184) in
metastatic castration-resistant prostate cancer
(mCRPC): Results from a phase II randomized
discontinuation trial [abstract]. J. Clin. Oncol.
29 (Suppl.), a4516 (2011).
7. Hugosson, J. et al. Mortality results from the
Goteborg randomised population-based
prostate-cancer screening trial. Lancet Oncol.
11, 725–732 (2010).
8. Schröder, F. H. et al. Screening and prostatecancer mortality in a randomized European
study. N. Engl. J. Med. 360, 1320–1328 (2009).
9. Andriole, G. L. et al. Mortality results from a
randomized prostate-cancer screening trial.
N. Engl. J. Med. 360, 1310–1319 (2009).
10. Chou, R. et al. Screening for prostate cancer:
a review of the evidence for the U. S. Preventive
Services Task Force. Ann. Intern. Med. 155,
762–771 (2011).
New therapeutic targets
and treatment strategies
Paula Cramer and Michael Hallek
2011 saw improvements in our understanding of B‑cell malignancies:
insights into the genomic basis of chronic lymphocytic leukemia were
achieved; reduced treatment intensity caused fewer toxic effects in
early-stage Hodgkin lymphoma; first-line rituximab maintenance therapy
improved outcome in follicular lymphoma; and selected patients with
diffuse large-cell lymphoma benefited from the addition of bortezomib.
Cramer, P. & Hallek, M. Nat. Rev. Clin. Oncol. 9, 72–74 (2012); published online 10 January 2012;
Chronic lymphocytic leukemia (CLL) is a
clonal B‑cell disorder initiated by genetic
aberrations in B cells that increase resistance to apoptosis. The survival of CLL cells
is strongly dependent on inter­actions with
surrounding cells (macro­phages and dendritic cells), biochemical cues (chemokines)
and specific receptors expressed on CLL
cells. 2011 saw publication of three important papers providing new insights into
the pathogenesis of CLL. Kikushige et al.1
reported that the capacity to generate clonal
B cells is acquired at the hematopoietic stem
cell (HSC) stage in CLL, suggesting multi­
potent, self-renewing HSCs are involved in
the initiation of CLL. These findings can lead
to the development of new therapies that
target these progenitor CLL cells.
The study by Puente et al.2 reported the
first data on whole-genome sequencing
in CLL. Four cases were analyzed and 46
somatic mutations were identified.2 Four
genes with recurrent mutations were confirmed in 363 patients with CLL: NOTCH1,
XPO1, MYD88 and KLHL6. Mutations in
MYD88 and KLHL6 were predominantly
found in patients with CLL who had a high
number of somatic hypermutations in the
variable region of IGHV, whereas mutations in NOTCH1 and XPO1 were mostly
detected in patients who did not have IGHV
mutations. These findings indicate a role
for mutations in NOTCH1, MYD88 and
XPO1 in the clinical evolution of CLL.
In a similar study on 91 patients with
CLL, Wang et al.3 performed massively parallel sequencing of 88 whole exomes and
genomes. They found nine genes that are
mutated at significant frequencies, including four with established roles (TP53 in 15%
of patients, ATM in 9%, MYD88 in 10%, and
NOTCH1 in 4%) and five genes without
established roles (SF3B1, ZMYM3, MAPK1,
FBXW7, and DDX3X). SF3B1, which functions at the catalytic core of the spliceosome,
was the second most frequently mutated
gene (15% of patients). SF3B1 mutations
occurred primarily in tumors with deletions in chromosome 11q, which are associated with a poor prognosis. The authors also
showed that mutations in SF3B1 induced
alterations in pre–mRNA splicing.
All three studies have considerable clinical implications for CLL. Firstly, the stem
cell origin of CLL offers a plausible explanation for the fact that conventional chemo­
immunotherapies regularly fail to cure CLL.
This type of therapy may not effectively
eradi­cate the HSC pool that regenerates CLL
cells. Secondly, our understanding of the
genetic basis of CLL has improved. These
advances will facilitate the development of
better-targeted therapies for this disease.
In the past 10 years, a remarkable improvement in overall survival was achieved for
several types of B‑cell lymphomas. This
improvement was largely a result of the use
of rituximab, an antibody targeting CD20.
In follicular lymphoma, several randomized trials have demonstrated that addition
of rituximab to different chemotherapeutic
regimens can prolong survival; therefore,
rituximab-based chemoimmuno­therapy is
the current standard first-line treatment.
Disease progression usually occurs 3–5 years
after initial therapy and rituximab maintenance therapy was previously shown to have
a clinical benefit in patients with relapsed
disease. The PRIMA study 4 evaluated the
benefit of rituximab maintenance therapy
KEY ADVANCES IN MEDICINE after first-line treatment for follicular lymphoma. In this study, 1,019 patients who
achieved a response after different chemo­
immunotherapies (rituximab plus cyclophosphamide, vincristine, doxo­r ubicin
and prednisone [R-CHOP]; rituximab plus
cyclophosphamide, vincristine and pred­
nisone; or rituximab plus fludarabine,
cyclo­phosphamide and mitoxantrone) were
randomly assigned to receive 2 years of maintenance therapy with rituximab (375 mg/m²
every 8 weeks) or no further therapy (observation). The maintenance treatment was
well tolerated. Infectious events were more
common in the maintenance arm than the
observation arm, but were mostly mild or
moderate and only a few patients withdrew
from the study because of toxicity. No signifi­
cant decrease in immunoglobulin levels was
observed and patient quality of life was not
affected by the repeated rituximab infusions. Maintenance therapy improved the
quality of response and prolonged progression-free survival and event-free survival.
No effect on overall survival was shown
for rituximab maintenance therapy, which
might be explained by a short follow-up
period and the efficacy of salvage therapies.
However, rituximab should be considered as
maintenance therapy after first-line chemo­
immunotherapy in follicular lymphoma, as
the duration of response to first-line therapy
has prognostic value.5
The outstanding outcome observed in
patients with Hodgkin lymphoma allowed
researchers to consider a reduction in
treatment intensity to reduce both acute
and long-term adverse events, such as
infertility and secondary neoplasias. The
Key advances
■■ Genomic analyses of patients with
chronic lymphocytic leukemia identified
mutations in genes such as Notch1 and
SF3B1 that have prognostic impact2,3
■■ Rituximab maintenance therapy after
first-line treatment with rituximab-based
chemoimmunotherapy significantly
improves the quality of response and
prolongs progression-free survival in
patients with follicular lymphoma4
■■ In advanced-stage, high-risk Hodgkin
lymphoma, reduction in treatment
intensity has not improved outcome and
further analyses are needed to define the
optimal treatment intensity7
■■ Patients with diffuse large B‑cell
lymphoma, in particular, those with a
non-germinal center B‑cell-like subtype,
benefit from the addition of bortezomib
to first-line chemoimmunotherapy9
© Eraxion |
HD10 trial of the German Hodgkin Study
Group (GHSG)6 showed that two cycles of
ABVD (doxo­rubicin, bleomycin, vinblastine and dacarbazine) therapy followed
by 20 Gy of involved–field radio­t herapy
was as efficacious as four cycles of ABVD
and 30 Gy of involved-field radiotherapy
in patients with stage I or II disease and
favorable risk factors. Reduced treatment
intensity may be considered as the new
standard treatment for early-stage, lowrisk Hodgkin lymphoma. By contrast, the
reduction of treatment intensity is debated
in advanced-stage Hodgkin lymphoma
(stage II–IV or IIB with extranodal lesions
or mediastinal mass).
The results of the HD12 trial by Borchmann
et al.7 were slightly disappointing, because
the reduction of the intensity of chemotherapy from eight cycles of high-dose
BEACOPP (bleomycin, etoposide, doxo­
rubicin, cyclophosphamide, vincristine,
procarbazine and prednisolone) to four
cycles of high-dose BEACOPP followed
by four cycles of BEACOPP at the baseline
dose did not reduce the severity of toxicity or
treatment-related mortality, but could possibly reduce efficacy. Thus, the GHSG considers eight cycles of high-dose BEACOPP
as the standard treatment for advancedstage Hodgkin lymphoma and is evaluating
other individualized, PET-controlled dose
reduction strategies in ongoing trials.
Viviani et al.8 compared BEACOPP and
ABVD in advanced-stage Hodgkin lymphoma. In this trial, 331 patients received
either four cycles of high-dose BEACOPP
plus four cycles of BEACOPP at baseline
dose or six to eight cycles of ABVD; each
treatment was followed by involved-field
radiotherapy. The estimated 7‑year rate of
freedom from first progression was 85% in
patients treated with BEACOPP and 73%
in patients treated with ABVD. However,
all patients with a relapse or less than complete response after initial therapy were
treated with multiple cycles of ifosfamidecontaining chemotherapy followed by
JANUARY 2012 | S15
high-dose chemo­t herapy with carmustine, etoposide, cytarabine, and melph­alan
plus autologous hemato­p oietic stem-cell
support. After completion of treatment
including the salvage therapy, the 7‑year
rate of overall survival was 89% in patients
initially treated with BEACOPP compared
to 84% in patients who had received ABVD.
As this difference in overall survival was not
significant, the authors concluded that treatment with BEACOPP resulted in a more
effective disease control, but not in a better
long-term outcome, compared with ABVD.
They argued that while a major proportion of patients could be cured by the initial
BEACOPP therapy, this clinical benefit
might be neutralized by a high rate of longterm adverse events, in particular myelo­
toxicity, infections and secondary neoplasias.
Viviani et al.8 deduced that the BEACOPP
regimen presents an over­treatment strategy for a large number of patients and
recommended ABVD for patients with
advanced-stage or unfavorable Hodgkin
lymphoma. This study has triggered an
intense discussion about the need for an
intensive BEACOPP therapy in advancedstage Hodgkin lymphoma, as few patients
who relapse can be rescued with a high-dose
salvage therapy. However, this trial8 was not
powered to demonstrate a survival difference
between the two treatments and the 7‑year
overall survival differed by 5% in favor of
the BEACOPP regimen. Moreover, median
observation time (61 months) and the
number of patients were low. As a result, an
appropriate treatment intensity in advancedstage or unfavorable Hodgkin lymphoma is
still debated.
The improvements in our understanding of
the molecular pathogenesis of non-Hodgkin
lymphoma have led to the develop­ment of
novel therapeutic agents that interfere with
signaling pathways involved in lympho­
magenesis. For instance, the addition of the
proteasome inhibitor bortezomib enhanced
the activity of chemo­immunotherapy with
R‑CHOP as first-line treatment for diffuse
large B‑cell lymphoma (DLBCL) and mantlecell lymphoma.9 Ruan et al.9 demon­strated
that, in particular, patients with a nongerminal center B‑cell-like (GCB) subtype,
which is usually associ­ated with a worse
outcome, benefit from the addition of borte­
zomib. This increased clinical benefit can
be explained by the fact that the non-GCB
subtype is characterized by a deletion in a
tumor-suppressor gene that activates NF-κB
and this activation is blocked by bortezomib.
Thus, a worse prognosis in this subgroup of
S16 | JANUARY 2012
patients with DLBCL could be overcome by
bortezomib treatment.
Overall, 2011 saw exciting developments
in the understanding and treatment of B‑cell
lymphomas. Whereas research in Hodgkin
lymphoma aims to reduce the adverse effects
of first-line treatment, in most other lymphomas more-effective therapies are needed.
The growing understanding of the pathogenesis of these lymphomas will facilitate
the develop­ment of novel therapeutic agents.
Department of Internal Medicine I and Center
for Integrated Oncology Köln-Bonn, University of
Cologne, Joseph-Stelzmann-Straße. 9, 50924
Cologne, Germany (P. Cramer, M. Hallek).
Correspondence to: M. Hallek
[email protected]
Competing interests
M. Hallek declares associations with the following
companies: Celgene, Mundipharma, Roche. See the
article online for full details of the relationships.
P. Cramer declares no competing interests.
Kikushige, Y. et al. Self-renewing hematopoietic
stem cell is the primary target in pathogenesis
of human chronic lymphocytic leukemia. Cancer
Cell 20, 246–259 (2011).
Puente, X. S. et al. Whole-genome sequencing
identifies recurrent mutations in chronic
lymphocytic leukaemia. Nature 475, 101–105
Wang, L. et al. SF3B1 and other novel cancer
genes in chronic lymphocytic leukemia. N. Engl.
J. Med.
Salles, G. et al. Rituximab maintenance for
2 years in patients with high tumor burden
follicular lymphoma responding to rituximab
plus chemotherapy (PRIMA): a phase 3,
randomised controlled trial. Lancet 377, 42–51
Bachy, E. et al. Long-term follow-up of patients
with newly diagnosed follicular lymphoma in the
prerituximab era: effect of response quality on
survival - a study from the Groupe d’Etude des
Lymphomes de l’Adulte. J. Clin. Oncol. 28,
822–829 (2010).
Engert, A. et al. Reduced treatment intensity in
patients with early-stage Hodgkin’s lymphoma
N. Engl. J. Med. 363, 640–652 (2010).
Borchmann, P. et al. Eight cycles of escalateddose BEACOPP compared with four cycles of
escalated-dose BEACOPP followed by four
cycles of baseline-dose BEACOPP with or
without radiotherapy in patients with advancedstage Hodgkin’s lymphoma: final analysis of
the HD12 trial of the German Hodgkin Study
Group. J. Clin. Oncol. 29, 4234–4242 (2011).
Viviani, S. et al. ABVD versus BEACOPP for
Hodgkin’s lymphoma when high-dose salvage
is planned. N. Engl. J. Med. 365, 203–212
Ruan, J. et al. Bortezomib plus CHOP-rituximab
for previously untreated diffuse large B‑cell
lymphoma and mantle cell lymphoma. J. Clin.
Oncol. 29, 690–697 (2011).
A new paradigm tumor for drug
Alexander M. M. Eggermont and Caroline Robert
Melanoma has emerged as the paradigm tumor for drug development
through mutation-targeted therapies (inhibitors targeting BRAF, MEK, and
c‑KIT) and immunotherapy. Exploring the combinations of both approaches
is a challenge that will require scientific rationale and the cooperation of
the pharmaceutical industry. But, with these challenges comes another
opportunity to change the paradigms in drug development.
Eggermont, A. M. M. & Robert, C. Nat. Rev. Clin. Oncol. 9, 74–76 (2012); published online 10 January 2012;
For the past 40 years no treatments developed for melanoma have significantly
improved survival over dacarbazine, a drug
with a response rate of around 10%. In the
past decade, we have witnessed—in the fields
of mutation-driven drug development and
immuno­modulation—the establishment of
treatments actively impacting survival. This
is just the beginning of a promising journey
to develop rational and biology-driven treatments for patients with melanoma with clini­
cally significant impact to greatly change the
thera­peutic landscape.
The treatment of melanoma has evolved
from using non-selective inhibitors to more
selective agents, with BRAF, MEK and c‑KIT
inhibitors leading the field. BRAF mutations
are the most frequently occurring and most
important mutations that provide drugable
targets in melanoma. The first selective
BRAF inhibitor developed in the clinical
setting was vemurafenib. 1 In a randomized phase III trial published in 2011, this
drug led to major tumor responses in 50%
of patients, and minor responses in >30% of
patients.2 The most common adverse events
were arthralgia, fatigue, and cutaneous
mani­festations such as rash, photo­sensitivity,
pruritis, and squamous-cell carcinoma of
the keratoacanthoma-type in around 20%
of patients.1,2 The phase III BRIM‑3 trial,
in which vemurafenib was compared with
dacarbazine in treatment-naive patients,
showed an improvement in progressionfree survival (PFS) of 5.3 months for vemurafenib. The impact on overall survival has
not yet been established and is probably
similar because, as the trial was unblinded at
the first interim analysis and cross over was
allowed, changes in overall survival will be
somewhat compromised by this cross over.
Responses to vemurafenib are immedi­
ate and evident on PET scans within
1–2 weeks after treatment commencement.
Disappointing, however, is the short median
duration of these responses, and essentially all patients relapse, 50% in the first
5–6 months, 50% any time after 6 months.
Various mechanisms of drug resistance have
been described, mostly reflecting primary
resistance (present from the beginning) and
heterogeneity of the tumors.
As BRAF inhibition leads to reactivation
of the MAPK pathway and enhances cell
growth through CRAF, combining a BRAF
inhibitor with a MEK inhibitor may both
prolong PFS and prevent the appearance
of squamous-cell carcinomas. The report
in 2011 of the use of a BRAF inhibitor
(GSK436) combined with a MEK inhibitor
(GSK212), clearly indicate these beneficial
effects.3 NRAS mutations occur in about
20% of melanomas but RAS remains a rather
elusive target in cancer, with no available
drugs that can directly antagonize its signaling activity. Dual targeting of the MAPK and
PI3K pathways might abrogate the effect of
NRAS mutation (Figure 1).
Mutations in c‑KIT are also important in
melanoma, especially in mucosal and acral
melanomas, in which mutations in c‑KIT
are found in 20–30% of the cases. These
types of melanoma represent less than 20%
of all melanomas in the white population
in the Western world, but incidence in the
Key advances
■■ BRAF and MEK inhibitors are key agents
in treating BRAF-mutated melanomas,
and their combination is essential to
optimize results2,3
■■ Immunomodulators, such as antibodies
blocking CTLA‑4 and PD‑1, establish
immunotherapy as a key component of
anti-tumor strategies, required for
long-term responses and potential cure6
■■ Ulcerated primary melanoma is a distinct
biologic entity that indicates sensitivity
to adjuvant IFN therapy determining when
such treatment should be indicated10
KEY ADVANCES IN MEDICINE Asian population is >70%.4 Response rates
and PFS rates with imatinib are in the range
of 15–20%, which is rather disappointing
compared to the activity of this inhibitor in
GIST tumors, and a reflection of a different
pattern of mutations.4 These mutations make
melanoma a tumor that often presents with
multiple perturbed pathways and with many
routes to escape from single or multiple drug
exposure. Therefore, although combinations
of drugs targeting two proteins in the same
signaling pathway or in different pathways
are currently being explored, it will be critical to combine targeted drugs with drugs
that modulate the immune system.
The cytotoxic T‑lymphocyte–associated
antigen 4 (CTLA‑4) ligand is a negative
regulator of T cells; therefore, blocking
CTLA‑4 action augments T‑cell activation
and proliferation, and inhibits immune
system tolerance. The anti-CTLA‑4 monoclonal antibody ipilimumab has recently
been assessed in randomized phase III
trials. Improved overall survival (4 months)
was demonstrated for ipilimumab alone
or in combination with the gp100 peptide
vaccine compared with the vaccine alone in
patients who did not respond to prior treatment.5 Improved overall survival was also
demonstrated for ipilimumab in the firstline setting when combined with dacarbazine (2.1 months).6 Thus, in 2011, the FDA
approved ipilimumab for advanced melanoma at a dose of 3 mg/kg, administered
every 3 weeks for a total of four doses for
all patients with advanced melanoma in
the first-line and second-line and settings.
In the pivotal trial assessing the combination of dacarbazine with ipilimumab as
first-line treatment at a dose of 10 mg/kg, a
high percentage of liver enzyme alterations
was observed.6 This adverse effect resulted
in significant percentage of patients abandoning the treatment, which is believed to
have reduced the impact of ipilimumab.
Because ipilimumab reactivates immune
responses that have been silenced, there is
a risk that it may reactivate lingering subclinical auto­immune responses in certain
patients, resulting in immune-related
adverse events. Some of the most prominent
were colitis, dermatitis, and hypo­physitis,
all of which must be closely monitored as
patients may need the immediate use of
corticosteroids. Ipilimumab has a modest
response rate of around 10%, but responses
are almost always durable with a median
of 20 months. Moreover, there was a clear
separation of the survival curves in both
trials, indicating durable responses that last
PI3K pathway
MAPK pathway
Cell growth, proliferation, survival
Figure 1 | Dual targeting of the MAPK and
PI3K pathways in melanoma treatment. This
figure illustrates how combining inhibition of
BRAF and MEK (intra-pathway combined
blocking) with AKT (inter-pathway
combined blocking) might abrogate the effect
of NRAS mutation.
for >3 years—with a significant number
of patients that have not relapsed after
5 years—which may indicate potential cure.
A reason why patients might have stopped
taking the treatment is that it takes time for
the responses to ipilimumab to develop.
Responses may even be preceded by progression of disease at 6 or 8 weeks, which
might reflect lymphocyte infiltration of the
lesions that will subsequently regress. These
observations have led to the development of
novel, immune-related response criteria that
might more accurately describe response
to immunotherapy and avoid premature
treatment cessation in patients with disease
progression before response.7
Studies that combine BRAF inhibitors
with cytokines such as IFN, IL2, GM-CSF
and anti-angiogenic agents are currently
under clinical development. Although early
reports indicate improved response rates
to these combinations, the interactions
between these drugs are poorly understood, and in the absence of clear biomarkers to provide rational guidance as to which
combinations should be used and how,
serious efforts should be made in this area
to allow for rational plans and progress.
Green et al.8 have proposed that various
chemotherapeutics may exert their action
through immunogenic cell death (leading
to antigen presentation and stimulating the
immune system) or tolerogenic cell death
(necrosis without immune-system priming
but instead leading to immuno­logical tolerance).9 This approach could provide rational
guidance to explore combination treatments.
JANUARY 2012 | S17
New immunomodulators are being tested
in early-phase studies. Most promising are
the results seen with the monoclonal antibody that acts against the programmed
death‑1 receptor (PD-1R), the ligand of
which (PD-1L) can be expressed directly
on melanoma cells. Good responses have
been observed in phase I studies with possibly fewer accompanying immune-related
adverse events than observed with ipili­
mumab treatment. 9 Another immuno­
modulator, pegylated IFN‑α-2b, was
approved in 2011 by the FDA as adjuvant
therapy for patients with stage III lymphnode positive disease after lymph-node
dissection, on the basis of a significant
improvement of relapse-free survival (RFS).10
This EORTC 18991 trial (n = 1,256 patients)
had been preceded by the EORTC 18952
trial, which evaluated intermediate doses
of regular IFN‑α-2b in 1,388 patients.10 In
both trials, patients were stratified according to nodal status and presence or absence
of ulceration in the primary tumor. In both
trials, low tumor-stage and the presence
of ulceration significantly correlated with
improved outcome in the IFN-treatment
arms, as reported by the recently published
meta-analysis of these trials.10 In patients with
an ulcerated primary tumor and stage IIB or
limited stage III disease (positive sentinel
node) there was a highly significant reduction
(30–40%) of the relative risk of RFS, distant
metastasis-free survival and overall survival.
In patients with stage III–N2 disease, these
benefits were progressively lost. This finding
strongly indicates that ulcerated melanoma
represents a distinct biologic entity that will
be assesed in the EORTC 18081 trial, which
will evaluate pegylated-IFN-α-2b versus
observation in node-negative patients with
ulcerated primary melanomas. Adjuvant
treatment with ipilimumab in patients with
advanced stage IIIb/IIIc is under evalu­
ation in the EORTC 18071 trial that has now
completed patient accrual.
There has never been a more exciting and
dynamic time in the treatment of melanoma.
To explore all potential combinations is
impossible. Thus, we need to stick to a rational approach—guided by principles such as
immunogenic cell death—and translational
research programs to identify biomarkers
predictive for outcomes of both the immuno­
therapeutic and the mutation-driven drug
development approach. Without such an
approach we could revert to the empiric ways
that characterized clinical drug development
of the past, and that is no longer acceptable
in the 21st century.
S18 | JANUARY 2012
Institut de Cancérologie Gustave Roussy, 114
Rue Edouard Vaillant, 94805 Villejuif, ParisSud, France (A. M. M. Eggermont, C. Robert)
Correspondence to: A. M. M. Eggermont
[email protected]
Competing interests
The authors declare associations with the following
companies: Bristol-Myers Squibb, GlaxoSmithKline,
Merck, Roche. See the article online for full details
of the relationships.
Flaherty, K. T. et al. Inhibition of mutated,
activated BRAF in metastatic melanoma.
N. Engl. J. Med. 363, 809–819 (2010).
Chapman, P. B. et al. Improved survival with
vemurafenib in melanoma with BRAF V600E
mutation. N. Engl. J. Med. 364, 2507–2516
Infante, J. R. et al. Phase I/II study to assess
safety, pharmacokinetics, and efficacy of the
oral MEK 1/2 inhibitor GSK1120212
(GSK212) dosed in combination with the oral
BRAF inhibitor GSK2118436 (GSK436)
[abstract]. J. Clin. Oncol. 29 (Suppl.),
CRA8503J (2011).
Carvajal, R. D. et al. KIT as a therapeutic target
in metastatic melanoma. JAMA. 305,
2327–2334 (2011).
5. Hodi, F. S. et al. Improved survival with
ipilimumab in patients with metastatic
melanoma. N. Engl. J. Med. 363, 711–723
6. Robert, C. et al. Ipilimumab plus dacarbazine
for previously untreated metastatic melanoma.
N. Engl. J. Med. 364, 2517–2526 (2011).
7. Hoos, A. et al. Improved endpoints for cancer
immunotherapy trials. J. Natl Cancer Inst. 102,
1388–1397 (2010).
8. Green, D. R., Ferguson, T., Zitvogel, L. &
Kroemer, G. Immunogenic and tolerogenic cell
death. Nat. Rev. Immunol. 9, 353–363 (2009).
9. Brahmer, J. R. et al. Phase I study of single-agent
anti-programmed death‑1 (MDX‑1106) in
refractory solid tumors: safety, clinical activity,
pharmacodynamics, and immunologic correlates.
J. Clin. Oncol. 28, 3167–3175 (2010).
10. Eggermont, A. M. et al. Ulceration and stage are
predictive of interferon efficacy in melanoma:
results of the phase III adjuvant trials EORTC
18952 and EORTC 18991. Eur. J. Cancer.
doi:10.1016/j.ejca.2011.09.028 BONE CANCER IN 2011
Prevention and treatment of bone
Robert E. Coleman
Bone-targeting treatments have transformed the quality of life of patients
with metastatic bone disease. 2011 saw the emergence of denosumab—a
RANK ligand-specific antibody—as a more-effective alternative treatment
to bisphosphonates and of data on the use of bone-targeting treatments to
prevent metastasis from breast and prostate cancers.
Coleman, R. E. Nat. Rev. Clin. Oncol. 9, 76–78 (2012); published online 20 December 2011;
Bone disease accounts for substantial morbidity and mortality in patients with cancer.
Advanced-stage solid tumors, notably
those arising from the breast and prostate,
and multiple myeloma are associated with
a heavy burden of skeletal disease, with
potentially debilitating or life-limiting skeletal-related events (SREs). As a result, bonetargeting treatments have been introduced
alongside specific anticancer treatments
to minimize skeletal morbidity and preserve the patients’ quality of life and physical function. An increased understanding
of the interplay among the disseminated
tumor cells, the bone marrow and both hostderived and tumor-derived growth factors
has shown that the complex interactions in
the bone marrow micro­environment provide
candidate therapeutic targets. Modification
of these inter­a ctions is emerging as an
important anticancer treatment strategy and,
in 2011, considerable developments have
Key advances
■■ Denosumab provides a more effective,
convenient and well-tolerated alternative
treatment to zoledronic acid for the
prevention of skeletal morbidity2,3
■■ The adjuvant use of zoledronic acid in
early-stage breast cancer should be
restricted to postmenopausal women, as
women with residual ovarian function do
not benefit from this treatment7
■■ Denosumab delays the development
of bone metastases in men with
castration‑resistant prostate cancer10
occurred in the treatment and prevention of
bone metastases.
In the past 20 years, bisphosphonates have
been the choice of treatment for preventing
skeletal morbidity associated with malignant bone disease. However, recognition of
the importance of the receptor activator of
nuclear factor κ‑B ligand (RANKL) in bone
metastasis has provided a new specific therapeutic target. Denosumab is a fully human,
synthetic IgG2 antibody that is administered via subcutaneous injection and binds
with high affinity to RANKL, preventing
the interaction of RANKL and RANK and
inhibiting osteoclast function.
In 2011, denosumab became available for
clinical use after the publication of three
large randomized double-blind registration
studies. These trials included 5,723 patients
with bone metastases secondary to breast
cancer (n = 2,046), 1 castration-resistant
prostate cancer (n = 1,901) 2 and other
advanced-stage solid tumors including
non-small-cell lung cancer (n = 702), other
types of solid tumors (n = 894) and myeloma
(n = 180).3 In these studies, patients were
randomly assigned to receive denosumab
or zoledronic acid every 3–4 weeks, plus
calcium and vitamin D supplements. The
primary end point was time to first onstudy SRE. Stopeck and colleagues1 showed
that denosumab was statistically superior
to zoledronic acid in delaying the first SRE
in patients with bone meta­stasis secondary
to breast cancer; the median time to first
on-study SRE was 26.4 months in the zoledronic acid-treated group and has not been
reached in the denosumab-treated group.
Similarly, Fizazi and colleagues2 reported
that denosumab was superior to zoledronic
acid in delaying the first SRE in patients
with bone meta­stasis secondary to prostate
cancer; the median time to first on-study
SRE was 17.1 months in the zoledronic
acid-treated group and 20.7 months in the
denosumab-treated group.
In patients with bone metastasis secondary to advanced solid tumors and myeloma,
Henry and colleagues3 showed that denosumab was not inferior to zoledronic acid in
delaying the first SRE, but was not superior
either. In an analysis of multiple events of
SREs, denosumab was superior to zoledronic acid in patients with breast cancer 1
and prostate cancer 2 (Table 1). In all three
studies, denosumab was well tolerated,
and a key difference in the safety profiles
of denosumab and zoledronic acid was the
lack of an effect on kidney function with
denosumab, which obviates the need for
monitoring renal function that is mandated
when treating patients with intravenous
bisphosphonates. Similar to the use of intravenous bisphosphonates used in oncology
settings, osteonecrosis of the jaw (ONJ) was
the most important adverse event associated
with denosumab, but it was uncommon.
The cumulative incidence rates of ONJ were
KEY ADVANCES IN MEDICINE Table 1 | Phase III registration trials of denosumab in advanced-stage cancer
Disease setting
End point
Hazard ratio
95% CI
P (relative to ZA)
Breast cancer1
Time to 1st SRE
Time to 1st and subsequent SRE
P = 0.01
P = 0.001
Time to 1st SRE
Time to 1st and subsequent SRE
P = 0.008
P = 0.008
OST and MM3
Time to 1st SRE
Time to 1st and subsequent SRE
P = 0.03
P = 0.14
Time to 1st SRE
Time to 1st and subsequent SRE
P <0.001
P <0.001
Abbreviations: CI, confidence interval; CRPC, castration-resistant prostate cancer; MM, multiple myeloma; OST, solid tumors
other than breast or prostate cancer; SRE, skeletal-related event; ZA, zoledronic acid.
similar for both treatments: 0.5% and 0.8%
in year 1, 1% and 1.8% in year 2, and 1.3%
and 1.8% in year 3 for zoledronic acid and
denosumab, respectively.4
The use of denosumab in patients with
advanced-stage cancer is likely to be
influenced by cost, in particular, because
denosumab did not improve either progression-free survival or overall survival
compared with zoledronic acid. However,
collectively these trials show that denosumab is slightly more effective and convenient to administer, has some advantages in
terms of safety over the current gold standard of treatment and represents an important advancement in the quality of life of
patients with advanced-stage cancer.
Other agents targeting the bone meta­
stasis pathway are in development. The
most interesting novel strategy for bone
metastasis treatment during 2011 was
reported by Parker and colleagues. 5 In
this study, 922 men with advanced-stage
castration-resistant prostate cancer who had
not responded or were considered unsuitable for chemotherapy were randomly
assigned (2:1) to receive a bone-localizing
α‑particle-emitting agent, 223RaCl 2 (also
called alpharadin) or placebo. Median
overall survival increased from 11.2 months
to 14 months in patients receiving alpharadin (hazard ratio [HR] 0.695; 95% CI
0.552–0.875, P = 0.002).5 In addition to the
effects of bone-targeting treatments on the
risk of skeletal complications in patients
with cancer, these agents could modify the
course of the disease via inhibitory effects
on the ‘vicious cycle’ of growth factor
and cytokine signaling between tumor
and bone cells within the bone marrow
In patients with early-stage breast
cancer, inconsistent effects on disease outcomes were previously reported in clinical
trials of adjuvant bisphosphonate treatment. However, in 2009, the results of the
ABCSG‑12 trial 6 stimulated a considerable amount of interest in the field of targeting bone micro­environment to modify
the course of early-stage breast cancer.
Zoledronic acid treatment every 6 months
for 3 years significantly improved diseasefree survival (DFS) in premenopausal
women with endocrine-sensitive earlystage breast cancer previously treated with
ovarian suppression therapy (goserelin)
and tamoxi­fen or anastrozole.6 In 2011, the
updated results of this trial7,8 showed that
DFS was still improved with adjuvant zoledronic acid treatment and overall survival
was also increased. Notably, these bene­
ficial effects of zoledronic acid were only
seen in patients over 40 years of age; goserelin treatment induced post­menopausal
levels of circulating reproductive hormones
consistently in these patients.
Improvements in DFS and overall survival was also reported in postmenopausal
women in the AZURE trial, a large randomized adjuvant trial that was designed
to determine whether treatment with zoledronic acid in addition to adjuvant therapy
improved disease outcomes in most women
with early-stage breast cancer.9 In this study,
3,360 patients were randomly assigned
to receive standard adjuvant systemic
therapy with or without zoledronic acid
every 3–4 weeks for six doses, then every
3–6 months thereafter, for a total of 5 years.9
After a median follow-up of 59 months, no
significant difference was seen in DFS or
overall survival. However, in a prespecified
subgroup analysis, a significant increase
in DFS was detected with zoledronic acid in
1,041 postmenopausal women (who were
postmenopausal for >5 years at the time of
diagnosis): 5‑year invasive DFS was 71% in
the control arm and 78.2% in those treated
with zoledronic acid (adjusted HR 0.75;
95% CI 0.59–0.96, P = 0.02).9 In addition,
zoledronic acid improved overall survival
in this group of women, with 5‑year overall
JANUARY 2012 | S19
survival rates of 79% in the control group
and 85% in patients treated with zoledronic
acid (adjusted HR 0.74; 95% CI 0.55–0.98,
P = 0.04). The data from both ABCSG‑12
and AZURE6,9 trials suggest that the bene­
ficial effects of adjuvant zoledronic acid
treatment are dependent on an environment
with low reproductive hormone levels. The
mechanisms underlying this observation
are under investigation.
Prostate cancer has the propensity to
metastasize almost exclusively to the bone
and, therefore, provides the ideal clinical
setting for the evaluation of bone-targeting
treatments to modify the disease course. In a
study by Smith and colleagues,10 1,432 men
with non-metastatic castration-resistant
prostate cancer at high risk for bone meta­
stasis (determined by either a PSA level
≥8.0 ng/ml and/or PSA doub­ling time of
≤10 months) were randomly assigned to
receive a monthly treatment of denosumab
(120 mg) or placebo. Bone metastasis-free
survival was significantly increased with
denosumab by a median of 4.2 months (HR
0.85; 95%CI 0.73–0.98, P = 0.028), and the
onset of first symptomatic bone metastases
was delayed. This effect, however, did not
translate into any improvement in overall
survival, and as the cumulative incidence of
ONJ was high (4% after 3 years), whether
the efficacy of denosumab to reduce bone
meta­stasis is sufficient to change clinical
practice is unclear.
Overall, 2011 saw important developments in the prevention and management
of bone metastasis. Integrating the current
bone-targeting treatments in routine clinical practice and achieving a better understanding of the cellular processes involved
in bone metastasis are the challenges for
the future.
Sheffield Cancer Research Center, Academic
Unit of Clinical Oncology, Weston Park Hospital,
Whitham Road, Sheffield S10 2SJ, UK.
[email protected]
Competing interests
The author declares associations with the following
companies: Amgen and Novartis. See the article
online for full details of the relationships.
Stopeck, A. T. et al. Denosumab compared with
zoledronic acid for the treatment of bone
metastases in patients with advanced breast
cancer: a randomized, double-blind study.
J. Clin. Oncol. 28, 5132–5139 (2010).
Fizazi, K. et al. Denosumab versus zoledronic
acid for treatment of bone metastases in men
with castration-resistant prostate cancer: a
randomised, double-blind study. Lancet 377,
813–822 (2011).
Henry, D. H. et al. Randomized, double-blind
study of denosumab versus zoledronic acid in
the treatment of bone metastases in patients
with advanced cancer (excluding breast and
prostate cancer) or multiple myeloma. J. Clin.
Oncol. 29, 1125–1132 (2011).
4. Saad, F. et al. Incidence, risk factors, and
outcomes of osteonecrosis of the jaw:
integrated analysis from three blinded activecontrolled phase III trials in cancer patients
with bone metastases. Ann. Oncol. http://
5. Parker, C. et al. Overall survival benefit of
radium‑223 chloride (alpharadin) in the
treatment of patients with symptomatic bone
metastases in castration-resistant prostate
cancer (CRPC): a phase III randomised trial
(ALSYMPCA) [abstract]. Eur. J. Cancer
47 (Suppl. 2), a3 (2011).
6. Gnant, M. et al. Endocrine therapy plus
zoledronic acid in premenopausal breast
cancer. N. Engl. J. Med. 360, 679–691 (2009).
7. Gnant, M. et al. Adjuvant endocrine therapy
plus zoledronic acid in premenopausal women
with early-stage breast cancer: 62-month
follow-up from the ABCSG‑12 randomised trial.
Lancet Oncol. 12, 631–641 (2011).
8. Gnant, M. et al. Overall survival with adjuvant
zoledronic acid in patients with premenopausal
breast cancer with complete endocrine
blockade: Long-term results from ABCSG‑12
[abstract]. J. Clin. Oncol. 29 (Suppl. 15), a520
9. Coleman, R. E. et al. Breast-cancer adjuvant
therapy with and without zoledronic acid.
N. Engl. J. Med. 365, 1396–1405 (2011).
10. Smith, M. R. et al. Denosumab and bone
metastasis-free survival in men with castrationresistant prostate cancer: results of a global
phase 3, randomised, placebo-controlled trial.
Don’t miss out!
Sign up to free e-ALERTS and be notified when new articles are published
in Nature Reviews Clinical Oncology. Research Highlights, News & Views, Reviews,
and Perspectives—you won’t miss a thing!
S20 | JANUARY 2012
Thyroid function—effects on mother
and baby unraveled
Anthony P. Weetman
The complex relationship between pregnancy and thyroid function, and its clinical effect on mother and baby,
continued to stimulate research in 2011. Key advances were made on three important issues: how long
maternal thyroid function affects fetal thyroid hormone levels; whether thyroid autoimmunity affects pregnancy
outcome; and the prevalence of permanent hypothyroidism after postpartum thyroiditis.
Weetman, A. P. Nat. Rev. Endocrinol. 8, 69–70 (2012); published online 6 December 2011; doi:10.1038/nrendo.2011.217
KEY ADVANCES IN MEDICINE possibility that maternal thyroid function
throughout pregnancy, rather than just the
first trimester, is an important determinant
of gestational outcome. Detailed sampling of
maternal thyroid function is required to
investigate this hypo­thesis further.
In a related study from 2011, maternal thy­
roid function variables measured at 28 weeks
gestation were compared with those measured in the umbilical cord at birth in 616
preg­nancies in the UK.3 Maternal free T4 and
cord free T4 values showed a positive correlation, and maternal TSH levels were inversely
corre­lated with cord free T4 levels. This evi­
dence by Shields and co-investigators is con­
sistent with that of the Dutch study 1 and
sug­gests that maternal thyroid function has
a measurable effect on fetal thyroid hormone
levels throughout pregnancy.
Shields et al. also examined the effect of
thyroid hormone levels on various para­
meters of fetal growth. Cord free T4 levels
were associated with birth weight, length and
skin-fold thickness. These results imply that
fetal thyroid hormone levels have an important role in regulating fetal growth, especially
when taken in conjunction with the results
of animal experiments. 4 An unexplained
finding was that birth weight was negatively
associated with maternal free T4 levels, given
that maternal and cord free T4 levels were
positively associated.
Together, these two studies highlight the
intricacy of the maternal–fetal relationship
with regard to thyroid function and its clinically important effects on fetal growth and
ges­tational age. Studies with detailed assessment of maternal thyroid function throughout the third trimester that incorpo­r ate
assess­ment of placental deiodinase activity
are needed. Variable iodine intake is also a
con­founding factor that needs addressing,
given the surprising discovery in 2011 of
inade­quate dietary supplies in the UK and
in other countries.5
Multiple studies have shown that euthyroid
pregnant woman with positive thyroid peroxidase (TPO) antibodies have an increased
risk of spontaneous mis­carriage and possibly
preterm delivery. Two mechanisms might be
responsible. Firstly, even though ‘euthyroid’,
these women may have subtle impairment
of thyroid function, and the effects of maternal thyroid function on the fetus, as discussed
above, are responsible. Secondly, thyroid
auto­immunity dis­torts the immunological
balance required to maintain tolerance to
the fetal semi-allograft.6
In 2011, a study of 245 TPO-antibodypositive women with TSH levels <2.5 mU/l
has shown that very preterm delivery
(<34 weeks of gestation) and respiratory
dis­tress at birth were more common than
© Gabriel Blaj |
The fetal thyroid gland starts to develop at
12 weeks of gestation, and for the first half of
pregnancy, transplacental passage of maternal thyroid hormone is essential for normal
fetal development. Failure to deliver sufficient
thyroid hormone to the fetus causes impaired
neurological development, although whether
other elements of fetal development are also
affected remains unclear. The essentially
normal outcome in neonates with congenital
hypothyroidism promptly treated with levothyroxine after birth suggests that maternal
thyroid hormones also have a fetal role in the
second half of pregnancy.
In a prospective survey of 886 pregnant
Dutch women published in 2011, neonates
born to mothers who had high-normal TSH
levels (above the 97.5th percentile) one or
more times during pregnancy had lower
total T4 levels when routine screening was
performed for congenital hypothyroidism
than neonates born to mothers whose TSH
levels had remained below the 97.5th percentile.1 This result contrasts with that of a
previous study that showed no relationship
between maternal and neonatal thyroid function,2 a difference possibly related to the use
of pregnancy-specific refer­ence ranges for
TSH measurement in the 2011 study.
Although the women with high-normal
TSH levels had normal free T4 levels, the TSH
levels might reflect a resetting of the
­pituitary–thyroid axis, which resulted in
less T4 being available to cross the placenta.
Alternatively, an unsuspected increase in
placental deiodinase activity might have
been responsible for low T4 availability to
the fetus. Low gestational age was associated with low neonatal thyroxine levels in
this and previous studies, which raises the
JANUARY 2012 | S21
in pregnancies of women who were negative for TPO antibodies.7 No association
with TPO-antibody positivity was observed
for 12 other adverse outcomes, including
preterm delivery between 34–37 weeks.
Although corrected for multiple comparisons, the statistical tests applied were onetailed rather than two-tailed. Given that the
association with respiratory distress is novel,
this finding could be a type 1 statis­tical error.
The key feature of this study was that all the
par­ticipants were chosen to have low-normal
TSH levels, and thus the results appear to
show that autoimmunity per se is responsible for these adverse out­comes. How­ever,
miscarriage rates, pre­viously shown to be
increased in TPO-antibody-positive women,
were not specifically increased and although
below 2.5 mU/l, TSH levels were significantly
higher in the TPO-antibody-positive women
than in control women. Although the results
clearly show that a trial of levothyroxine
during pregnancy is warranted in this subset
of TPO-antibody-positive women with lownormal TSH levels, the pathogenic mechanism behind the adverse outcomes of such
pregnancies is still in question.
Postpartum thyroiditis is a transient flaring
of pre-existing autoimmune thyroidi­tis in
the year after delivery, which seems to be a
consequence of the restoration of nor­mal
immune function after a period of altered
immuno­r egulation during pregnancy. 6
Tran­sient abnormalities of thyroid dysfunction are common in postpartum thyroiditis; however, as has been clear for over two
decades, 2–21% of such patients have permanent hypo­thyroidism at the end of the year
after delivery, and this figure rises during
follow-up over subsequent years. A survey of
4,394 women from Southern Italy reported
in 2011 that 3.9% had postpartum thyroiditis
and 54% of these were hypo­thyroid at 1 year
after delivery.8 This figure for the prevalence of post­partum thyroiditis is rather low
compared with that in many other studies, a
difference that could be related to the exclusion of women with any thyroid dysfunction
during the first trimester of pregnancy in the
2011 study. In addition, the women in this
Italian study were only screened twice postpartum, so some cases of dysfunction might
have been missed. By contrast, the figure for
the prevalence of hypothyroidism at 1 year
is strikingly high. This finding might also be
related to the study design—if women with
milder forms of postpartum thyroiditis were
excluded because of limited sampling, this
exclusion could have distorted the overall
prevalence results, given that such women are
S22 | JANUARY 2012
Key advances
■■ Maternal thyroid function has a
measurable effect on fetal thyroid
hormone levels throughout pregnancy1,3
■■ Fetal thyroid hormone levels have an
important role in regulating fetal growth,
as measured by birth weight, length and
skin-fold thickness3
■■ Very preterm delivery (<34 weeks of
gestation) and respiratory distress may
be associated with maternal thyroid
peroxidase antibody positivity7
■■ Hypothyroidism following an episode of
postpartum thyroiditis could be more
common in some populations than
previously described, and the risk of this
outcome correlates with the severity of
the preceding destructive thyroiditis8,9
probably less likely to become hypothyroid.
In this context, a related study has shown
that low urinary iodine levels postpartum are
associated with long-term hypothyroidism,
probably because the low iodine excretion
reflects more severe destructive thyroiditis
that causes a discharge of thyroid iodine
content.9 Together, these results reinforce
the need for careful clinical evaluation of all
women who have postpartum thyroiditis at
1 year after delivery and annually thereafter.
In summary, five studies from 2011 have
shown a pervasive effect of maternal thyroid
function on the fetus throughout pregnancy.
Abnormalities in maternal or fetal thyroid
function are associated with an increasing
number of adverse pregnancy outcomes. The
additional role of autoimmunity per se on
pregnancy, rather than through its effect
on maternal thyroid function, remains to be
fully established. However, no doubt exists
that changes to the maternal autoimmune
response in the postpartum period have
considerable clinical effects on the mother’s
thyroid reserve.
Department of Human Metabolism, University
of Sheffield, The Medical School, Beech Hill
Road, Sheffield S10 2RX, UK.
[email protected]
Competing interests
The author declares no competing interests.
Kuppens, S. M. et al. Neonatal thyroid
screening results are related to gestational
maternal thyroid function. Clin. Endocrinol. (Oxf.)
75, 382–387 (2011).
Oken, E. et al. Neonatal thyroxine, maternal
thyroid function, and child cognition. J. Clin.
Endocrinol. Metab. 94, 497–503 (2009).
Shields, B. M., Knight, B. A., Hill, A.,
Hattersley, A. T. & Vaidya, B. Fetal thyroid
hormone level at birth is associated with fetal
growth. J. Clin. Endocrinol. Metab. 96,
E934–E938 (2011).
Bassett, J. H. et al. A lack of thyroid hormones
rather than excess thyrotropin causes abnormal
skeletal development in hypothyroidism. Mol.
Endocrinol. 22, 501–512 (2008).
Vanderpump, M. P. et al. Iodine status of UK
schoolgirls: a cross-sectional survey. Lancet
377, 2007–2012 (2011).
Weetman, A. P. Immunity, thyroid function and
pregnancy: molecular mechanisms. Nat. Rev.
Endocrinol. 6, 311–318 (2010).
Negro, R. et al. Thyroid antibody positivity in the
first trimester of pregnancy is associated with
negative pregnancy outcomes. J. Clin.
Endocrinol. Metab. 96, E920–E924 (2011).
Stagnaro-Green, A. et al. High rate of persistent
hypothyroidism in a large-scale prospective
study of postpartum thyroiditis in southern Italy.
J. Clin. Endocrinol. Metab. 96, 652–657 (2011).
Stuckey, B. G. et al. Low urinary iodine
postpartum is associated with hypothyroid
postpartum thyroid dysfunction and predicts
long-term hypothyroidism. Clin. Endocrinol. (Oxf.)
74, 631–635 (2011).
Towards a better understanding
of causation and consequences
Michael Stowasser
With primary aldosteronism now widely acknowledged as common and
associated with both hypertension-related and non-hypertension-related
pathology, research interest into its causes and consequences continues
to grow. In 2011, major breakthroughs occurred in understanding the
role and nature of underlying genetic disturbances and elucidating
the pathophysiology of its cardiovascular sequelae.
Stowasser, M. Nat. Rev. Endocrinol. 8, 70–72 (2012); published online 13 December 2011;
The concept that primary aldo­steronism has
a genetic basis is not new. Origi­nally reported
in 1966,1 the rare, glucocorticoid-remediable
familial hyperaldosteronism type 1 (FH‑I) is
due to inherit­ance of a hybrid gene containing the regulatory elements of CYP11B1—a
or because testing of family members
was inconclusive.4
In a major breakthrough towards further
understanding the genetic basis of pri­
mary aldosteronism, Choi and col­leagues
reported somatic mutations (Gly151Arg and
Leu168Arg) in KCNJ5, which encodes an
inwardly-rectifying K+ channel, in eight of 22
patients with APA.5 A third, germline mutation in KCNJ5 (Thr158Ala) had pre­viously
been identified in affected members of a
US family with a new familial form of pri­
mary aldosteronism (FH-III).6 The affected
father and two daughters demon­strated
very severe primary aldo­steronism, with
childhood-onset hyper­tension, hypo­kalemia
and markedly elevated aldosterone and
sup­pressed renin levels. As in patients with
FH‑I, 18-hydroxycortisol and 18-­oxocortisol
levels were elevated, albeit to a greater
degree. How­ever, the primary aldosteronism
was not ­glucocorticoid-suppressible, and
the hybrid gene mutation was not present.
Resected adrenal glands showed marked,
diffuse hyperplasia of the ZF, with the combined weight of both adrenal glands in one
daughter reaching 81 g (normal <12 g).6
None of the three KCNJ5 mutations had
been reported in genetic databases. Choi
et al. discovered that all three amino acid sub­
stitutions were in close proximity to the channel’s selectivity filter and, when expressed in
human embryonic kidney (HEK293T) cells,
were associated with the loss in channel selectivity for K+.5 These findings suggest that
KCNJ5 mutations pre­dispose indivi­duals
to chronic adreno­cortical cell membrane
depolarization via increased channel Na+
permeability, the extra­cellular concentrations of which are much higher than those
of K+. The resulting influx of Ca2+, in turn,
leads to upregulation of enzymes involved in
al­dosterone synthesis and cell proliferation.5
In normal adrenal glands, expression of
wild-type KCNJ5 is restricted to the ZG,5
but the massive hyperplasia seen in patients
with FH-III involved the ZF.6 The fact that
these patients had markedly elevated hybrid
steroid levels also suggests that, as in FH‑I,
the adrenal cells express both aldosterone
synthase and 17α-hydroxylase and, hence,
have molecular characteristics of both the
ZG and the ZF. Neither hybrid steroid levels
nor the cellular composition of the tumors
were reported for patients with APA.
However, on the basis of findings in patients
with FH‑III, those with somatic KCNJ5
mutations are likely to have elevated levels
of hybrid s­ teroids, and their APAs probably
consist of mostly ZF-like cells.
gene that encodes steroid 11β-hydroxylase
and is regulated by adrenocorticotropic
hor­mone (ACTH)—and coding sequences
from CYP11B2—a gene that encodes aldosterone synthase and is primarily regulated
by angiotensin II. Expression of this hybrid
gene leads to ACTH-regulated aldo­sterone
overproduction.2 Glucocorticoids, in small
doses, suppress hybrid gene expression
and ameliorate primary aldosteronism and
hypertension. Because wild-type CYP11B1
is expressed in the zona fasciculata (ZF)
of the adrenal cortex, so is the hybrid gene
(Figure 1). Thus, aberrant aldosterone synthase activity, normally restricted to the
zona glomerulosa (ZG), occurs in the ZF. In
contrast to the ZG, which lacks the steroid
17α-hydroxylase enzyme necessary for the
formation of cortisol precursors, cortisol is
available as a substrate for aldosterone synthase in the ZF. Hence, cortisol is converted
into the ‘hybrid steroids’ 18-hydroxycortisol
and 18-oxocortisol, the levels of which are
elevated in patients with FH‑I.2
The second recognized form of fami­
lial hyperaldosteronism (FH-II) is neither
gluco­corticoid-suppressible nor associ­ated
with the hybrid gene mutation.3 Approxi­
mately 30% of patients with FH‑II have
uni­lateral forms of primary aldosteronism
(mostly due to an aldosterone-producing
ade­n oma [APA]), with the remainder
being bilateral (due to bilateral adrenal
hyper­plasia [BAH]). The genetic basis of
FH‑II remains uncertain and is almost certainly hetero­geneous; several families have
demon­strated linkage of FH‑II with a locus
at chromosome 7p22.3
In 2011, Mulatero and colleagues assessed
prevalence rates of familial hyper­aldo­stero­
nism among 300 patients consecutively
diag­nosed with primary aldosteronism,
after exclud­ing those suspected or diagnosed as having a form of familial primary
aldo­steronism.4 Hybrid gene testing yielded
only two (0.7%) patients with FH‑I. Of
the remaining 298 study participants, 199
had relatives available for and consenting
to bio­chemical screening by aldosterone–
renin ratio (ARR) testing and, where positive, definitive confirmation or exclusion
of primary aldo­steronism by saline infusion testing. This analysis revealed that 12
patients had at least one affected relative,
result­ing in a preva­lence of FH‑II of at least
6% (almost 10-fold that of FH‑I). The prevalence could be higher, as another 54 patients
were classified as having ‘uncertain’ status,
either because not all family members
with hypertension underwent ARR testing
Steroid 17αhydroxylase
Steroid 11βhydroxylase
Figure 1 | Cellular and functional zones of
steroid synthesis in the adrenal cortex.
Normally, production of cortisol is restricted
to ZF and ZR, owing to the enzymatic activity
of steroid 17α-hydroxylase (encoded by
CYP17A1) and 11β-hydroxylase (encoded
by CYP11B1). Similarly, production of
aldosterone is restricted to ZG, as enzymatic
activity of aldosterone synthase (encoded by
CYP11B2) is limited to this zone. In familial
hyperaldosteronism type I, aldosterone
synthase is active in ZF and ZR, as the hybrid
CYP11B1/CYP11B2 gene has regulatory
elements derived from CYP11B1. Hence,
aldosterone synthase has access to cortisol
as a substrate, leading to the production of
18-hydroxycortisol and 18-oxocortisol.
Abbreviations: ZF, zona fasciculata; ZG, zona
glomerulosa; ZR, zona reticularis.
Gordon et al. previously described two
dis­tinct forms of APA.7,8 Angiotensin-II-­
unresponsive APA are composed predominantly of ZF-like cells; patients with this form
of APA have elevated hybrid steroid levels
and lack responsiveness of aldo­sterone to
upright posture or angiotensin-II infusion.
Angiotensin-II-responsive APA contain predominantly non-ZF-like morphology (ZG
or hybrid ZF/ZG), and patients with this
form usually have normal hybrid steroid
levels and are aldosterone-responsive to
upright posture or angiotensin-II infusion.
Taking the above findings into account, APAs
bear­ing KCNJ5 mutations are likely to be of
the angiotensin-II-unresponsive subtype.
Further studies comparing patients with
APAs with or without mutations in terms
of histology, steroid hydroxylase expression,
aldosterone responsiveness to posture and/or
­angiotensin-II, and peripheral hybrid steroid
levels are awaited with great interest.
If KCNJ5 mutation-bearing cells have
both ZG and ZF characteristics, what is
JANUARY 2012 | S23
Key advances
■■ Familial hyperaldosteronism type II (FH-II)
may account for at least 6% of primary
aldosteronism cases and is approximately
5–10 times more common than FH‑I4
■■ A germline mutation in KCNJ5 is
associated with severe, early-onset
familial primary aldosteronism, and
somatic KCNJ5 mutations are common in
aldosterone-producing adenoma5
■■ Patients with primary aldosteronism have
reduced circulating levels of endothelial
progenitor cells, which might contribute
to the development of vasculopathy in
these individuals9
■■ The degree of left ventricular enlargement
in primary aldosteronism is largely
determined by dietary salt; dietary salt
restriction might reduce cardiovascular
risk in this condition10
their origin? Do KCNJ5 mutations bestow
ZF-like characteristics (including his­tol­
ogy, 17α-hydroxylase expression, hybrid
steroid formation, and loss of aldosterone
responsive­ness to angiotensin-II) to ZG
cells, or are ZG-like characteristics (aldo­
sterone synthase expression) bestowed onto
ZF cells? Further studies should shed light on
this fascinating issue.
Other studies in 2011 have focused on
the mechanisms underlying adverse cardio­
vascular effects of aldosterone excess. In a
novel study by Wu et al.,9 levels of endothelial progenitor cells (EPCs), which are
thought to protect against cardiovascular
disease by repairing endothelial injury, were
lower in 113 patients with primary aldo­
steronism (APA n = 87; BAH n = 26) than
in 55 patients with essential hypertension.9
Differences in pulse-wave velocity, a marker
of arterial stiffness, and high-­s ensitivity
C‑reactive protein (hsCRP) levels, a marker
of cardiovascular inflammation, and EPC
counts were attenuated following uni­
lateral adrenalectomy or during treatment
with aldosterone antagonists. Overall, the
findings suggest that increased circulating
aldosterone levels in patients with primary
aldosteronism contribute to vasculopathy
by reducing EPC numbers, partly by activating EPC mineralocorticoid receptors and
possibly indirectly by raising hsCRP levels.9
Animal studies have demonstrated a cri­
tical role for salt in the development of aldosterone-induced cardiovascular damage;
how­ever, corroborative data in humans were
lack­ing. In a case–control study of 21 patients
with primary aldosteronism and 21 matched
individuals with essential hyperten­sion,
Pimenta and co-workers reported in 2011
S24 | JANUARY 2012
that patients with primary aldosteronism
had greater thickness of the left ventricular
wall, end-diastolic dia­meter and mass. More­
over, urinary sodium excretion, a mar­ker
of dietary salt intake, positively correlated
with and was an indepen­dent predictor of
left ventricular wall thickness and mass in
patients with pri­mary aldosteronism, but
not in those with essential hypertension.10
Hence, as in animal studies, salt appears
to interact with auto­nomous aldosterone
excess to bring about cardio­vascular damage
in patients with primary aldosteronism. The
lack of a posi­tive correlation between plasma
aldoster­one and left ventricular dimen­sions
in patients with primary aldosteronism
raises the possibility that, above a certain
thres­hold, aldosterone plays a more permissive part, whereas salt has a more graduated
effect. Either way, these results argue for a
role of dietary salt restriction to reduce the
risk of cardiovascu­lar disease in patients with
primary aldosteronism.
Endocrine Hypertension Research Center,
University of Queensland School of Medicine,
Greenslopes and Princess Alexandra Hospitals,
Ipswich Road, Woolloongabba, Brisbane 4102,
[email protected]
Competing interests
The author declares no competing interests.
Sutherland, D. J., Ruse, J. L. & Laidlaw, J. C.
Hypertension, increased aldosterone secretion
and low plasma renin activity relieved by
dexamethasone. Can. Med. Assoc. J. 95,
1109–1119 (1966).
2. Lifton, R. P. et al. Hereditary hypertension
caused by chimaeric gene duplications and
ectopic expression of aldosterone synthase.
Nat. Genet. 2, 66–74 (1992).
3. Stowasser, M., Pimenta, E. & Gordon, R. D.
Familial or genetic primary aldosteronism and
Gordon syndrome. Endocrinol. Metab. Clin. North
Am. 40, 343–368, viii (2011).
4. Mulatero, P. et al. Prevalence and characteristics
of familial hyperaldosteronism: the PATOGEN
study (Primary Aldosteronism in TOrino-GENetic
forms). Hypertension 58, 797–803 (2011).
5. Choi, M. et al. K+ channel mutations in adrenal
aldosterone-producing adenomas and
hereditary hypertension. Science 331, 768–772
6. Geller, D. S. et al. A novel form of human
mendelian hypertension featuring
nonglucocorticoid-remediable aldosteronism.
J. Clin. Endocrinol. Metab. 93, 3117–3123
7. Gordon, R. D., Hamlet, S. M., Tunny, T. J. &
Klemm, S. A. Aldosterone-producing adenomas
responsive to angiotensin pose problems in
diagnosis. Clin. Exp. Pharmacol. Physiol. 14,
175–179 (1987).
8. Tunny, T. J., Gordon, R. D., Klemm, S. A. &
Cohn, D. Histological and biochemical
distinctiveness of atypical aldosteroneproducing adenomas responsive to upright
posture and angiotensin. Clin. Endocrinol. (Oxf.)
34, 363–369 (1991).
9. Wu, V. C. et al. Endothelial progenitor cells in
primary aldosteronism: a biomarker of severity
for aldosterone vasculopathy and prognosis.
J. Clin. Endocrinol. Metab. 96, 3175–3183
10. Pimenta, E. et al. Cardiac dimensions are largely
determined by dietary salt in patients with
primary aldosteronism: results of a case–
control study. J. Clin. Endocrinol. Metab. 96,
2813–2820 (2011).
Genes, aging and sleep apnea
in polycystic ovary syndrome
Andrea Dunaif
Polycystic ovary syndrome (PCOS) is a complex genetic disease that
affects approximately 7% of women of reproductive age worldwide. From
novel pathways implicated in the etiology of PCOS through genome-wide
association to characterization of the reproductive and metabolic changes
that occur in ageing women with PCOS, the year 2011 has seen a number
of studies published that highlight the intricacies of this condition.
Dunaif, A. Nat. Rev. Endocrinol. 8, 72–74 (2012); published online 20 December 2011;
The year 2011 saw the advent of the first
genome-wide association study (GWAS) in
polycystic ovary syndrome (PCOS).1 GWAS
have been widely used, since the publication of the human haplotype map in 2005,
to localize susceptibility genes for com­plex
traits, such as obesity and type 2 dia­betes
mellitus (T2DM).2 This analysis per­mits
an unbiased examination of the entire
genome for novel disease susceptibility loci
and, unlike candidate gene approaches, is
­hypothesis-generating. 2 The first GWAS
of PCOS was conducted in Han Chinese
women with PCOS, who were diagnosed
using the Rotterdam cri­teria (two of three
of the following findings: oligomenor­rhea,
hyperandrogen­ism and/or poly­cystic ovaries
detected on ultrasono­graphy).1 Asso­cia­tions
achiev­ing genome-wide levels of significance3 between PCOS and loci on chromosomes 2p16.3 (OR 0.71), 2p21 (OR 0.67)
and 9q33.3 (OR 1.34) were observed in a
dis­covery sample of 744 PCOS cases and
895 con­trols and in two indepen­dent replication cohorts of 2,840 PCOS cases and 5,012
­controls, and 498 PCOS and 780 controls.1
Several known genes are located near
the locus at 2p16.3, which showed the
great­est association with PCOS, includ­ing
GTF2A1L and LHCGR. GTF2A1L, which
encodes TFIIAα/β‑like factor (ALF), is a
germ-cell-specific transcription factor that
is highly expressed in adult testis and may
play a part in spermato­genesis. How variation in GTF2A1L might contribute to PCOS
is unclear. LHCGR encodes the lutropin–­
choriogonadotropic hormone recep­t or
(LH/CG‑R) and is a highly plau­­si­ble PCOS
candidate gene. The strongest asso­ci­ation
at the locus on chromo­some 2p21 was with
THADA, a gene origi­nally iden­tified in thy­­
roid adenomas. A GWAS in white indivi­
duals of European ances­t ry reported an
associa­tion of THADA with T2DM.4 The
region on chromosome 9q33.3 associ­ated
with PCOS was located within DENND1A,
which encodes a domain differen­t ially
expressed in normal and neoplastic cells
(DENN) that can bind to and negatively
regulate endo­plasmic reticulum amino­
peptidase 1. Elevations of circulating endoplasmic reti­culum amino­peptidase 1 have
been reported in obese women with PCOS.
Thus, it is possible that varia­tion in THADA
and DENND1A contribute to T2DM and
obesity risk in women with PCOS.
Confirmation that the GWAS signals
reflect a variation in these genes, rather
than in other genes that are in linkage dis­
equilibrium with these loci, requires fur­
ther genetic analyses. In addition, the roles
of these loci in PCOS suscepti­bility in other
racial or eth­nic groups and in other pheno­
types, for instance, in women with PCOS
diag­nosed using the National Insti­tute of
Child Health and Human Develop­ment
(NICHD) criteria (history of oligo­menorrhea
and clinical and/or biochemi­cal evidence of
hyper­androgenism), will require further
investigation. Studies in monozygotic twins
indicate that the herit­ability of PCOS is
close to 80%. However, the loci identified
in the PCOS GWAS confer modest (~30%)
changes in disease risk and do not account
Key advances
■■ Three genetic susceptibility loci
have been mapped by genome-wide
association study in Han Chinese women
with PCOS1
■■ Increased ovarian and adrenal androgen
production and insulin resistance persist
in postmenopausal women with PCOS6,7,8
■■ Obstructive sleep apnea contributes
to insulin resistance and continuous
positive airway pressure improves insulin
sensitivity in women with PCOS10
for the observed herit­ability of PCOS.5 This
so-called ‘missing heritability’ has been seen
in other common complex traits, such as
T2DM, and might reflect the fact that rare
rather than common variants contribute
to complex diseases.5 Nevertheless, GWAS
has been important for implicating novel
b­iologic pathways in disease pathogenesis.
Little was known about the reproductive
and metabolic changes that occur with age
in women with PCOS. In 2011, two crosssectional studies compared postmenopausal
women in the sixth decade of life with
and without PCOS.6,7 In both studies, the
diagnosis of PCOS was made on the basis
of the NICHD diagnostic criteria, although
the study by Puurunen et al. 7 also contained premenopausal women with PCOS
diagnosed by the Rotterdam criteria and a
control group consisting of reproductively
normal, premenopausal women. The study
by Markopoulos et al. 6 found that postmenopausal women with PCOS had significantly elevated levels of circulating total
testosterone, androstenedione, dehydro­
epiandrosterone sulfate (DHEAS) and
17-hydroxyprogesterone and an increased
free androgen index (FAI). Sex hormonebinding globulin (SHBG) levels, how­ever,
were significantly decreased compared
with the control group. No evidence of
increased sensitivity to corticoliberin (also known as
hormone) or adreno­
cortico­t ro­pic hormone
was reported in women
with PCOS, which suggests that increases in
adrenal androgen production
did not result from changes in responsiveness to tropic hor­mones. Dexamethasone
suppression sug­gested that elevated total
testosterone and DHEAS levels were partly
adrenal in origin. Puurunen et al.7 also found
increased androstenedione levels, basally
and in response to chorio­gonado­t ropin,
in postmenopausal women with PCOS
compared with postmenopausal control
women. Furthermore, they found persistent
evidence for insulin resistance and increased
inflammation, independent of BMI, in post­
menopausal women with PCOS compared
to those without the disease.
In a 2011 prospective study, Schmidt et al.8
diagnosed PCOS on the basis of ovarian
histology from ovarian wedge resec­tion or
unilateral oophorectomy per­formed in 1956–
1965. These women were examined in 1987
and compared with control women from a
population-based study. Both groups were
recalled for repeat examination in 2008, in
their eighth decade of life. The PCOS group
had higher FAI and lower follitropin (also
known as follicle-stimulating hormone;
FSH) and SHBG levels than the control
group, as was the case in 1987. DHEAS, total
testo­sterone and androstenedione levels were
significantly increased in premenopausal
women with PCOS in 1987, but were similar
to those of the control group after menopause. The prevalence of hypo­thyroidism
was lower in post­menopausal women with
PCOS than in control individuals.
Taken together, these three studies6,7,8 sug­
gest that hyperandrogenism of both ova­rian
and adrenal origin, as well as insulin resistance, persist in postmenopausal women
with PCOS. Increased androgen levels
declined with age in older post­menopausal
women with PCOS such that only differences in FAI were present in women aged
>70 years.8 By contrast, decreases in SHBG
and FSH levels persist into old age. All of
the studies are limited by small sample sizes,
which might account for the differences
in the prevalence of hypothyroidism. In
JANUARY 2012 | S25
the cross-sectional studies, the accuracy of
diagnosis of PCOS on the basis of medical
records is a potential limitation, although
oligo­menorrhea and symptoms of androgen
excess, such as hirsutism, have been shown
to be excellent proxies for PCOS.9
The prevalence of obstructive sleep apnea
(OSA) is significantly increased in women
with PCOS, independent of obesity—a wellknown risk factor for OSA. Androgen levels
and insulin resistance are positively associated with OSA in PCOS. In 2011, Tasali and
colleagues proposed that OSA contributes
to insulin resistance in PCOS, as it does in
other OSA populations.10 The investigators
directly tested this hypothesis by treating
affected women with continuous positive
airway pressure (CPAP), which resulted in a
modest but significant improvement in
insulin sensitivity.10 A dose–response effect
of CPAP on insulin sensitivity was observed,
with greater improvements in sensitivity
with longer duration of CPAP. A significant decrease in circulating norepinephrine
levels was also reported, without changes in
epinephrine, cortisol or leptin levels. This
observation suggests that the decrease in
insulin sensitivity is mediated by sympathetic nervous system activation. Diastolic
blood pressure, which was not elevated in
women with PCOS who had OSA, decreased
significantly with CPAP.
The limitations of this study were the small
number of women (n = 9) who completed the
CPAP intervention with acceptable compliance (≥4 h of use per night).10 Further­
more, the study population was extremely
obese, with a mean BMI of 48.4 kg/m 2.
The improvements in insulin sensitivity,
although significant, were very modest (on
average 7%); the study par­ticipants remained
profoundly insulin resistant after CPAP
treatment. Modeling of the data suggested
that the beneficial effect of CPAP would
be greater in overweight or mildly obese
patients with PCOS. How­ever, the robustness of the model is questionable given the
small sample size. Never­the­less, this study
suggests that OSA con­tributes to insulin
resistance in PCOS and that CPAP improves
insulin sensitivity in affected women.
In conclusion, these 2011 publications
will have a major effect on the field. It is now
clear from an adequately powered and replicated study that genetic variation contri­
butes to the etiology of PCOS. Furthermore,
reproductive and metabolic features of the
dis­order persist beyond the reproductive
years, increasing the impact of PCOS as
a leading women’s health issue. Finally, a
S26 | JANUARY 2012
component of the insulin resistance associated with PCOS is secondary to OSA and
improves with CPAP treatment.
Division of Endocrinology, Metabolism and
Molecular Medicine, Northwestern University
Feinberg School of Medicine, 303 East Chicago
Avenue, Tarry 15‑745, Chicago, IL 60611, USA.
[email protected]
Competing interests
The author declares no competing interests.
Chen, Z. J. et al. Genome-wide association
study identifies susceptibility loci for polycystic
ovary syndrome on chromosome 2p16.3, 2p21
and 9q33.3. Nat. Genet. 43, 55–59 (2011).
Hirschhorn, J. N. & Daly, M. J. Genome-wide
association studies for common diseases and
complex traits. Nat. Rev. Genet. 6, 95–108
Hoggart, C. J., Clark, T. G., De Iorio, M.,
Whittaker, J. C. & Balding, D. J. Genome-wide
significance for dense SNP and resequencing
data. Genet. Epidemiol. 32, 179–185 (2008).
Zeggini, E. et al. Meta-analysis of genome-wide
association data and large-scale replication
identifies additional susceptibility loci for type 2
diabetes. Nat. Genet. 40, 638–645 (2008).
Manolio, T. A. et al. Finding the missing
heritability of complex diseases. Nature 461,
747–753 (2009).
6. Markopoulos, M. C. et al. Hyperandrogenism in
women with polycystic ovary syndrome
persists after menopause. J. Clin. Endocrinol.
Metab. 96, 623–631 (2011).
7. Puurunen, J. et al. Unfavorable hormonal,
metabolic, and Inflammatory alterations persist
after menopause in women with PCOS. J. Clin.
Endocrinol. Metab. 96, 1827–1834 (2011).
8. Schmidt, J., Brännström, M., LandinWilhelmsen, K. & Dahlgren, E. Reproductive
hormone levels and anthropometry in
postmenopausal women with polycystic ovary
syndrome (PCOS): a 21-year follow-up study of
women diagnosed with PCOS around 50 years
ago and their age-matched controls. J. Clin.
Endocrinol. Metab. 96, 2178–2185 (2011).
9. Taponen, S. et al. Hormonal profile of women
with self-reported symptoms of
oligomenorrhea and/or hirsutism: Northern
Finland Birth Cohort 1966 Study. J. Clin.
Endocrinol. Metab. 88, 141–147 (2003).
10. Tasali, E., Chapotot, F., Leproult, R.,
Whitmore, H. & Ehrmann, D. A. Treatment of
obstructive sleep apnea improves
cardiometabolic function in young obese
women with polycystic ovary syndrome. J. Clin.
Endocrinol. Metab. 96, 365–374 (2011).
Epigenetics, the life-course
and metabolic disease
Peter D. Gluckman
Clinical and experimental studies suggest that early life experiences,
perhaps spanning multiple generations, affect lifelong risk of metabolic
dysfunction through epigenetic mechanisms. Data published in 2011
suggest that epigenetic analysis could potentially have utility as a marker
of early metabolic pathology and might enable early life prophylaxis.
Gluckman, P. D. Nat. Rev. Endocrinol. 8, 74–76 (2012); published online 20 December 2011;
Both population and individual variation
in sus­ceptibility to an obesogenic environ­
ment exist. Genome-wide association studies
have been disappointing in explaining this
variation, and there is a growing focus on
possible epigenetic explanations. Epigenetic
variation is most likely to arise early in the
life-course. An extensive body of experimental, clini­cal and epidemiological evidence
demon­strates that early life develop­mental
factors—­including the maternal diet, but
operating within the normative range of
develop­mental exposures—affect the risk
of developing meta­bolic disease later in life.1
This phe­nomenon is sometimes termed
developmental programming. The early
observations were largely ignored because
of the absence of a plausible biological
mechanism, but explana­t ions developed
over the past decade have been framed in
terms of developmental p
­ lasticity—the
adaptive processes enabling an organism
to respond to environmental cues acting in
early life and adjust its develop­mental trajectory. Developmental plasticity is at least
partly underpinned by epigenetic mecha­
nisms, including DNA methylation and
histone modifications. A number of papers
published during 2011 have re­inforced the
validity of these conclusions.
Extensive previous work has shown that
rats whose mothers are underfed during
preg­nancy and lactation develop an insulin
resis­t ant and obese phenotype in adulthood. Moreover, mice lacking the transcription factor hepatocyte nuclear factor 4α
© Paul Hakimata |
(HNF‑4α), a crucial regulator of gene expression in pancreatic islet β cells, have impaired
glucose tolerance. In 2011, Sandovici et al.
reported how maternal diet in rats affects the
epigenetic regulation of the Hnf4a locus in
pancreatic islets of offspring.2 The researchers found that a low-protein maternal diet
led to underexpression of Hnf4a in the islets
of the offspring. Mechanistically, the under­
expression could be attributed in part to
his­tone modifications at the gene enhancer
region of Hnf4a, which led to changes in
chromatin looping that ultimately weakened
the interaction between the enhancer region
and a specific promoter region. The normative age-dependent epigenetic silencing
of the Hnf4a locus was exacerbated in rats
exposed in utero to a low-protein diet. These
data add to the growing experimental literature showing epigenetic consequences from
the manipulation of early life undernutrition
in metabolically relevant tissues.
Previous studies in rats have shown that
neonatal administration of the adipokine
leptin can reverse developmental programming leading to metabolic compromise as
well as some of the associated epi­genetic
changes. 3 Such studies are provocative
in that they suggest that develop­mental
program­ming in humans might be reversi­
ble. Pinney et al. examined how adminis­
tration of exenatide, an analogue of the
incretin glucagon-like peptide 1, acts to
prevent the development of glucose intolerance in adulthood in rats subjected to intrauterine growth restriction.4 Rats subjected
to this abnormal intrauterine environ­ment
develop glucose intolerance and are charac­
terized by progressive epi­genetic silencing of pancreatic islet transcription factor
Pdx1. The 2011 paper found that injection
of exenatide during postnatal day 1–6 in
rats after intrauterine growth restriction
increased Pdx1 transcription by normalizing
post-translational and epigenetic changes
Key advances
■■ Changes in promoter–enhancer interactions underlie the epigenetic regulation of a
transcription factor in rat pancreatic islets induced by poor maternal diet 2
■■ The trajectory towards onset of type 2 diabetes mellitus (T2DM) in the growth-compromised
newborn rat can be reversed by an incretin analogue that normalizes epigenetic
modifications associated with pancreatic β‑cell dysfunction4
■■ Strong relationships exist between epigenotype at birth and later adiposity, and between
maternal nutrition and offspring epigenotype in humans6
■■ The epigenetic states of genes associated with T2DM and obesity might be a
presymptomatic marker of the lifetime risk of T2DM in humans7
■■ The phenotypic effects of early life overnutrition can be transgenerationally transmitted
through the paternal line in mice9
related to this gene, including Pdx1 activator phosphorylation, histone acetylation and
trimethylation, and DNA methylation at the
CpG island in the Pdx1 promoter. Exenatide
is already in use in the treatment of adults
with type 2 diabetes mellitus (T2DM), and
this study provides an experimental proof
of concept for its usage early in life prior to
disease onset.
Prophylactic treatment of at-risk indivi­
duals early in life is generally more effica­
cious and provides greater health and cost
benefits than treatment after a disease
de­velops. This is particularly the case for
diseases such as diabetes mellitus, where
the cost of treat­ing associated morbidities
and complica­tions compounds over many
years. How­­ever, beyond the obvious research
program that would be needed to validate
such interventions, a challenge remains:
how do we identify infants whose particular developmental programming has placed
them at increased metabolic risk? Birth
weight is not a satisfactory proxy, given that
metabolically adverse developmental programming can occur independent of birth
weight.5 A recent paper by Godfrey et al.
—to which I contributed—has suggested that
the epigenetic state at birth can both predict
childhood adiposity and be a measure of prenatal nutritional exposures.6 Umbilical cord
tissue-derived DNA was sourced from two
independent large-scale prospective cohorts,
and the methylation status at specific CpGs
in the promoter region of RXRα, which
encodes a transcription factor involved in
fat metabolism and insulin sensitivity, was
measured after a preliminary genome-wide
scan. A positive correlation between degree
of RXRα methyla­tion at one specific site in
the promoter and childhood body adipo­
sity at age 6 years or 9 years was found in
both within-cohort and between-cohort
replicates. Additionally, a negative association was observed between mater­nal carbo­
hydrate intake during the first tri­mes­ter of
pregnancy and the degree of methylation at
this site. In contrast to the small effect sizes
typically seen with genetic polymorp­hisms
associated with metabolic disease risk, this
single site epigenotype could account for
at least 25% of the variance in child­hood
adipo­sity. This association was obser­ved
across a healthy population, indicating
that develop­mental programming of metabolic risk and associated epigenetic variation occurs across the normative range of
developmental experiences.
Genes for which polymorphisms have
strong associations with disease risk might
be anticipated to have epimutations (aberrant epigenetic marks). One such gene is
FTO, which has been identified by genomewide association studies as being involved
in obesity risk. In a prospective study of
nonsymptomatic indivi­duals of a mean age
of 30 years by Toperoff et al.,7 those who
showed hypomethylation at an intronic
CpG site of FTO in blood samples had
an increased risk of develop­ing impaired
glucose metabolism by a mean age of
43 years. This pattern was also seen in a
sepa­rate case–control analysis, although the
level of hypomethylation in cases relative to
controls was small, as was the effect size.7
These findings were made in a tissue not
specifically related to metabolic function,
and the differential methylation levels may
have been established in early develop­
ment before major cell differentiation had
occurred, although this inference is contingent on the long-term stability of the FTO
epigenotype after its establishment. The
methylation levels could not be explained
by the presence or absence of the risk allele,
indicating that their association with disease
risk was independent of DNA sequence.
Global prevalence of maternal obesity
and ges­tational diabetes mellitus is rising,
and both are risk factors for the develop­
ment of meta­­bolic disorder in offspring;8
there­fore, the potential for an escalating
prevalence of intergenerationally induced
T2DM is a par­ticular concern. Epigenetic
JANUARY 2012 | S27
mechanisms could underlie the inter­
generational transmis­sion of predisposition
to T2DM, but unraveling the mechanisms
involved is highly complex owing to the
multiple pathways by which nongenomic
inheritance, involving either direct or indirect epi­genetic mechanisms, could occur.
Dunn and Bale utilized a maternal high-fat
diet mouse model and sought to eliminate
maternal physiological or behavioral effects
and grand-maternal effects by studying
pheno­typic outcomes up to the F3 generation through the paternal line.9 They found
that female-specific increased body weight
and length was inherited through the
pater­nal, not the maternal, lineage. Taken
together with other research show­ing that
specific histone marks located at developmental loci in sperm are maintained beyond
spermio­genesis,10 this research suggests that
the phenotype was transmitted through the
male germline via stable epigenetic marks
in sperm.
Our knowledge of developmental path­
ways to metabolic disease is rapidly increas­
ing. It is becoming apparent that epi­genetic
variations induced early in life or that are
present later in life are associated with metabolic risk. How far environ­mentally induced
plasticity, associated with epi­genetic changes
that have metabolic sequelae, operates into
the life-course is, how­e ver, not yet clear.
Research to date has largely been based on a
candidate gene approach, but newer metho­
dologies in sequencing that enable whole
genome approaches, such as ChIP-Seq
(chromatin immuno­precipitation together
with massively parallel DNA sequencing),
should allow more rapid develop­ment of
understandings. Many potential metho­
dological pitfalls exist, and the importance
of validation across multiple cohorts is
emphasized in the work of Godfrey et al.,
who found an apparently impressive associa­
tion between adiposity and methylation at
the endothelial nitric oxide synthase gene
NOS3, which could not be validated in a
second cohort.
Strategies could emerge whereby epi­
genetic state at birth or in infancy could be
used to assess developmental exposures, to
pre­dict metabolic risk and to assist in the
iden­tification of new preventative or therapeutic approaches. Indeed the work with
exena­tide discussed above4 could be con­
sidered an experimental proof of concept;
the chal­lenge would be to find valid risk
markers and acceptable interventions. The
outlook for epigenetic research in human
metabolic and other diseases is promising.
S28 | JANUARY 2012
Liggins Institute, The University of Auckland,
Private Bag 92019, Auckland 1142,
New Zealand.
[email protected]
F. Low provided assistance with reviewing
the literature.
Competing interests
The author declares no competing interests.
Fernandez-Twinn, D. S. & Ozanne, S. E. Early
life nutrition and metabolic programming. Ann.
NY Acad. Sci. 1212, 78–96 (2010).
Sandovici, I. et al. Maternal diet and aging alter
the epigenetic control of a promoter–enhancer
interaction at the Hnf4a gene in rat pancreatic
islets. Proc. Natl Acad. Sci. USA 108,
5449–5454 (2011).
Gluckman, P. D. et al. Metabolic plasticity
during mammalian development is directionally
dependent on early nutritional status. Proc.
Natl Acad. Sci. USA 104, 12796–12800
Pinney, S., Jaeckle Santos, L., Han, Y.,
Stoffers, D. & Simmons, R. Exendin-4
increases histone acetylase activity and
reverses epigenetic modifications that silence
Pdx1 in the intrauterine growth retarded rat.
Diabetologia 54, 2606–2614 (2011).
5. Gale, C. R. et al. Maternal diet during pregnancy
and carotid intima-media thickness in children.
Arterioscler. Thromb. Vasc. Biol. 26, 1877–1882
6. Godfrey, K. M. et al. Epigenetic gene promoter
methylation at birth is associated with child’s
later adiposity. Diabetes 60, 1528–1534
7. Toperoff, G. et al. Genome-wide survey reveals
predisposing diabetes type 2-related DNA
methylation variations in human peripheral
blood. Hum. Mol. Genet.
8. Poston, L., Harthoorn, L. F. & Van Der
Beek, E. M. on behalf of contributors to the ILSI
Europe workshop. Obesity in pregnancy:
implications for the mother and lifelong health
of the child. a consensus statement. Pediatr.
Res. 69, 175–180 (2011).
9. Dunn, G. A. & Bale, T. L. Maternal high-fat diet
effects on third-generation female body size via
the paternal lineage. Endocrinology 152,
2228–2236 (2011).
10. Hammoud, S. S. et al. Distinctive chromatin in
human sperm packages genes for embryo
development. Nature 460, 473–478 (2009).
Osteoporosis therapy—dawn
of the post-bisphosphonate era
Roland Baron
Over the past decade, investigators have actively searched for
safer therapeutic approaches to replace or complement the use
of bisphosphonates and/or parathyroid hormone, exploring both
antiresorptive and osteoanabolic pathways. Besides marked progress
in basic research, the year 2011 has seen several compounds for the
treatment of osteoporosis enter or progress within clinical trials.
Baron, R. Nat. Rev. Endocrinol. 8, 76–78 (2012); published online 6 December 2011;
Osteoporosis is characterized by a marked
decrease in BMD and strength, resulting
in fragility fractures associated with high
morbidity and mortality. The development of bis­phosphonates has substantially
altered the course of the disease in treated
patients. This class of drugs binds rapidly
to the mineral­ized skeleton and impairs the
process of bone resorption, which leads to a
progressive increase in BMD and a decrease
in the risk of fractures, albeit mostly in verte­
bral rather than hip or other non­vertebral
fractures. However, as bone resorption and
bone formation are linked by the process of
bone remodeling, the anti­resorptive effect
of these drugs rapidly decreases bone formation and tissue turnover, limiting the
gain in BMD. This progressive decrease in
bone turnover and the persistent effect of
bis­phosphonates even years after their withdrawal might contribute to two undesirable
long-term adverse effects: osteonecrosis of
the jaw and atypical femoral fractures.
In addition to bisphosphonates, within
the past few years osteoporosis treatment
has seen the arrival of the first osteo­anabolic
drugs: injectable parathyroid hormone
(PTH) fragments. When administered
daily, PTH, unlike bisphosphonates, initially induces clear increases in the levels
of bone formation markers, bone turnover
and BMD. The combination of PTH with
bisphosphonates, especially when the latter
are given at long intervals, has some additive effects over monotherapy with either
compound.1 Yet, the effects of PTH wane
with time, and a concomitant increase in
bone resorption can be observed. Moreover,
some concerns have been raised over the
drug’s oncogenic potential. As a result, academic and industry laboratories have been
engaged in active research over the past
decade to find safer therapeutic approaches,
both antiresorptive and osteoanabolic,
to replace or complement the use of bis­
phosphonates and/or PTH. Besides significant advances in basic science, the year
2011 has seen several compounds enter or
progress within clinical trials.
Mutations in the gene encoding sclero­
stin, an osteocyte-derived Wnt antagonist,
cause sclerosteosis or van Buchem disease,
two high-bone-mass syndromes.2,3 Together
with pre­vious findings, these observations
made a strong case for develop­ing antagonists to this protein. The year 2011 has seen
the first clinical trial of a humanized monoclonal antibody to sclerostin (AMG785) in
healthy men and postmenopausal women.4
In this phase I, randomized, double-blind,
placebo-­controlled study, 72 healthy indivi­
duals received a single subcutaneous or
intravenous injection of ascending doses
of the antibody and were followed up for a
maximum of 85 days. The results showed
not only a dose-dependent increase in the
levels of several bone formation markers
but also a decrease in those of bone resorption markers. Despite the short duration of
the study, significant increases in BMD
of up to 5% in the spine and 3% in the hip
were measured. In this short period, the
compound was safe and well-tolerated. Six
patients, however, developed anti-AMG785
antibodies, which were neutralizing in two
individuals—a reason for some con­cern.
Never­t heless, this trial demonstrates the
valid­ity of this novel therapeutic approach,
which stands out not only because of its
efficacy but also owing to its mechanism
of action, being possibly both anabolic and
antiresorptive. A phase II study has been
completed, and next year will undoubtedly bring more data on this development
program. In the long term, and despite the
fact that haploinsufficient sclerosteosis
or van Buchem mutation carriers experience no significant adverse effects, the
risks of foramen closures due to bone overgrowth and the potential oncogenicity of
this thera­peutic approach will have to be
carefully examined.
The search for antiresorptive compounds
that do not remain in bone for long periods
of time and/or that decrease bone resorption
without negative effects on bone formation
has also been active. One approach has been
to block the differentiation of osteoclasts by
Key advances
■■ A single injection of humanized
antibodies to sclerostin, an endogenous
Wnt antagonist secreted mostly by
osteocytes, markedly raised markers
of bone formation and decreased
markers of bone resorption, increasing
BMD over a short period of time in a
phase I clinical trial4
■■ Orally delivered inhibitors of cathepsin K
increased BMD continuously at all bone
sites over a 3‑year period in extensions
of phase II clinical trials9,10
■■ Antibodies to RANKL reduced the
incidence of vertebral and hip fractures
in women at high risk7
■■ Discontinuation of cathepsin K inhibitors
or antibodies to RANKL induced a rapid
increase in bone turnover and return to
pre-treatment BMD values8,9
antagonizing RANKL, a cytokine that is
required for osteoclast formation, through
the use of antibodies. Detailed analysis of
the effect of denosumab, a human mono­
clonal antibody against RANKL, on bone
turnover markers confirmed that it induces
a very rapid, profound and sustained
decrease in bone resorption markers, but
also in bone formation markers.5 This find­
ing con­firmed that denosumab induces a
marked decrease in bone turnover, which
indeed reaches levels below the pre­
menopausal reference interval. Although
this decrease could be of some concern on
the basis of our previous experi­ence with
bis­phosphonates, no adverse skeletal effects
were noted after 3 years of denosumab
therapy in the FREEDOM trial.6 The study
is being extended for 10 years, which should
adequately address long-term safety issues.
Des­pite these theoretical concerns, another
study in postmenopausal women at high
risk of fractures has demonstrated that
denosumab efficiently reduces the incidence
of new vertebral and hip fractures in highrisk sub­groups.7 Of the 7,808 women enrol­
led in the FREEDOM trial in 213 centers
worldwide, a subgroup with known risk
factors (multiple prevalent vertebral fractures, aged ≥75 years and/or with a femoral
neck BMD T‑score of ≤–2.5) was analyzed
post hoc.7 In all three groups, denosumab
significantly reduced the risk of fracture
to a degree consistent with that in low-risk
patients. Thus, denosumab has a consistent
antifracture efficacy in patients with varying
degrees of fracture risk. Interestingly, the
higher absolute risk observed in the highrisk group was associated with the greatest
denosumab-induced risk reduction.
Another novel antiresorptive approach has
been to antagonize not the differen­tiation
but the function of osteoclasts through
inhi­bition of cathepsin K, an enzyme utilized by osteoclasts to degrade the dense
col­lagen that constitutes most of the bone
matrix. Following the report of a 2‑year
phase II trial,8 Eisman et al.9 reported in
2011 the results of a 1‑year extension study.
Con­tinued treatment of postmenopausal
women with 50 mg odanacatib (orally once
a week for 3 years) produced significant
increases from baseline and from year 2 in
BMD at both the spine and the hip. Bone
resorption markers remained suppressed,
whereas the bone formation marker BALP
was unchanged from baseline, a feature
that distinguishes this treatment modality
from other antiresorptive therapies, including bisphosphonates and denosumab (see
above). Confirming this unique profile of
cathepsin K inhibition, Eastell et al.10 showed
that another inhibitor (ONO‑5334), given
to postmenopausal women orally once daily
for 12 months, also resulted in significant
increases in BMD at the lumbar spine, total
hip and femoral neck. These studies sug­gest
that inhibition of cathepsin K, which does
not decrease osteoclast differentiation, can
reduce osteoclast acti­vity while maintaining
the ‘coupling’ to bone formation and osteoblast activity, thereby efficiently increasing
BMD. Interestingly, this approach has a
less negative effect on bone turnover than
bisphosphonates or denosumab.
The year 2011 also saw the first reports
on a novel and potentially troublesome phenomenon: the resolution of drug effects.8,9
Whereas most therapeutic fields would not
expect anything other than a resolution of
effect upon discontinuation of treatment,
the osteoporosis field was somewhat caught
by surprise. This surprise can be explained
by the fact that, for the past 20 years, we
have been used to the bisphosphonates,
a class of drugs which remains bound to
the skeleton and continues to have therapeutic acti­vity for several years after cessation of treat­ment. This characteristic has
both advantages (nothing happens when
you stop treat­ment) and drawbacks (possible long-term adverse effects). A more
troublesome finding is that the studies on
discontinuation of treatment of both denosumab8 and odanacatib9 have shown not
only a resolution of effect—that is, rapid
reversibility of treatment, which would be
an advantage over ­bisphosphonates—but
also a ‘rebound’ effect. High bone turnover
activity has been shown to rapidly (within
JANUARY 2012 | S29
6–12 months) return the patient’s BMD (at
least spine and hip) to values found before
initiation of treatment. Biologically, these
results illustrate the long-suspected existence of a ‘skeletal mechanostat’ whereby
BMD is determined by the balance of a
large number of factors intrinsic to each
individual. Unsurprisingly, if these factors
remain unchanged, the ‘excess’ bone generated by antiresorptive treatments is rapidly
removed by excessive bone resorption once
we allow osteoclasts to do their job. Yet, this
phenomenon will create new challenges for
osteoporosis treatment, probably calling
for the combined or sequential use of several
thera­peutic principles. Already, the use of
bis­phosphonates to stabilize bone gains after
PTH treatment has been suggested. Whether
a similar resolution of effect will occur
with other anabolic therapies, in particular
sclero­stin anti­bodies, will be an important
­question to answer in the coming years.
In summary, 2011 has seen significant
pro­gress in the understanding and implementation of several new therapeutic options
for the treatment of osteoporosis, in what
could be called the post-­bisphosphonate
era. Although very promising novel anti­
resorptive and anabolic agents have entered
or been further assessed in clinical trials,
showing extraordinary efficacies and pro­
mises for the future, they will have to be
tested in the long term to better establish
their safety relative to bisphosphonates.
These studies have also uncovered novel
chal­lenges, in particular the resolution of
drug effects, which will need to be addressed
in order for novel therapeutic strategies to be
implemented for osteoporosis.
Department of Medicine, Harvard Medical
School, Massachusetts General Hospital,
Endocrine Unit, 188 Longwood Avenue, Boston,
MA 02115, USA.
[email protected]
Competing interests
The author declares associations with the following
companies: Amgen, Arcarios, Bone Therapeutics,
Merck, Novartis, Roche. See the article online for full
details of the relationships.
Cosman, F. et al. Effects of intravenous
zoledronic acid plus subcutaneous teriparatide
[rhPTH(1–34)] in postmenopausal
osteoporosis. J. Bone Miner. Res. 26, 503–511
Balemans, W. et al. Increased bone density in
sclerosteosis is due to the deficiency of a novel
secreted protein (SOST). Hum. Mol. Gen. 10,
537–543 (2001).
Balemans, W. et al. Identification of a 52kb
deletion downstream of the SOST gene in
patients with van Buchem disease. J. Med.
Genet. 39, 91–97 (2002).
S30 | JANUARY 2012
Padhi, D. et al. Single-dose, placebo-controlled,
randomized study of AMG 785, a sclerostin
monoclonal antibody. J. Bone Miner. Res. 26,
19–26 (2011).
Eastell, R. et al. Effects of denosumab on bone
turnover markers in postmenopausal
osteoporosis. J. Bone Miner. Res. 26, 530–537
Cummings, S. R. et al. Denosumab for
prevention of fractures in postmenopausal
women with osteoporosis. N. Engl. J. Med. 361,
756–765 (2009).
Boonen, S. et al. Treatment with denosumab
reduces the incidence of new vertebral and hip
fractures in postmenopausal women at high
risk. J. Clin. Endocrinol. Metab. 96, 1727–1736
Bone, H. G. et al. Effects of denosumab
treatment and discontinuation on bone mineral
density and bone turnover markers in
postmenopausal women with low bone mass.
J. Clin. Endocrinol. Metab. 96, 972–980
9. Eisman, J. A. et al. Odanacatib in the
treatment of postmenopausal women with
low bone mineral density: three-year
continued therapy and resolution of effect.
J. Bone Miner. Res. 26, 242–251
10. Eastell, R. et al. Safety and efficacy of the
cathepsin K inhibitor ONO-5334 in
postmenopausal osteoporosis: the OCEAN
study. J. Bone Miner. Res. 26, 1303–1312
Heterogeneity of T1DM raises
questions for therapy
Paolo Pozzilli
Research in 2011 regarding β‑cell destruction, early
immunointervention trials and development of late complications
in type 1 diabetes mellitus have highlighted the heterogeneity
of this disease. Patient phenotyping should be performed for the
implementation of tailored therapies, especially taking into account
the age at which the disease is diagnosed.
Pozzilli, P. Nat. Rev. Endocrinol. 8, 78–80 (2012); published online 20 December 2011;
Type 1 diabetes mellitus (T1DM) is an autoimmune disorder involving the destruction
of pancreatic β cells. Why T1DM develops is
still debated, but a role of oxidative stress in
genetically susceptible individuals has been
proposed in the past few years. Oxidative
stress in these individuals might be caused
by a lack of antioxidant enzymes and/or by
environ­mental triggers such as viral infections, and has been linked to β‑cell cyto­
toxicity; 1,2 however, the role of reactive
oxygen species (ROS) in the pathogenesis
of T1DM remains controversial.
In 2011, Thayer et al. 3 reported the
required cellular sources of ROS for T1DM
induction in mice models of T1DM. The
researchers showed that nonobese, diabetic
(NOD) mice were protected from T1DM as a
consequence of macrophage and neutrophil
depletion. Interestingly, depletion of neutrophils alone did not protect or delay the onset
of the disease. In other experiments, using
mice with spleno­c ytes genetically altered
to prevent the production of ROS, development of T1DM was significantly delayed.
These findings suggest that ROS produced
by macrophages are critical factors in inducing damage to the β cells and that targeting
this mechanism could be a novel approach
worth considering for T1DM prevention and
early treatment.
The results of two trials published in
2011 revealed the potential of immunotherapy to protect β‑cell function in patients
with recent onset of T1DM. 4,5 Abatacept
is a drug already in use in the treatment of
rheumatoid arthritis. This fusion protein
of cytotoxic T‑lymphocyte protein 4 and
Key advances
■■ Signaling of reactive oxygen species
appears to be a relevant mechanism for
the initiation of β‑cell autoimmunity by
T cells in mice3
■■ Immunointervention trials in patients with
recent onset of type 1 diabetes mellitus
did not meet everyone’s expectations4,5
■■ Poor glycemic control is associated with
an increased risk of heart failure in adult
patients with T1DM8
■■ Protective factors for late complications
exist in some patients with type 1
diabetes mellitus,9 and their
identification would help prevention
of retinopathy, neuropathy and
cardiovascular disease
immunoglobulin modulates costimulatory
receptors to prevent full T‑cell activation. In a
multi­center, double-blind, random­ized controlled trial of abatacept, 112 patients with
recent onset of T1DM received abatacept or
placebo infusions intravenously on days 1,
14, 28 and monthly for a total of 27 infusions
over 2 years.4 The pri­mary outcome was the
C‑peptide response following a mixed meal
evaluated at 2 years. Abata­cept, as compared
to placebo, increased the adjusted C‑peptide
area under the curve (AUC) by 59%, with a
difference between the two groups persisting
through­out the trial and with a delay in the
reduc­tion of C‑peptide of 9.6 months using
abata­cept. The investigators concluded that
modulation of T cells with abatacept slowed
reduction in β‑cell function over 2 years;
however, after the 6-months time point, the
decrease in β‑cell function in the abata­cept
group was parallel to that in the placebo
group, which suggests that the effect of
a­batacept declined over time.
The second trial tested the capacity of a
humanized, anti-CD3 monoclonal anti­body
(teplizumab) to preserve β‑cell func­t ion
and decrease insulin needs in patients with
T1DM of recent onset aged 8–35 years.5 In
a phase III trial, 516 patients diag­nosed with
T1DM for <3 months were randomly allocated to receive various doses of teplizumab
or placebo at baseline and at 26 weeks. The
primary composite outcome was the percentage of patients with insulin dose <0.5 U/kg
per day and HbA1c <6.5% at 1 year. The pri­
mary outcome did not differ between groups;
however, 5% of patients in the teplizumab
groups were not taking insulin at 1 year compared with no patients in the placebo group.
Why did immunointervention not achieve
what everyone expected? The answer lies in
patient heterogeneity, the chosen sample
size and the dose of drugs used. T1DM is
a hetero­geneous disease in terms of age at
onset, HLA genotype, residual β‑cell function at the time of diagnosis, insulin resistance, compliance to insulin treatment
and HbA1c level. These two trials included
patients diagnosed with the disease as chil­
dren and those diagnosed as adults. Com­
pared with children diagnosed with T1DM,
those diagnosed as adults show higher
C‑peptide secretion 6 and higher insulin
resistance. If the primary end point of an
immuno­intervention trial is the C‑peptide
response following a mixed meal (as is the
case in nearly all recent trials), the above
variables should have been taken into
account when calculating the tar­get sample
size. The problem is that the decline of
C‑peptide after diagnosis varies among different age groups, which means that large
sample sizes are required if strict age groups
are not used. If this issue is not sorted out, we
will inevitably continue to fail. Experience
shows that subgroups of patients exist who
respond to immuno­intervention and indeed
their β cells still function, especi­ally in those
with low risk HLA genotypes.7 By means of
post-hoc analysis and/or meta-analysis of
data we should be able to identify what fac­
tors determine a beneficial effect of a specified treatment so that future trials can be
designed accordingly.
Poor glycemic control is associated with
microvascular and macrovascular complications in T1DM, but is it associated with
heart failure? In 2011, Lind et al. revealed the
association in 20,985 patients with T1DM
identified from the Swedish national dia­
betes registry.8 Patients aged ≥18 years with
T1DM and no known heart failure were followed up until hospital admission for heart
failure, death or end of follow-up. Patients
had a mean age of 38.6 years at baseline.
During a median follow-up of 9 years, 635
(3%) of the patients were admitted to hospital with a diagnosis of heart failure, which
corresponds to an incidence of 3.38 events
per 1000 patient-years. In patients with an
HbA1c ≥10.5%, as compared with a reference
group with an HbA1c <6.5%, the hazard ratio
for developing heart failure was 3.98 (95%
CI 2.23–7.14). Increased age and duration
of the disease raised the risk of heart failure.
Therefore, it looks like that by achieving good metabolic control, prevention of
heart failure can be obtained in patients
with T1DM.
Evidence for a decline in the inci­dence
of microvascular complications in T1DM
dur­ing the past few decades has been
reported; therefore, the question arises
whether protective factors for late dia­betic
com­plications in T1DM can be identi­fied.
To assess complication prevalence and identify protective factors in patients with T1DM
of ≥50 years duration, a complication-free
sub­group was analyzed. A cross-sectional,
observational study was carried out in 351
US residents who survived with T1DM for
≥50 years (known as the Medalists).9
A substantial number of Medalists did not
develop proliferative retinopathy (42.6%),
nephropathy (86.9%), neuropathy (39.4%)
or cardiovascular disease (51.5%). Meta­
bolic control (current and over the pre­vious
15 years) was unrelated to complica­tions,
whereas a 7.2-fold increased risk was obser­
ved in patients with raised plasma levels of
© Dml5050 |
the advanced glycation end pro­ducts (AGEs)
carboxyethyl-lysine and pentosidine.
The population of Medalists is of interest,
as protective factors are probably present in
these patients. Once identified, such factors
could be used to protect other patients with
the disease from the development of longterm chronic complications. A link between
formation or processing of AGE and the
development of diabetic vasculo­pathy is a
well-known finding. One possible mechanism of damage occurs via CML-collagen,
mediated through the soluble receptor for
AGE (RAGE).10 The measurement of circulating RAGE in the Medalists would have
aided understanding of the protection from
complications achieved by these patients.
What have we learnt about T1DM in
2011? If one looks at disease patho­genesis,
protection of β cells from apoptosis or at the
development of or protection from late complications, the disease appears more heterogeneous than ever. Therapeutic approaches,
in whichever field, should be established
with this concept in mind. Too often,
pharma­ceutical companies aim to tackle a
clinically relevant issue by treating the entire
population of patients with a disease with
one target drug; unfortunately, this approach
does not work for T1DM. Specific patient
subgroups must be integral in the design
of studies either for etiopathogenesis or the
prevention and treatment of β‑cell damage.
The same concept applies for late diabetic
complications, as different process pathways might be involved. Therapies should
be designed accordingly.
JANUARY 2012 | S31
Department of Endocrinology and Diabetes,
University Campus Bio-Medico, Via Álvaro del
Portillo, 21, 00128 Rome, Italy.
[email protected]
Competing interests
The author declares associations with the
following companies: Andromeda Biotech,
Bristol-Meyers Squibb, GlaxoSmithKline, Novartis,
Sanofi-Aventis. See the article online for full details
of the relationships.
Bottino, R. et al. Response of human islets to
isolation stress and the effect of antioxidant
treatment. Diabetes 53, 2559–2568 (2004).
Tran, P. O et al. Adenoviral overexpression of
the glutamylcysteine ligase catalytic subunit
protects pancreatic islets against oxidative
stress. J. Biol. Chem. 279, 53988–53993
Thayer, T. C, et al. Superoxide production by
macrophages and T cells is critical for the
induction of autoreactivity and type 1 diabetes.
Diabetes 60, 2144–2151 (2011).
Orban, T. et al. Co-stimulation modulation
with abatacept in patients with recent-onset
type 1 diabetes: a randomised, double-blind,
placebo-controlled trial. Lancet 378, 412–419
Sherry, N. et al. Teplizumab for treatment of
type 1 diabetes (Protégé study): 1-year results
from a randomised, placebo-controlled trial.
Lancet 378, 487–497 (2011).
Klinke, D. J. Age-corrected beta cell mass
following onset of type 1 diabetes mellitus
correlates with plasma C-peptide in humans.
PLoS ONE 6, e26873 (2011).
Buzzetti, R. et al.C-peptide response and HLA
genotypes in subjects with recent-onset type 1
diabetes after immunotherapy with
DiaPep277: an exploratory study. Diabetes 60,
3067–3072 (2011).
8. Lind, M. et al. Glycaemic control and incidence
of heart failure in 20,985 patients with type 1
diabetes: an observational study. Lancet 378,
140–146 (2011).
9. Sun, J. K. et al. Protection from retinopathy and
other complications in patients with type 1
diabetes of extreme duration: the Joslin
50-Year Medalist Study. Diabetes Care 34,
968–974 (2011).
10. Thomas, M. C, et al. Soluble receptor for AGE
(RAGE) is a novel independent predictor of allcause and cardiovascular mortality in type 1
diabetes. Diabetologia 54, 2669–2677
Don’t miss out!
Sign up to free e-ALERTS and be notified when new articles are published
in Nature Reviews Endocrinology. Research Highlights, News & Views, Reviews,
and Perspectives—you won’t miss a thing!
S32 | JANUARY 2012
A new standard of care and the race towards
IFN-free therapy
Wolf Peter Hofmann and Stefan Zeuzem
Chronic HCV infection is a leading cause of liver-related morbidity and mortality. In 2011, treatment options for
patients infected with HCV genotype 1 changed dramatically with the approval of two nonstructural protein 3
protease inhibitors—boceprevir and telaprevir—by the FDA and the European Medicines Agency.
Hofmann, W. P. & Zeuzem, S. Nat. Rev. Gastroenterol. Hepatol. 9, 67–68 (2012); published online 20 December 2011; doi:10.1038/nrgastro.2011.249
In phase III clinical trials, the combination of boceprevir or telaprevir with PEGIFN-α–ribavirin has been shown to result
in increased sustained virological response
(SVR) rates compared with PEG-IFN-α–
ribavirin (67–75% and 38–44%, respectively) in therapy-naive patients. 1,2 As
these new triple therapies also resulted in
increased rapid virological response (RVR;
undetectable levels of HCV RNA at week 4
of triple therapy) rates, response-guided
therapy to shorten the duration of treatment from 48 weeks to 24–28 weeks is now
possible for a large proportion of patients.
Additional phase III clinical trials showed
that many patients who did not respond
100 –
90 –
well to previous PEG-IFN-α–ribavirin
therapies also benefited from re-treatment
with PEG-IFN-α–ribavirin and boceprevir
or telaprevir.3,4 Patients who had previously
relapsed achieved SVR rates of 69–88%
when re-treated, and previous partial
responders showed SVR rates of 40–59%.
In those who had a previous null response,
SVR rates following re-treatment were still
poor (33% for telaprevir).4
Despite these encouraging achievements
for patients infected with HCV genotype 1,
the low tolerability, particularly for PEGIFN-α, and the emergence of resistant
variants associated with treatment failure
of triple therapies that include boce­previr
or telaprevir are still major drawbacks.
Furthermore, current dosing schedules are
complex and the pill burden is high, which
might result in suboptimal adherence to
treatment. Combination therapies with two
direct-acting antiviral agents (DAAs) that
have different modes of action, and an alloral IFN-free DAA therapy should overcome
resistance, reduce the incidence of adverse
events and improve treatment adherence.
Different substance classes with anti-HCV
activity include nonstructural protein (NS)
3 protease inhibitors, NS5A inhibitors, and
nucleoside inhibitors and non-nucleoside
inhibitors of the HCV NS5B polymerase.
Results of several trials of DAA combination
RVR and SVR (%)
80 –
70 –
60 –
50 –
40 –
30 –
20 –
10 –
BMS-790052 +
BMS-790052 +
BMS-650032 +
Study 1
BMS-790052 +
Study 2
PSI-7977 +
PSI-7977 +
RBV12 +
PEG-IFN-α (4)
PSI-7977 +
RBV12 +
PEG-IFN-α (8)
PSI-7977 +
RBV12 +
PEG-IFN-α (12)
Study 3
Figure 1 | The effectiveness of an all-oral IFN-free therapy for patients with HCV has been demonstrated in some study arms of three recent trials.
Patients in study 1 had HCV genotype 1a or 1b infection and previously a null response to PEG-IFN-α–RBV.8 These patients received 200 mg of the
protease inhibitor BMS‑650032 twice a day plus 60 mg of the NS5A inhibitor BMS‑790052 once a day with or without PEG-IFN-α–RBV for 24 weeks.
Patients with viral breakthrough during dual therapy received PEG-IFN-α–RBV and achieved SVR thereafter (not shown). Patients in study 2 all had HCV
genotype 1b and had previously had a null response to PEG-IFN-α–RBV.9 These patients received 200 mg of the protease inhibitor BMS‑650032 twice
a day and 60 mg of the NS5A inhibitor BMS‑790052 once a day for 24 weeks. Patients in study 3 had HCV genotypes 2 or 3 and were treatment
naive.10 These patients all received 400 mg of the nucleoside inhibitor PSI‑7977 once a day as monotherapy or with RBV for 12 weeks. Three arms
also received PEG-IFN-α for 4, 8 or 12 weeks. Abbreviations: RBV, ribavirin; RVR, rapid virological response; SVR, sustained virological response.
Key advances
■■ The newly licensed protease inhibitors
boceprevir and telaprevir substantially
increase sustained virological response
rates in treatment-naive1,2 patients
infected with HCV genotype 1 and those
who failed previous therapy
■■ Several IFN-free regimens that combine
protease inhibitors and polymerase
inhibitors with or without ribavirin achieve
undetectable HCV RNA levels early during
therapies that had at least one IFN-free treatment arm were published in 2011 and open
the race for an all-oral combination therapy
for patients with chronic hepatitis C.
In the SOUND‑C1 study, 32 patients who
were infected with HCV genotype 1 were
randomly assigned to receive the protease
inhibitor BI 201335 in combination with one
of two doses of the non-­nucleoside inhibitor
BI 207127 and ribavirin for 4 weeks, followed
by PEG-IFN-α–­ribavirin and BI 201335 until
week 24.5 Only patients without optimal
initial virologic response received PEGIFN-α–ribavirin until week 48. Patients in
the high-dose treatment group achieved
better RVR and SVR rates than those in the
low-dose treatment group (100% and 94%
versus 73% and 73%). The all-oral triple
combination administered for 4 weeks was
generally well tolerated and did not cause
serious adverse events.
In a phase II trial, 46 treatment-naive
patients infected with HCV genotype 1 were
assigned to receive one of three regimes: the
protease inhibitor GS‑9256 in combination
with the non-nucleoside inhibitor tego­
buvir (dual combination); GS‑9256, tegobuvir and ribavirin (triple combination); or
GS‑9256, tegobuvir, PEG-IFN-α and riba­
virin (quadruple combination) for 4 weeks.6
All patients received PEG-IFN-α–ribavirin
thereafter, or immediately if they did not
respond or had virological breakthrough.
In the dual, triple and quadruple combination therapy arms, RVR rates were 7%, 38%
and 100%, respectively. The low RVR rate
in the dual DAA combination was mainly
attributable to the emergence of resistant
variants, particularly in patients infected
with HCV genotype 1a. The addition of
ribavirin to the DAA combina­tion further
reduced viral levels and decreased viral
breakthrough. None of the patients receiving the quad­ruple therapy experienced viral
plateau or breakthrough during the 4 weeks
of treatment. The combination of GS‑9256
and tegobuvir was generally well tolerated
S34 | JANUARY 2012
and a transient increase in levels of bilirubin
was observed in all treatment arms.
In the ZENITH study, 106 patients were
randomly assigned to receive telaprevir in
combination with two different doses of
the non-nucleoside inhibitor VX‑222 either
as a dual DAA therapy or as a quadruple
therapy in combination with PEG-IFN-α–
ribavirin for 12 weeks.7 Patients received
PEG-IFN-α–ribavirin for an additional 12
or 24 weeks, depending on their virological
results at weeks 2 and 8. The RVR rates in
the low-dose and high-dose VX‑222 dual
DAA combination arms were 17% and
59%, respectively, and a significant proportion of patients developed virological
breakthrough. In the quadruple combination arms, RVR rates for patients receiving
low-dose and high-dose VX‑222 were 86%
and 87%, respectively, and no virologic
breakthrough was observed. The DAA
combina­tion was generally well tolerated.
In two independent trials from the USA
and Japan, small cohorts of patients with
HCV genotype 1 infection who were previous null responders were treated with the
protease inhibitor BMS‑650023 in combination with the NS5A inhibitor BMS‑790052
with or without PEG-IFN-α–ribavirin
(Figure 1).8,9 In the US study, patients with
HCV genotype 1a and 1b infection who
received quadruple therapy with PEGIFN-α–ribavirin had a better response than
those who received dual DAA combination therapy.8 Treatment failure with the
dual therapy was mainly associated with
HCV genotype 1a. In the Japanese study,
10 patients, all infected with HCV genotype 1b, who had previously not responded
to therapy completed IFN-free DAA
combina­tion therapy with BMS‑650023 and
BMS‑790052 for 24 weeks.9 HCV RNA was
undetectable from treatment week 8, and all
patients achieved SVR.
In addition, a separate study found that
100% of patients infected with HCV geno­
types 2 or 3 who were treated with the
nucleoside inhibitor PSI‑7977 plus ribavirin
for 12 weeks achieved an SVR (Figure 1).10
The combination with ribavirin (but not
PEG-IFN-α) remained necessary, as relapses
occurred with PSI‑7977 monotherapy
in four of 10 patients.
A new standard of care is now available
for patients infected with HCV genotype 1.
Furthermore, exciting results from several
clinical trials demonstrate that a high proportion of patients can achieve undetectable levels of HCV RNA during therapy
with an all-oral IFN-free combination of
DAAs. Treatment failure of IFN-free DAA
combina­tions is associated with host factors,
virus factors and the barrier to resistance of
different compounds and substance classes.
In this context, the future role of host-­
targeting antiviral agents such as alis­porivir
(cyclophilin inhibitor) and miravirsen
(miR‑122 inhibitor) must be explored. The
development of a pan-genotypic and an
all-oral regimen for patients with chronic
hepatitis C is rapid and promising.
GmbH, Rubensstraβe 119, 12157 Berlin,
Germany (W. P. Hofmann). Medizinische
Klinik 1, Klinikum der Johann Wolfgang GoetheUniversität, Theodor Stern-Kai 7, Frankfurt am
Main 60590, Germany (S. Zeuzem).
Correspondence to: S. Zeuzem
[email protected]
Competing interests
W. P. Hofmann declares associations with the
following companies: Bristol-Meyers Squibb, Gilead,
Roche, MSD and Janssen Cilag. S. Zeuzem declares
associations with the following companies: Abbott,
Achillion, Boehringer Ingelheim, Bristol-Meyers
Squib, Gilead, iTherX, Merck, Pharmasset, Roche,
Santaris, Tibotec and Vertex. See the article online
for full details of the relationships.
Poordad, F. et al. Boceprevir for untreated
chronic HCV genotype 1 infection. N. Engl. J.
Med. 364, 1195–1206 (2011).
2. Jacobson, I. M. et al. Telaprevir for previously
untreated chronic hepatitis C virus infection.
N. Engl. J. Med. 364, 2405–2416 (2011).
3. Bacon, B. et al. Boceprevir for previously treated
chronic HCV genotype 1 infection. N. Engl. J.
Med. 364, 1207–1217 (2011).
4. Zeuzem, S. et al. Telaprevir for retreatment of
HCV infection. N. Engl. J. Med. 364, 2417–2428
5. Zeuzem, S. et al. Efficacy of the protease
inhibitor BI 201335, polymerase inhibitor
BI 207127, and ribavirin in patients with chronic
HCV infection. Gastroenterology 141,
2047–2055 (2011).
6. Zeuzem, S. et al. The protease inhibitor GS‑9256
and non-nucleoside polymerase inhibitor
tegobuvir alone, with RBV or peginterferon plus
RBV in hepatitis C. Hepatology http://
7. Di Bisceglie, A. et al. VX‑222 with TVR alone or in
combination with peginterferon alfa2a and
ribavirin in treatment-naive patients with chronic
hepatitis C: ZENITH study interim results
[abstract]. J. Hepatol. 54, S540 (2011).
8. Lok, A. S. et al. Quadruple therapy with
BMS‑790052, BMS‑650032 and PEG-IFN/RBV
for 24 weeks results in 100% SVR12 in HCV
genotype 1 null responders [abstract].
J. Hepatol. 54, S536 (2011).
9. Chayama, K. et al. Dual oral combination therapy
with the NS5A inhibitor BMS‑790052 and the
NS3 protease inhibitor BMS‑650032 achieved
90% sustained virologic response (SVR12) in
HCV genotype 1b-infected null responder [LB-4].
Hepatology 54 (Suppl. 1), 100A (2011).
10. Gane, E. et al. Once daily PSI‑7977 plus RBV:
pegylated interferon-ALFA not required for
complete rapid viral response in treatment-naive
patients with HCV GT2 or GT3 [abstract 34].
Hepatology 54 (Suppl. 1), 377A (2011).
Key advances
Genomics in hepatocellular
carcinoma—a big step forward
■■ Next-generation high-throughput
sequencing has enabled the
comprehensive description of genomic
alterations throughout the genome at
extremely high resolution; the first report
using this technology in hepatocellular
carcinoma (HCC) from an individual with
chronic hepatitis C revealed previously
uncharacterized mutation patterns,
intra-chromosomal rearrangements and
genetic heterogeneity within the tumor4
■■ The first application of genome-wide
association studies to HCV-related HCC
has identified new susceptibility loci6,7
Ryosuke Tateishi and Masao Omata
Despite enthusiastic efforts using the latest advanced molecular
technologies, no specific universal genetic alteration has been found
in hepatocellular carcinoma (HCC). The application of whole-genome
sequencing using next-generation sequencing technologies is starting
to clarify the intraindividual and intratumoral diversity in genomic
alterations in HCC. A new sequencing era in HCC has begun.
Tateishi, R. & Omata, M. Nat. Rev. Gastroenterol. Hepatol. 9, 69–70 (2012); published online 10 January 2012;
Primary liver cancer is the fifth most frequently diagnosed cancer worldwide in men
and the seventh in women.1 Hepatocellular
carcinoma (HCC) is the predominant histological type, accounting for 70–85% of total
liver cancers.1 Globally, chronic HBV infection is the most prevalent cause of HCC, followed by chronic HCV infection. Although
HCC is a typical example of a virus-related
cancer, it is also strongly associated with
certain lifestyle factors. Chronic alcoholism is a classic risk factor; obesity is also
recognized to strongly affect HCC development compared with other malignancies.2
In either case (virus-related or lifestyle
factors), it is assumed that accumulated
DNA damage, caused by long-standing
necroinflammation and regeneration, has a
major role in the process of hepatocarcinogenesis. Despite enthusiastic efforts using
the latest advanced molecular technologies, no specific universal genetic alteration has been found in HCC. However, the
success of sorafenib, a multikinase inhibitor
of Raf‑1 and B‑Raf, has provided proof that
molecularly targeted agents have a role in
the treatment of HCC.3 An understanding
of the comprehensive picture of mol­ecular
alterations in HCC is now profoundly
needed (Figure 1).
The International Cancer Genome
Consortium (ICGC) is an international
project aiming to obtain a comprehensive
description of genomic, transcriptomic
and epigenomic changes in 50 different
tumor types and/or subtypes that are of
clinical and societal importance across the
globe. Japanese and French researchers
are investigating virus-related and virusunrelated HCC, respectively. A study by
Totoki et al.4 is the first report in HCC as
part of this project. The authors sequenced
HCV-related HCC and lymphocytes from
KEY ADVANCES IN MEDICINE alleles harbored this substitution, which
demonstrates the variability of genetic
alterations among cancer cells.
Li et al.,5 in the USA, also used wholeexome sequencing to determine the
sequences of approximately 18,000 proteincoding genes in the cancers and normal
tissues of ten individuals with HCV-related
HCC. They identified 689 potential somatic
mutations, including 429 non­synonymous
somatic mutations in 411 genes. Five genes
that were found to be mutated in more than
one tumor (CTNNB1 in four tumors, TP53
in three tumors and ARID2, DMXL1 and
NLP1 in two tumors each) were selected for
further analysis. Li et al.5 investigated the
coding exons of these five genes in 23 additional HCV-related HCCs, which revealed
that CTNNB1, TP53, ARID2, DMXL1 and
NLRP1 were mutated in eight (24.2%), four
(12.1%), six (18.2%), two (6.1%) and two
(6.1%) of the total 33 HCCs, respectively.
Among these genes, the authors identified
ARID2 as a novel tumor suppressor gene in
a Japanese male using massively parallel
sequencing. By comparing tumor and nontumor sequences, they identified 11,731
somatic mutations in the tumor. The prevalence of somatic substitutions was substantially less in the genic (intronic, noncoding
exon and coding exon) regions relative to
the intergenic regions, which suggests that
negative selection of lethal mutations or
repair of transcribed regions occurs. Totoki
et al.4 also detected 90 somatic substitutions and seven small somatic insertions
and deletions in protein-coding regions,
including in two well-known tumor suppressor genes for HCC (TP53 and AXIN1).
They found somatic alterations in five
genes that have previously been reported
in other cancers but were unknown in
HCC. Whole-exome sequencing revealed
47 somatic substitutions. Among the validated substitutions, a nonsense substitution
in TSC1 was not detected by whole-genome
sequencing as only 13.2% of the tumor
Outer ring: chromosome ideograms
Somatic mutations in coding regions
Blue: substitution
Red: small deletion
Orange: small insertion
Copy number changes
Red: copy number loss
Green: copy number gain
Inner circle: chromosomal rearrangements
Green: inversion
Red: deletion
Purple: translocation
Figure 1 | Whole-genome view of somatically acquired alterations in the liver cancer genome.
Permission obtained from Nature Publishing Group © Totoki et al. Nat. Genet. 43, 464–469 (2011).
JANUARY 2012 | S35
HCC, as mutations were found throughout the coding region of this gene, all of
which were predicted to inactivate protein
function. They evaluated the prevalence
of ARID2 mutations in 106 additional
HCC samples of other etiology; the results
revealed that ARID2 mutations were more
common in HCV-related HCC (14%)
compared with HBV-related HCC (2.0%).
In addition, the prevalence of TP53 mutations was significantly higher in HCC of
individuals from China than those from
the USA and Europe. These findings could
be another proof of principle that etiological background and mutation profiles are
mutually correlated.
In 2011, two genome-wide association
studies (GWASs) by Japanese investigators
reported on genetic susceptibility to HCVrelated HCC. Kumar et al. 6 performed a
GWAS using 432,703 autosomal single
nucleotide polymorphisms (SNPs) in 721
individuals with HCV-related HCC and
2,890 HCV-negative controls of Japanese
origin. The validation study in 673 patients
with HCC and 2,596 controls using eight
candidate SNPs identified a locus in the
5' flanking region (rs2596542) of the gene
encoding major histocompatibility complex
class I polypeptide-related sequence A
(MICA) protein as strongly associated
with HCV-related HCC (odds ratio 1.39).
MICA is a membrane protein that activates natural killer cells and CD8 + T cells
by acting as a ligand for NKG2D. The risk
allele of rs2596542 was associated with low
levels of soluble MICA, which were shown
to be proportional to levels of membranebound MICA, which suggests that MICA
potentially has a suppressive role in HCV
infection (and thus progression to HCC).
Miki et al.7 applied the same technique in
212 individuals with HCV-related HCC and
765 individuals with chronic HCV without
HCC. The validation study in 710 cases
and 1,625 controls identified one intronic
SNP in the DEPDC5 gene as associated with
susceptibility to HCV-related HCC (odds
ratio 1.75).
Carcinogenesis is basically a collapse in
a biological system caused by the random
accumulation of genomic, transcriptomic
and epigenomic alterations. Using highthroughput sequencing, Totoki et al. 4
found more than 10,000 mutations in the
HCC tumor, which suggests that many of
these are meaningless passenger mutations. To find a major player or a driver
mutation, whole-genome sequencing in
large sample sets is needed; the ICGC
S36 | JANUARY 2012
intend sample sets of 500 (in fact the study
group updated its whole-genome data on
27 HCCs as of July 2011).8 Increasing the
number of sample sets will contribute not
only to finding common genetic alterations among indivi­duals but also to the
establishment of molecular classifications
of cancers.
GWASs are an undoubtedly powerful
tool for elucidating genetic susceptibility
to cancer development. However, the odds
ratios in both of the studies described here
were less than two. Given that clinical risk
factors for HCC are well established by past
epidemiological studies (for example, it is
widely recognized that males are at least
two times more susceptible to HCC than
females), genetic risk factors revealed by
GWASs should yield odds ratios much
higher than two in order to influence daily
clinical practice. Thus, the 1000 Genomes
Project,9 which intends to develop a catalog
of common human genetic variants with a
frequency of ≥1% by using next-generation
sequencing technologies, could have an
important role.
Despite difficulties in handling large
amounts of data, next-generation sequencing technologies will lead to revolutionary
change in biomedical research. We have
reason to be optimistic; if the major problems are associated with limited computing resources, according to Moore’s law
(that is, the number of transistors on a chip
doubles approximately every 2 years), these
resources are likely to progress rapidly. A
new sequencing era in HCC has begun.
Department of Gastroenterology, Graduate
School of Medicine, The University of Tokyo,
7‑3‑1 Hongo, Bunkyo-ku, Tokyo 113‑8655,
Japan (R. Tateishi). Yamanashi Prefectural
Hospital Organization, 1‑1‑1, Fujimi, Kofu-shi,
Yamanashi-ken 400‑8506, Japan (M. Omata).
Correspondence to: M. Omata
[email protected]
The authors appreciate the advice given by
Dr Yoshinari Asaoka in preparing the manuscript.
Competing interests
The authors declare no competing interests.
Jemal, A. et al. Global cancer statistics. CA
Cancer J. Clin. 61, 69–90 (2011).
Calle, E. E., Rodriguez, C., Walker-Thurmond, K.
& Thun, M. J. Overweight, obesity, and mortality
from cancer in a prospectively studied cohort of
US adults. N. Engl. J. Med. 348, 1625–1638
Llovet, J. M. et al. Sorafenib in advanced
hepatocellular carcinoma. N. Engl. J. Med. 359,
378–390 (2008).
Totoki, Y. et al. High-resolution characterization
of a hepatocellular carcinoma genome. Nat.
Genet. 43, 464–469 (2011).
Li, M. et al. Inactivating mutations of the
chromatin remodeling gene ARID2 in
hepatocellular carcinoma. Nat. Genet. 43,
828–829 (2011).
Kumar, V. et al. Genome-wide association study
identifies a susceptibility locus for HCV-induced
hepatocellular carcinoma. Nat. Genet. 43,
455–458 (2011).
Miki, D. et al. Variation in the DEPDC5 locus is
associated with progression to hepatocellular
carcinoma in chronic hepatitis C virus carriers.
Nat. Genet. 43, 797–800 (2011).
International Cancer Genome Consortium.
Dataset Summary. International Cancer
Genome Consortium [online], http:// (2011).
1000 Genomes. About the 1000 Genomes
Project. 1000 Genomes [online], (2011).
IBD IN 2011
Advances in IBD management
—towards a tailored approach
Guillaume P. Pineton de Chambrun and William J. Sandborn
Important advances have been made in the management of IBD in 2011.
Research has focused on optimizing the currently available therapies and
taking a more tailored approach to each individual patient.
Pineton de Chambrun, G. P. & Sandborn, W. J. Nat. Rev. Gastroenterol. Hepatol. 9, 70–72 (2012);
published online 10 January 2012; doi:10.1038/nrgastro.2011.248
During the past decade, large genome-wide
association studies in patients with Crohn’s
disease or ulcerative colitis have brought
new insights into IBD patho­g enesis. 1
Despite this improved understanding of the
underlying mechanisms regulating mucosal
inflammation, most of the newly developed
targeted therapies for IBD are yet to reach
clinical practice. Anti-tumor necrosis factor
(TNF) agents, immuno­suppressive drugs
(thiopurines and methotrexate), steroids
and mesalazine remain the only thera­
peutic options for patients with IBD. A large
part of IBD clinical research in 2011 was
focused on the optimization of currently
available therapies.
One important advance in 2011 was an
extension of the indications for currently
available IBD drugs. Three anti-TNF agents
are presently available for Crohn’s disease—
infliximab, adalimumab and certolizumab
pegol (only approved in the USA)—whereas
infliximab is the only approved agent for
ulcerative colitis.2 A 2011 phase III trial3
evaluated the efficacy of two adalimumab
dosing regimens in patients with moderateto-severe ulcerative colitis. Patients (n = 390)
were randomly assigned to receive one
of two adalimumab induction protocols:
ADA160/80 (adalimumab subcutaneously
160 mg at week 0, 80 mg at week 2 and 40 mg
at week 4 and 6); and ADA80/40 (adalimumab subcutaneously 80 mg at week 0,
40 mg at weeks 2, 4 and 6) or placebo. At
week 8, the rates of clinical remission were
18.5% and 10% in patients treated with
ADA160/80 and ADA80/40, compared
with 9.2% in those on placebo. In the subgroup analysis, increased baseline C‑reactive
protein concentrations and body weight
were associated with reduced remission
rates at week 8 (especially in the ADA160/80
group), indicating that some patients might
require a higher dose of adalimumab to
induce remission. These data provide
the basis for extension of adalimumab to
ulcerative colitis, and raises the possibility
that treatment optimization (potentially
through therapeutic drug monitoring)
could further increase drug efficacy.
In this context, new research should
be undertaken to tailor IBD treatment
to each individual. Research addressing
the selection of patients and drugs based
on predefined and validated criteria that
predict treatment efficacy and the need
for treatment intensification or even treatment discontinuation are required to
further improve the outcome of patients
with IBD. Among potential criteria useful
for treatment decision-­making, complete
mucosal healing has emerged as an important surrogate marker of ‘deep remission’
in IBD and evidence is accumulating that
mucosal healing can alter the course of the
disease when achieved.4 Indeed, anti-TNF
agents can induce and maintain mucosal
healing in patients with IBD, and mucosal
healing was associated with longer clinical remission and fewer disease-related
hospitalizations in patients with Crohn’s
disease treated with infliximab.4 In 2011,
data from the ACT‑1 and ACT‑2 studies
demonstrated the importance of mucosal
■■ Adalimumab, a fully humanized
recombinant monoclonal antibody
against tumor necrosis factor, is effective
for induction of remission in patients
with moderate-to-severe active ulcerative
■■ Early achievement of mucosal healing
in patients with ulcerative colitis treated
with infliximab decreases the risk of
colectomy during follow-up5
■■ Switching to adalimumab for reasons
of convenience or cost in patients with
Crohn’s disease in clinical remission on
maintenance therapy with infliximab is
not advised7
■■ In Crohn’s disease, discontinuing
infliximab in those in remission on
combination therapy (infliximab and
thiopurines) might be possible in
selected patients, but cannot as yet be
recommended in clinical practice8
■■ Thiopurines do not increase the risk
of complications during pregnancy in
women with IBD10
healing as a therapeutic goal in patients with
ulcerative colitis.5 Patients with refractory,
moderate-­to-severe active disease (n = 364
in each study) received intravenous placebo,
inflix­imab 5 mg/kg or infliximab 10 mg/kg
at weeks 0, 2 and 6 and then every 8 weeks
until week 46 in ACT‑1 and week 22 in
ACT‑2. The 2011 re-analysis of previously
published data showed that infliximabtreated patients with mucosal healing at
week 8 after treatment were less likely to
progress to colectomy during the first year
of follow-up.5 Thus, early evaluation of a
‘hard’ end point such as mucosal healing
during the treatment induction phase could
be useful to determine which patients need
treatment intensification, and to decrease
disease-related complica­tions.
The optimal strategy for inducing and
maintaining remission for 1 year in Crohn’s
disease is combination therapy with antiTNF agents and thiopurines. 6 For those
in clinical remission on anti-TNF maintenance therapy, a number of questions exist
regarding switching between or discontinuing these agents. In Crohn’s disease,
both inflix­imab and adalimumab are more
effective than placebo for the maintenance
of clinical remission, 7 but prospective
trials directly comparing the efficacy of
infliximab and adalimumab are lacking.
Although switching between these two
agents is effective in patients with Crohn’s
disease who previously responded to one
agent and then lost response, switching in
patients in clinical remission has not been
studied. Because adalimumab can be selfadministered, patients and physicians might
wish to consider switching from infliximab
to adalimumab for convenience or costs.
The clinical outcome of electively switching from infliximab to adalimumab in
patients with Crohn’s disease who were in
clinical remission on infliximab maintenance therapy was investigated in a 2011
prospective trial.7 Patients (n = 73) treated
with a stable infliximab regimen (infusion
intervals at least every 6 weeks) for the past
6 months were randomly assigned to continue infliximab (5 mg/kg intravenously)
at the same interval or to switch to adali­
mumab (80 mg subcutaneously at inclusion, then 40 mg every other week). At
1 year, discontinuation of treatment was
observed in 10 of 36 (six due to intolerance, four due to loss of efficacy) patients
in the adalimumab group and one patient
in the infliximab group. Moreover, markedly more patients preferred subcutaneous
adalimumab over intravenous infliximab.
Nevertheless, within 1 year, almost one in
three patients who switched to adalimumab
returned to infliximab therapy to control
Crohn’s disease. On the basis of this inferior clinical outcome, switching should be
reserved only for those with loss of response
or intolerance.
Another important issue is whether,
and when, to stop treatment with antiTNF agents in patients on maintenance
therapy who have long-standing remission.
Indeed, patients and physicians might wish
to consider discontinuation of treatment
for a variety of reasons—concerns about
long-term safety, perceived risk during
pregnancy and costs. A 2011 prospective study 8 investigated the risk of relapse
after withdrawal of infliximab in patients
with Crohn’s disease (n = 125 on infliximab
and anti­metabolite combination therapy)
in long-standing remission. Overall, 52
relapses occurred in 115 patients who
stopped inflix­i mab and continued anti­
metabolites, with an estimated risk of
relapse over 1 year and 2 years of 43.9% and
52.2%, respectively. Factors associated with
low risk of relapse were identified, most corresponding to surrogate markers of intestinal inflammation (Box 1). Re-treatment
of relapsing patients with inflix­imab was
generally effective and well tolerated
when evaluated just before the third infusion. Notably, about one-half of selected
patients who stopped infliximab—those
receiving combination therapy for at least
JANUARY 2012 | S37
Box 1 | Relapse in Crohn’s disease*
Risk factors
■■ Corticosteroid use 6–12 months before
infliximab discontinuation
■■ No previous surgical resection
■■ Male sex
■■ Hemoglobin level ≤145 g/l
■■ Leukocyte count >6 × 109 per l
■■ CDEIS >0
■■ hsCRP level ≥5 mg/l
■■ Infliximab trough level ≥2 mg/l
■■ Fecal calprotectin level ≥300 μg/g
Clinical relapse rates‡
■■ <4 risk factors: <20%
■■ 4 risk factors: ≈40%
■■ 5–6 risk factors: ≈80%
■■ >6 risk factors: 100%
*Factors associated with time to relapse in
patients with Crohn’s disease in clinical
remission on maintenance infliximab therapy and
antimetabolites who stopped infliximab.8
Number of risk factors before infliximab
discontinuation. Abbreviations: CDEIS, Crohn’s
disease endoscopic index of severity; hsCRP,
high-sensitivity C-reactive protein.
1 year and in corticoid-free clinical remission—experienced a clinical relapse within
1 year. By comparison, such patients would
be expected to have ~10% relapse rate over
1 year with continued stable combination
therapy. Overall, the risk of relapse is high
after discontinuing anti-TNF therapy and,
at the individual patient level, the ability
to predict sustained remission after withdrawal of anti-TNF therapy is limited. These
data should help the design of future clinical
trials, but should not be used as a basis for
changing clinical practice at present.
One reason for the interest in discontinuing anti-TNF therapy relates to their
long-term safety. Similar concerns exist
for thiopurines, the alternative long-term
therapy for patients with IBD. In 2009,
results from the French CESAME cohort 9
demonstrated that patients receiving thiopurines for IBD have a fivefold increased
risk of developing lymphoproliferative
disorders; the absolute risk of developing
lymphoproliferative disorders was largely
limited to older patients (>65 years), with
thiopurines fairly safe in younger patients
(<50 years). Thus, selection of patients and
treatment type should include criteria that
predict drug safety. In particular, the use
of thiopurines during pregnancy is still
controversial. In 2011, a nested study 10 in
the CESAME cohort (19,486 patients with
IBD total; 11,006 women) investigated the
safety of thiopurines during pregnancy. Of
215 pregnancies reported in 204 women,
no differences were observed between the
S38 | JANUARY 2012
patient groups—those on thiopurines, other
IBD drugs or no medication—in terms of
the risk of pregnancy failure, prematurity,
low birth weight or intrauterine growth
retardation. Although further epidemio­
logical studies are needed to evaluate the
frequency of low event rates (such as fetal
malformations), use of thiopurines in
women with IBD who wish to become pregnant seems to be generally safe and should
not be discouraged.
Development of new promising mol­
ecules (for example, ustekinumab or
vedolizumab) is ongoing and the results of
large therapeutic clinical trials are expected
soon. However, currently, tailoring available
thera­pies to each patient’s needs might be
the best opportunity to improve the treatment outcomes in IBD and is a key message
of the 2011 findings.
Inserm U995, Université Lille Nord de France,
Lille F‑59000, France (G. P. Pineton de
Chambrun). Division of Gastroenterology,
University of California San Diego, 9500
Gilman Drive, La Jolla, CA 92093‑0956, USA
(W. J. Sandborn).
Correspondence to: W. J. Sandborn
[email protected]
Competing interests
W. J. Sandborn declares associations with the
following companies: Abbott, Janssen, Merck. See
the article online for full details of the relationships.
G. P. Pineton de Chambrun declares no competing
Barrett, J. C. et al. Genome-wide association
defines more than 30 distinct susceptibility loci
for Crohn’s disease. Nat. Genet. 40, 955–962
2. Rutgeerts, P. et al. Infliximab for induction and
maintenance therapy for ulcerative colitis.
N. Engl. J. Med. 353, 2462–2476 (2005).
3. Reinisch, W. et al. Adalimumab for induction of
clinical remission in moderately to severely
active ulcerative colitis: results of a
randomised controlled trial. Gut 60, 780–787
4. Pineton de Chambrun, G., Peyrin-Biroulet, L.,
Lemann, M. & Colombel, J. F. Clinical
implications of mucosal healing for the
management of IBD. Nat. Rev. Gastroenterol.
Hepatol. 7, 15–29 (2010).
5. Colombel, J. F. et al. Early mucosal healing with
infliximab is associated with improved longterm clinical outcomes in ulcerative colitis.
Gastroenterology 141, 1194–1201 (2011).
6. Colombel, J. F. et al. Infliximab, azathioprine, or
combination therapy for Crohn’s disease.
N. Engl. J. Med. 362, 1383–1395 (2010).
7. Van Assche, G. et al. Switch to adalimumab in
patients with Crohn’s disease controlled by
maintenance infliximab: prospective
randomised SWITCH trial. Gut http://‑2011‑300755.
8. Louis, E. et al. Maintenance of remission
among patients with Crohn’s disease on
antimetabolite therapy after infliximab therapy
is stopped. Gastroenterology http://dx/
9. Beaugerie, L. et al. Lymphoproliferative
disorders in patients receiving thiopurines for
inflammatory bowel disease: a prospective
observational cohort study. Lancet 374,
1617–1625 (2009).
10. Coelho, J. et al. Pregnancy outcome in patients
with inflammatory bowel disease treated with
thiopurines: cohort from the CESAME Study.
Gut 60, 198–203 (2011).
Translating the microbiota
to medicine
Fergus Shanahan
Interest in the gut microbiota has escalated with growing appreciation
of the role of indigenous microbes in the health of the host and in the
pathogenesis of several intestinal and extraintestinal disorders. The
microbiota has become a plausible target for drug and dietary therapy
and a repository from which bioactive agents or new drugs can be mined.
Shanahan, F. Nat. Rev. Gastroenterol. Hepatol. 9, 72–74 (2012); published online 20 December 2011;
The commensal microbiota is one of the
hottest areas in medicine, with converging
interests from diverse disciplines including
nutrition and food science, micro­biology,
gastroenterology, immunology and metabolic
medicine. Microbial signaling is required
for the development and maintenance of
digestive, immunologic and neurobiologic
functions within the host. Therefore, it can
be anticipated that the list of disorders within
and outside the gut involving some disturbance of host–microbe interactions will
increase. Several high impact publications
attest to the scope and richness of human
micro­biome explora­tion, a representative
sample of which from 2011 is addressed here.
Microbes account for 10-fold more cells
and 100-fold more genes than those of their
human host. The information encoded in
the microbiome supplements that of the
human genome by provision of trophic,
metabolic and protective signals to the
host. Although metagenomic sequencing
has shown a predominance of two major
families or phyla in the human gut in
roughly equal proportions (Firmicutes and
Bacteroidetes), a report on over 20 newly
sequenced metagenomes from four countries, in combination with previous data,
demonstrated that variation of the micro­
biota at the species level is not continuous
and tends to congregate within distinct
clusters or enterotypes. 1 These enterotypes are identifiable by their enrichment
in Bacterioides (enterotype 1), Prevotella
(enterotype 2) or Ruminococcus (enterotype 3) and are known to be unrelated to
nationality or host characteristics such as
BMI, age or gender, although their diag­
nostic potential remains to be explored.
Diet seems to be the main lifestyle or
environ­m ental modifier of the microbiota, and it is noteworthy that although
short-term dietary interventions have a
rapid influence on microbial composition
in the gut, only long-term diets are associated with the microbial enterotypes. For
example, dietary protein and animal fat
favors Bacteroides, whereas carbohydrates
seem to favor Prevotella.2
The relationship between diet, microbes
and the metabolic health of the host, including the risk of obesity and diabetes, continues to generate intriguing information.3
The translation of this information to clinical medicine took a step forward in 2011
with the discovery of a direct link between
the gut microbiota and the risk of athero­
sclerosis caused by microbial-­dependent
metabolism of dietary phospholipids
and the generation of pro-atherosclerotic
metabolites. 4 Three metabolites of the
dietary lipid phosphatidylcholine were
identified by a metabolomics approach to
predict the risk of cardiovascular disease.
The participation of the microbiota in the
generation of the metabolites is obligate and
suggests that a pro­biotic or other form of
dietary manipulation may have prophy­lactic
potential against heart disease.
Although the microbiota shapes intestinal immunity and the metabolic welfare
of the host, host–microbe interactions
are bidirectional. The mucosal immune
system influences the composition and
pro­inflammatory potential of the gut
Family size
Mucosal immunity
human cells
microbial cells
Disease risk
Repository of bioactive agents
(e.g. antimicrobial, anti-inflammatory, metabolic)
Figure 1 | Clinical importance of the gut microbiota. The human metabolome is the composite
product of the human genome and the microbiome, the latter providing trophic, metabolic and
protective signals to the host that are crucial for development and homeostasis. Host–microbe
signaling is bidirectional; in susceptible individuals, aberrant host–microbe interactions
represent a risk factor for disease. Several aspects of a modern lifestyle influence the
composition of the microbiota, with the most important being diet. Microbe–host signaling, as
well as microbe–microbe signaling, within the gut represents a rich repository from which
bioactive agents for drug discovery can be mined.
microbiota, and disturb­a nces of innate
immunity have been linked with aberrant
expansion of components of the microbiota that have been associated with risk
of inflammatory and metabolic disease.
For example, mice lacking Toll-like receptor (TLR) 5 develop obesity and metabolic
syndrome, which seems to be dependent
on alterations in the microbiota.5 Other
models of defective innate immunity have
been associated with the emergence of a
‘colitogenic’ microbiota with the capacity to
transfer colitis to normal recipients and, in
one report, Koch’s postulates were fulfilled
in a host-genotype-specific way.6
The regulatory mechanisms used by the
host to maintain compositional equilibrium
within the microbiota are uncertain, but a
newly identified molecular pathway highlights the pivotal role of intestinal epithelial
cells.7 The epithelium contributes to both
afferent and efferent limbs of the mucosal
immune response by detecting microbeassociated molecular patterns (MAMPs),
using surface receptors and intra­cellular
sensors, and by responding to danger
signals with the production of bacteriocides,
cytokines or chemokines. Intracellular
inflammasomes are multiprotein complexes comprised, in part, of nod-like
receptors (NLRs), such as NLRP6, and they
sense endogenous and exogenous stress
or damage-associated molecular patterns.
In response to pathogenic components of
the commensal microbiota, the intestinal
epithelium mobilizes the NLRP6 inflamma­
some, triggering a cascade of events that
includes activation of caspase 1, maturation of IL‑18, recruitment of γ‑interferonproducing natural killer cells and T cells
and enhanced bactericidal activity of local
macrophages. When NLRP6 is deficient,
the epithelium fails to respond appropriately to pathogens and the composition of
the commensal bacteria may become colito­
genic. This provokes epithelial production
Key advances
■■ The diversity of species within the gut
microbiota is organized into identifiable
clusters or enterotypes that are
correlated with long-term but not
short-term dietary patterns1,2
■■ A direct link between the intestinal
microbiota, dietary phosphatidylcholine
and risk of atherosclerosis has been
■■ Mechanisms by which the host
distinguishes harmless commensals
from pathogens are uncertain, but
a molecular cascade within enteric
epithelial cells (triggered by the NLRP6
inflammasome) could have a role in
maintaining a balanced composition
within the microbiota7
■■ The metabolic diversity within the
microbiota represents a repository from
which bioactive agents can be mined,
as shown with the discovery of an
antimicrobial bacteriocin with specific
activity against Clostridium difficile10
JANUARY 2012 | S39
of the CCL5 chemokine, which recruits
neutrophils and manifests as IBD.7
The identity of the activating trigger for
the NLRP6 inflammasome in epithelial cells
is unknown. The trigger might be a damageinduced molecular pattern, which is one
mechanism by which the host could distinguish harmless commensals from pathogens,
both of which express similar molecular patterns as ligands for TLRs. However, in some
instances, the pathogen and commensal
dichotomy might simply relate to the context
of their encounter with the immune system.
Thus, a commensal in the wrong place will
be treated as a pathogen; likewise, the handling of commensals in certain genetically
susceptible individuals might be similar
to that of pathogens. Another explanation
for immunologic discrimination between
pathogens and commensals could involve
recognition of symbiotic bacterial molecules in a process that favors colonization
with commensals.8 An immunomodulatory
polysaccharide produced by the prominent
gut commensal, Bacteroides fragilis, has
been reported to suppress T H17 effector
cells by signaling through TLR2 on regulatory T cells, thereby enabling the commensal
to avoid an adverse immune response and
successfully colonize the host. The response
to polysaccharide is distinct from that seen
with other TLR2 ligands that promote clearance of pathogens. The immunomodulatory
properties of polysaccharide also have efficacy in an animal model of IBD, confirming
the potential for mining the microbiota for
drug discovery (Figure 1).
Other examples of bacterial-derived
metabolites with therapeutic potential
include the production of a soluble protein
ligand for the epidermal growth factor
receptor by Lactobacillus rhamnosus GG
(which attenuates intestinal inflammation
by inhibiting cytokine-induced apoptosis in
intestinal epithelial cells), and the discovery of an antimicrobial agent with narrow-­
spectrum activity against Clostridium
difficile.9,10 The latter was uncovered by an
extensive screen of fecal colonies for antimicrobial producers and resulted in the
identification of a strain of B. thuringiensis
that produces a heterodimeric bacteriocin, thuricin CD, which has potent activity against C. difficile. Using a distal colon
model, thuricin CD was shown to be as
effective as vancomycin and metronid­azole
but exhibited a narrower spectrum of activity without causing ‘collateral damage’ to the
dominant phyla within the surrounding
commensal microbiota.
S40 | JANUARY 2012
What can we expect from this field in the
immediate future? Microbial enterotypes
are likely to be refined and correlated with
human genotypes with respect to disease
risk, and longitudinal studies will shed light
on the impact of lifestyle variables over time.
However, as molecular profiling continues
apace, studies of the microbiota should be
complemented with a return to culturebased in vitro studies to fulfil the promise
of mining the microbiota and to understand the molecular basis of host–microbe
interactions in health and disease.
Department of Medicine, Clinical Sciences
Building, Cork University Hospital, Wilton, Cork,
[email protected]
F. Shanahan is supported, in part, by Science
Foundation Ireland.
Competing interests
The author declares associations with the following
companies: Alimentary Health Ltd, GlaxoSmithKline,
Procter & Gamble. See the article online for full
details of the relationships.
Arumugam, M. et al. Enterotypes of the human
gut microbiome. Nature 473, 174–180 (2011).
Wu, G. U. et al. Linking long-term dietary
patterns with gut microbial enterotypes.
Science 334, 105–108 (2011).
3. Shanahan, F. & Murphy, E. The hybrid science
of diet, microbes, and metabolic health. Am.
J. Clin. Nutr. 94, 1–2 (2011).
4. Wang, Z. et al. Gut flora metabolism of
phosphatidylcholine promotes cardiovascular
disease. Nature 472, 57–63 (2011).
5. Vijay-Kumar, M. et al. Metabolic syndrome
and altered gut microbiota in mice lacking
Toll-like receptor 5. Science 328, 228–231
6.Bloom, S. M. et al. Commensal Bacteroides
species induce colitis in
host‑genotype‑specific fashion in a mouse
model of inflammatory bowel disease. Cell
Host Microbe 9, 390–403 (2011).
7. Elinav, E. et al. NLRP6 inflammasome
regulates colonic microbial ecology and risk
for colitis. Cell 145, 745–757 (2011).
8. Round, J. L. et al. The Toll-like receptor 2
pathway establishes colonization by a
commensal of the human microbiota. Science
332, 974–977 (2011).
9. Yan, F. et al. Colon-specific delivery of a
probiotic-derived soluble protein ameliorates
intestinal inflammation in mice through an
EGFR-dependent mechanism. J. Clin. Invest.
121, 2242–2253 (2011).
10. Rea, M. C. et al. Effect of broad- and narrowspectrum antimicrobials on Clostridium
difficile and microbial diversity in a model of
the distal colon. Proc. Natl Acad. Sci. USA 108
(Suppl. 1), 4639–4644 (2011).
Emerging concepts in
neurogastroenterology and motility
Keith A. Sharkey and Gary M. Mawe
Neurogastroenterology encompasses intrinsic and extrinsic neural
processes that regulate gut functions, sensation and related behaviors
such as ingestion. In 2011, key advances were made in understanding
gut–brain interactions, visceral sensation, serotonin signaling,
neurogenesis and neuromuscular transmission.
Sharkey, K. A. & Mawe, G. M. Nat. Rev. Gastroenterol. Hepatol. 9, 74–76 (2012);
published online 13 December 2011; doi:10.1038/nrgastro.2011.247
Neural control of the gastrointestinal tract
in both health and disease is a rapidly
evolving and intriguing subject area. Key
advances have been made on several fronts
in neurogastroenterology in 2011. Here, we
highlight a breadth of studies that represent
major milestones in our understanding of
the effect of nutrients and gut microbiota on
emotion and food intake; the role of stress
in visceral hypersensitivity; the concept
that enteric glia can serve as neuronal pre­
cursors; and the roles of serotonin signaling in the gut. In addition, we discuss the
identifica­tion of a novel class of cells that
could mediate inhibitory neuromuscular
signaling in the gastrointestinal tract.
It is becoming increasingly clear that
signals arising in the lumen of the gastro­
intestinal tract can lead to changes in emotional state and behaviors such as food
intake. The notion that foods with a high fat
content are ‘comfort foods’ was substantiated this year by MRI studies demon­strating
that intragastric infusion of fatty acid positively enhanced emotional states, decreased
hunger scores and increased neural activity in the regions of the brain that process
emotions. 1 These findings indicate that
luminal nutrients can have acute effects on
mood as well as satiety. Evidence indicates
that endocannabinoid signaling in the gut
regulates fat consumption. Thus, the capacity to regulate fat intake exists within the
gut, and this process could, in turn, have
an effect on emotional state and long-term
energy balance.
In addition to nutrients, gut–brain
communication can also be influenced
by enteric microflora, including resident
microbes and ingested probiotics. A recent
study has shown that probiotic bacteria
influence emotional behavior by modulating the subunits of receptors of the neuro­
transmitter γ-aminobutyric acid, and
attenuate anxiety via activation of vagal
pathways.2 Probiotic treatment strategies
might, therefore, prove to be beneficial in
stress-related disorders (such as anxiety and
depression), which are common comorbidi­
ties of functional and inflammator y
bowel disorders.
Although stress is known to potentiate visceral pain and discomfort, a lack of
adequate animal models has meant that
the mechanisms that underlie this form
of visceral hypersensitivity have not been
resolved. Advances in the past year have
provided insights into peripheral and central
mechanisms and have helped to explain how
stress exacerbates visceral pain. Following
the resolution of infectious colitis in mice,
induction of stress resulted in exaggerated
peripheral nociceptive signaling—which is
analogous to post­infectious IBS.3 The hyperexcitabity of primary afferent neurons in this
model is associated with enhanced expression of β-adrenergic and gluco­c orticoid
receptors in these cells. Interestingly, the
effects of stress are mimicked by agonists
of these receptors, thus providing potential new therapeutic targets. In addition to
changes in primary afferent neurons, stressinduced activation of astroglial cells in the
spinal cord also seems to contribute to visceral hypersensitivity through the modulation of glutaminergic signaling.4 These novel
observations highlight the importance of
spinal glia and glutamate metabolism in the
sensation of pain.
Glia in the brain and gut serve a wide
array of functions beyond their original
definition as the ‘glue’ that holds neurons
together. In the gut, these functions are
known to include metabolic regulation,
neurotransmission and support of barrier
integrity. Two independent studies published during the past year provide compelling evidence that enteric glia have the
■■ Nutrients and bacteria in the lumen of
the gut can affect mood and ingestive
behavior through vagal pathways1,2
■■ Peripheral and central mechanisms
contribute to stress-induced visceral
■■ Enteric glia can give rise to new enteric
■■ Neuronal serotonin protects the integrity
of the enteric nervous system and
regulates gastrointestinal motility and
■■ Fibroblast-like cells mediate inhibitory
purinergic neuromuscular transmission10
potential to give rise to neurons in adult
gut or in culture under certain restricted
conditions. Laranjeira and colleagues5 used
genetic lineage tracing to confirm previous results showing that neurogenesis does
not seem to occur in the enteric nervous
system under steady state conditions. This
observa­t ion was corroborated by Joseph
and colleagues6 who used incorporation
of a thymidine analogue to investigate
cell division. Remarkably, after injury to
the myenteric plexus, glia were shown to
generate new neurons in vivo.5 However,
the conditions under which neurons can
be replaced seem to be limited to injury to
the plexus. Gliogenesis was observed both
in steady-state conditions and in response
to injury, but the function of new glial cells
remains to be determined. In culture conditions, enteric glia could readily form new
neurons, which indicates that endogenous
pre­c ursors exist within a patient’s own
bowel and could be used for transplantation to replace neurons lost or damaged as
a result of idiopathic or acquired enteric
Serotonin (5-hydroxytryptamine; 5-HT)
in the gastrointestinal tract can trigger
motor, secretory and vasodilator reflexes
under physiological conditions, and acts
as a proinflammatory mediator and stimulator of emesis, pain and discomfort in
pathophysiological conditions. Changes
in serotonin signaling have been reported
in patients with functional gastrointestinal
disorders; however, the causative role of
serotonin in the symptoms of these conditions is not yet fully established. A report
suggests that mucosal serotonin could contribute to visceral pain in these individuals.7
In patients with IBS, spontaneous serotonin
release from the mucosa is increased, which
correlates with the severity of abdominal pain. Moreover, biopsy supernatants
from these individuals activate discharge
of extrinsic afferent fibers in an ex vivo rat
preparation, and this response is inhibited by granisetron—an antagonist of the
5-HT3 receptor.
The majority of serotonin is synthesized,
stored and released by entero­chromaffin
cells in the gastrointestinal mucosa; serotonin also serves as an enteric neuro­
transmitter, but the physiological role of
enterochromaffin cell and neuronal serotonin signaling has not been fully determined. Li and colleagues8 addressed this
issue using mice that lack the genes for
tryptophan hydroxylase 1 or 2 (enzymes
required for serotonin biosynthesis in
enterochromaffin cells and neurons, respectively). Although mice lacking mucosal
serotonin did not exhibit a clear phenotype
with regard to gut function, mice deficient
in neuronal serotonin exhibited lower neuronal density, slower intestinal transit and
accelerated gastric emptying when compared with healthy mice. These findings
indicate that neuronal serotonin protects
the integrity of the enteric nervous system
and contributes to normal gastrointestinal
motility. Mucosal serotonin can act as a
proinflammatory mediator, but Tsuchida
et al.9 demonstrated that activation of 5-HT4
receptors on enteric nerve terminals triggers an anti-inflammatory effect. 5-HT4
agonists facilitate acetylcholine release,
which, in turn, can dampen proinflammatory cytokine induction via α7 nicotinic
receptors on macrophages. This finding
suggests that 5-HT4 agonists might, by
inhibiting the inflammatory response and
promoting propulsive motility, have a bene­
ficial effect in certain conditions, such as
postoperative ileus.
One of the ongoing controversies in
neuro­gastroenterology over the past decade
has been the mechanism by which smooth
muscle cells receive inhibitory purinergic
signals from enteric motor neurons. These
signals do not seem to be mediated either
directly by smooth muscle or indirectly
by interstitial cells of Cajal because mice
lacking interstitial cells of Cajal still exhibit
purinergic inhibitory junction potentials,
and isolated smooth muscle cells exhibit
mixed excitatory and inhibitory responses
to ATP. Kurahashi and colleagues 10 shed
light on this dilemma in a report demonstrating that a novel class of excitable cells
(referred to as ‘fibroblast-like cells’), which
express platelet-derived growth factor
receptor α, exhibit all of the properties necessary to detect and transmit puri­nergic
JANUARY 2012 | S41
signals from nerve terminals to smooth
muscle. These interstitial cells should be
investigated for potential contributions to
gastrointestinal motor disorders.
The gastrointestinal dysfunctions that
fit under the umbrella of neurogastro­
enterology represent a considerable burden
to society with limited treatment options.
Continued efforts, such as those highlighted
here, will provide a better understanding of
these enigmatic disorders and open new
avenues for therapies of the future.
Hotchkiss Brain Institute & Snyder Institute of
Infection, Immunity and Inflammation,
Department of Physiology & Pharmacology,
University of Calgary, AB T2N 4N1, Canada
(K. A. Sharkey). Department of Anatomy &
Neurobiology, University of Vermont, Burlington,
VT 05405, USA (G. M. Mawe).
Correspondence to: K. A. Sharkey
[email protected]
Competing interests
The authors declare no competing interests.
Van Oudenhove, L. et al. Fatty acid-induced gutbrain signaling attenuates neural and behavioral
effects of sad emotion in humans. J. Clin. Invest.
121, 3094–3099 (2011).
Bravo, J. A. et al. Ingestion of Lactobacillus strain
regulates emotional behavior and central GABA
receptor expression in a mouse via the vagus
nerve. Proc. Natl Acad. Sci. USA 108,
16050–16055 (2011).
Ibeakanma, C. et al. Brain-gut interactions
increase peripheral nociceptive signaling in mice
with postinfectious irritable bowel syndrome.
Gastroenterology 141, 2098–2108 (2011).
Bradesi, S. et al. Role of astrocytes and altered
regulation of spinal glutamatergic
neurotransmission in stress-induced visceral
hyperalgesia in rats. Am. J. Physiol. Gastrointest.
Liver Physiol. 301, G580–G589 (2011).
Laranjeira, C. et al. Glial cells in the mouse
enteric nervous system can undergo
neurogenesis in response to injury. J. Clin.
Invest. 121, 3412–3424 (2011).
6. Joseph, N. M. et al. Enteric glia are multipotent
in culture but primarily form glia in the adult
rodent gut. J. Clin. Invest. 121, 3398–3411
7. Cremon, C. et al. Intestinal serotonin release,
sensory neuron activation, and abdominal pain
in irritable bowel syndrome. Am. J.
Gastroenterol. 106, 1290–1298 (2011).
8. Li, Z. et al. Essential roles of enteric neuronal
serotonin in gastrointestinal motility and the
development/survival of enteric dopaminergic
neurons. J. Neurosci. 31, 8998–9009 (2011).
9. Tsuchida, Y. et al. Neuronal stimulation with
5hydroxytryptamine 4 receptor induces antiinflammatory actions via α7nACh receptors on
muscularis macrophages associated with
postoperative ileus. Gut 60, 638–647 (2011).
10. Kurahashi, M. et al. A functional role for the
‘fibroblast-like cells’ in gastrointestinal smooth
muscles. J. Physiol. 589, 697–710 (2011).
Don’t miss out!
Sign up to free e-ALERTS and be notified when new articles are published
in Nature Reviews Gastroenterology & Hepatology. Research Highlights, News & Views,
Reviews, and Perspectives—you won’t miss a thing!
S42 | JANUARY 2012
New clues to environmental influences
in glomerular disease
Peter J. Nelson and Charles E. Alpers
Research into genetic susceptibilities for proteinuric glomerular diseases has uncovered key pathogenic
contributions from inheritable defects in podocytes. However, much less is known about environmental factors
that may initiate or propagate podocyte injury. Seminal reports in 2011 provided new mechanistic insights into
how this may occur.
Nelson, P. J. & Alpers, C. E. Nat. Rev. Nephrol. 8, 65–66 (2012); published online 13 December 2011; doi:10.1038/nrneph.2011.214
Over the past 10 years, numerous high-­
profile studies in forward and reverse genetics have mapped several genetic causes for
proteinuric glomerular diseases. 1 Many
identified genes encode mutant proteins
(for example, nephrin, podocin, or TRPC6)
or variant proteins (for example, APOL1 or
HLA-DQ) that render podocytes susceptible
to injury, highlighting the critical contribution of podocytes to the glomerular filtration barrier (GFB). This research is leading
towards prog­n ostic strategies to gauge
hereditary risk for disease and response to
treatment in protein­uric glomerular diseases. The findings have also, however,
exposed the comparatively little attention
given to identi­fying environmental factors
that unmask genetic susceptibilities.
One reason for this imbalance of research
in gene–environment interactions in
protein­uric glomerular diseases is the lack
of validated methods to measure and integrate effects from the environment.2 Unlike
genome-wide association studies, there is a
lack of techniques for ‘environment-wide’
association studies that can identify and
assign hierarchy to complex exposures from
diet, the microbiome, xenobiotics, and other
potential environmental insults that contribute to podocyte injury. With this deficiency
in mind, we briefly review seminal studies
in proteinuric glomerular diseases published
in 2011 which reported new pathogenic
mechanisms that may heighten the focus of
research in glomerular disease on the role of
the environment.
Membranous nephropathy, among
the most common causes of nephrotic
syndrome in adults, results from the
forma­t ion of glomerular subepithelial
KEY ADVANCES IN MEDICINE antibody–antigen immune complexes that
injure podocytes. Studies have shown that
antibodies to bovine serum albumin (BSA)
can induce membranous nephropathy in
animals, and exposure to BSA is common
in the human diet. Based on these findings,
Debiec, Ronco and colleagues hypothesized
that antibody–BSA immune complexes
cause secondary membranous nephro­pathy
in humans.3 The authors found that 11 of
50 patients with idiopathic membranous
nephropathy, but only two of 172 controls,
harbored high-titer serum antibody that
recognized epitopes in BSA not present in
human albumin. Seven of the 11 patients
with membranous nephropathy, but no controls, also had higher than normal serum
levels of BSA, and in four of these patients
the cationic BSA colocalized with the BSAspecific antibody in their subepithelial
immune complex deposits.
Key advances
■■ Bovine serum albumin has been identified
as a dietary cause of membranous
nephropathy,3 raising the possibility that
other dietary antigens may also cause
membranous nephropathy
■■ Soluble urokinase-type plasminogen
activator receptor is a likely cause of focal
segmental glomerulosclerosis that can
be induced by lipopolysaccharide,5 and
probably by other pathogen-associated
molecular patterns, potentially linking
podocyte injury to heightened exposure to
commensal and non-commensal microbes
■■ Overactivation of the podocyte
mammalian target of rapamycin pathway
in response to the dysregulated nutrient
environment of diabetes may contribute to
the development of diabetic nephropathy9
The study by Debiec, Ronco and colleagues
is a landmark study in that it identi­fied a circulating exogenous antigenic protein derived
from dietary exposure that is deposited in
glomerular walls and serves as a ‘planted’
antigen to induce membranous nephro­
pathy. The findings of this study raise the possibility that other dietary antigens may also
cause secondary membranous nephropathy.
Peripheral tolerance to protect podocytes
and the GFB from dietary antigens filtered
within the kidney is maintained by the renal
mononuclear phagocytic system. 4 This
system may not operate when immunogenic
dietary antigen is retained at the anionic GFB,
such as may occur for cationic BSA, leading
to its recognition by antibody and subsequent immune complex formation in situ.
Without epidemio­logic or experimental
clues, however, the challenge that lies in the
untargeted discovery of other dietary antigens in idiopathic membranous nephropathy
will be in knowing how best to sample candidates and map them to dietary sources and
potentially to other environmental stimuli.
Wei, Reiser and colleagues similarly translated knowledge gained through animal
studies to discover that soluble urokinase-­
type plasminogen activator receptor (suPAR)
is a probable cause of recurrent focal segmental glomerulosclerosis (FSGS).5 The system
involving suPAR, urokinase-type plasminogen activator receptor (uPAR) and urokinase-­
type plasminogen activator (uPA) is an
innate immune mediator of tissue remodeling during inflammation.6 Wei, Reiser et al.
showed that endotoxin (lipopolysaccharide),
produced by commensal bacteria in the gut
and ubiquitous in the environment,7 caused
proteinuria in mice by inducing systemic
JANUARY 2012 | S43
release of suPAR, which crossed the GFB
and activated podocyte β3 integrin in both
native uPAR+/+ kidneys and transplanted
uPAR –/– kidneys. The researchers also
found that experi­mentally sustained secretion of suPAR from sites outside the kidney
induced FSGS. In addition, they found that
serum suPAR level was significantly elevated in patients with primary FSGS (but
not in those with minimal-change disease
or membranous nephro­pathy) compared
with levels in healthy people; among those
with FSGS, serum suPAR levels were highest
in those who developed recurrent FSGS
after transplanta­tion. In patients in whom
FSGS recurred, glomerular lesions showed
activated podocyte β3 integrin, and suPAR
removed from serum by plasma­pheresis
correlated with remission of recurrent FSGS.
The results of this study provide a major
breakthrough in the study of FSGS on several
fronts. They identify a long-sought-after circulating factor that is causative of recurrent
FSGS, is a potential biomarker for the development of or the risk of development of FSGS,
and is a new potential therapeutic target for
treating a disease in which current therapies
show limited efficacy. Although not emphasized in the article, this study also provides
suggestive evidence that FSGS may be an
outcome of otherwise ‘clinically silent’ inflammation from environ­mental pathogens. The
marked upregulation of suPAR by lipopolysaccharide in mice suggests that chronic,
heightened exposure to lipopolysaccharide
or other pathogen-associated mol­ecular patterns (PAMPs) derived from commensal or
non-commensal microbes might cause elevated suPAR in humans, resulting in FSGS
in susceptible individuals.8 If found to be
true, this hypothesis could lead to prophylactic strategies that manipulate the microbiome to lessen exposure and secondary
damage to podocytes from these PAMPs.
Podocytes may also be injured by serving
as sensors and not simply by being bystanders of environmental insults. The mammalian target of rapamycin (mTOR) signaling
pathways are of great interest at present;
sirolimus (also known as rapamycin), a
drug that targets these pathways, is clinically
useful in immune suppression, as an antiproliferative agent in cardiac stenting, and
as a potential therapeutic agent for the treatment of renal cancer and polycystic disease.
More recently it has been shown that mTOR
pathways control aging in multiple model
systems. In 2011, the mTOR pathway in
podocytes emerged as a key molecular sensor
of the nutrient environment. Godel, Huber
S44 | JANUARY 2012
and colleagues examined gene expression
in microdissected glomeruli from human
kidneys with early diabetic nephropathy.9
They found marked upregulation of target
genes of the mTOR pathway, suggesting that
overactivation of this pathway in response
to the dys­regulated nutrient environment of
diabetes contributes to the development of
diabetic nephropathy.9 To test this hypothesis, the authors bred mice haplo­insufficient
for mTORC1 in podocytes, thereby curtailing
the activity of the mTOR pathway, and then
induced experimental diabetic nephro­pathy
with streptozotocin. Intriguingly, the mice
showed significant amelioration of diabetic
nephropathy, clearly demonstrating that
podocyte gene–environment interactions
can drive the progression of diabetic nephro­
pathy, and also identifying mTOR pathways
as a potential thera­peutic target to abrogate
the progression of diabetic nephropathy.
This selection of studies published in
2011 highlight three seemingly disparate
glomerular diseases, each with a different
pathogenesis, but which together account
for the majority of cases of nephrotic syndrome and end-stage renal disease in adults.
Each of these studies has identified new and
compelling mechanisms underlying glomerular injury. A common feature of these
studies is that each also indicates the importance of under­standing the complex role of
the environ­ment in mediating proteinuric
glomerular diseases. Although we may continue to hope that gene replacement therapy
can eliminate inheritable disease risks, we
also need to identify concurrent modifiable
environ­mental factors to protect susceptible
patients from developing disease. These
studies are signaling the way forward in
this quest.
Division of Nephrology (P. J. Nelson), Division of
Nephrology and Department of Pathology
(C. E. Alpers), University of Washington Medical
Center, 1959 NE Pacific Street, Seattle,
WA 98195, USA.
Correspondence to: P. J. Nelson
[email protected]
Competing interests
The authors declare no competing interests.
Machuca, E., Benoit, G. & Antignac, C. Genetics
of nephrotic syndrome: connecting molecular
genetics to podocyte physiology. Hum. Mol.
Genet. 18, R185–R194 (2009).
Renz, H. et al. Gene-environment interactions in
chronic inflammatory disease. Nat. Immunol.
12, 273–277 (2011).
Debiec, H. et al. Early-childhood membranous
nephropathy due to cationic bovine serum
albumin. N. Engl. J. Med. 364, 2101–2110
Nelson, P. J. et al. The renal mononuclear
phagocytic system. J. Am. Soc. Nephrol. (in
Wei, C. et al. Circulating urokinase receptor as
a cause of focal segmental glomerulosclerosis.
Nat. Med. 17, 952–960 (2011).
Smith, H. W. & Marshall, C. J. Regulation of cell
signalling by uPAR. Nat. Rev. Mol. Cell Biol. 11,
23–36 (2010).
Michel, O. Severity of asthma is related to
endotoxin in house dust. Am. J. Respir. Crit.
Care Med. 154, 1641–1646 (1996).
Kau, A. L, Ahern, P. P., Griffin, N. W.,
Goodman, A. L. & Gordon, J. I. Human nutrition,
the gut microbiome and the immune system.
Nature 474, 327–336 (2011).
Godel, M. et al. Role of mTOR in podocyte
function and diabetic nephropathy in humans
and mice. J. Clin. Invest. 121, 2197–2209
Connecting the dots toward a
polycystic kidney disease therapy
Vicente E. Torres and Peter C. Harris
Understanding the complex interactions between the various pathways
disrupted in polycystic kidney and liver disease is essential to identify and
optimize therapies for these disorders. Studies published in the past year
have demonstrated a functional interaction between the main proteins
implicated in these diseases and identified novel therapeutic approaches.
Torres, V. E. & Harris, P. C. Nat. Rev. Nephrol. 8, 66–68 (2012); published online 13 December 2011;
The enlarged kidneys and liver character­istic
of polycystic kidney disease (PKD) have for
centuries attracted attention. Pathologists
in the 19th century debated whether cysts
are develop­mental or the result of tubular
obstruction or excessive epithelial cell
prolifera­t ion. Microdissection, electron
micro­scopy and physiology studies carried
out during the 20th century showed that cysts
in patients with autosomal dominant PKD
(ADPKD) detach from renal tubules and
grow as blind sacs by a process that requires
cell prolifera­tion, net fluid secretion and
remodeling of extracellular matrix. After the
PKD1 and PKD2 genes and their encoded
proteins, polycystin‑1 and polycystin‑2,
were identified in 1994 and 1996, respectively, research in this area greatly accelerated. In the following years, genes mutated
in autosomal recessive PKD (PKHD1 encoding fibro­cystin) and in autosomal dominant
polycystic liver disease (PRKCSH encoding
glucosidase 2 subunit β, and SEC63 encoding translocation protein SEC63 homolog)
were identified. Polycystin‑1, polycystin‑2
and fibrocystin are membrane glyco­proteins,
whereas glucosidase 2 subunit β and SEC63
are endoplasmic reticulum resident proteins needed for translocation and folding
of integral membrane proteins and secreted
proteins. Polycystin‑1, polycystin‑2 and
fibrocystin have numerous interacting partners and their disruption affects multiple
signaling pathways. Despite the multi­plicity
of genes and signaling pathways involved in
PKD, interventions affecting many diverse
targets have been effective in animal models
of PKD, which indicates the presence of
connections between the various pathways.
Understanding these complex interactions
will be essential to effectively treat PKD.
Fedeles et al. used mouse mutants to
show that the main proteins implicated in
ADPKD, autosomal recessive PKD and
autosomal dominant polycystic liver disease
functionally interact.1 Loss of gluco­sidase 2
subunit β or SEC63 reduces polycystin‑1
and, to a lesser extent, polycystin‑2 expression, blocks polycystin‑1 trafficking to
primary cilia, causes renal and hepatic cysts,
worsens cystic disease in hetero­z ygous
Pkd1 +/− and Pkd2 +/− mice and induces
renal cystogenesis in Pkhd1 del 4/del 4 mice.
Transgenic overexpression of poly­cystin‑1,
but not polycystin‑­2, rescues the renal and
hepatic pheno­type of tissue-­selective Prkcshknockout or Sec63-knockout mice and
the renal phenotype of Pkhd1del4/del4 mice.
Polycystin‑1 is the rate-limiting component
in the polycystin‑1/poly­cystin‑2 complex as
its level of expression determines the severity
of the cystic pheno­type, but some functional
polycystin‑2 is essential for polycystin‑1
to exert this effect. Immunocytochemical
analysis revealed that the collecting duct is
the segment most susceptible to the cystogenic effect of reduced polycystin‑1 dosage.
Proteasome inhibition increases poly­cystin‑1
levels and attenuates cystic disease in Prkcshknockout models, thus offering a conceptual
■■ The main proteins implicated in polycystic
kidney disease (PKD)—polycystin‑1,
polycystin‑2 and fibrocystin—and in
autosomal dominant polycystic liver
disease (glucosidase 2 subunit β and
SEC63) functionally interact1
■■ The vasopressin V2 receptor antagonist,
tolvaptan, inhibits cystogenesis in vitro,2
and reduced the volume of polycystic
kidneys after 1 week of treatment, and
slowed the growth of polycystic kidneys
in a 3‑year, open-label study of patients
with PKD3
■■ Two insulin-sensitizing drugs used to
treat type 2 diabetes mellitus, metformin
and a peroxisome proliferator-activated
receptor‑γ agonist, inhibit cyst growth in
rodent models of PKD by different and
possibly complementary mechanisms4–6
■■ Macrophage infiltration contributes to the
proliferation of cyst-lining cells and PKD
■■ STAT3, a transcription factor that is
essential during development but
dispensable postnatally, is a novel
therapeutic target in PKD8–10
therapeutic approach to auto­somal dominant
polycystic liver disease and possibly ADPKD.
The central role of cyclic AMP in PKD,
and the ability to hormonally modulate
cyclic AMP in a cell-specific manner,
enables targeting of an important cystogenic pathway with relative safety. Among
hormonal systems that affect renal cyclic
AMP is the vasopressin–V2 receptor axis.
This target is attractive because V2 receptors are mostly restricted to the thick
ascending limb and collecting ducts, which
are main sites of cystogenesis where vasopressin is the major agonist of cyclic AMP.
These sites are continuously subjected to
tonic vaso­pressin action. Circulating vaso­
pressin levels are increased in ADPKD and
V2 receptors are overexpressed in poly­
cystic kidneys. Pharmacological and genetic
inhibition of vasopressin or V2 receptor
expression has been shown to be effective
in rodents. Reif et al. examined the effects of
tolvaptan on human ADPKD cyst epithelial
cells.2 Low concentrations inhibited vaso­
pressin-induced cyclic AMP production,
cell prolifera­tion, chloride secretion and cyst
growth in collagen matrices. Other analyses
showed that tolvaptan administration for
1 week reduced total kidney volume by 3.1%
in patients with ADPKD, and that 3 years of
tolvaptan therapy reduced kidney growth
by 70% compared to historical controls
(1.7% versus 5.8% increase per year, respectively).3 Changes in kidney volume and in
estimated glomerular filtration rate were
significantly and negatively correlated. This
finding supports the use of kidney volume
as a biomarker to monitor ADPKD progression. Further evaluation of tolvaptan for the
treatment of patients with ADPKD awaits the
conclusion of a randomized, double-blind
clinical trial (NCT00428948) in 2012.
Metformin and peroxisome proliferatoractivated receptor (PPAR)‑γ agonists are
widely used to treat type 2 diabetes melli­tus.
Takiar et al. found that metformin stimulates
the energy-sensing molecule AMP-activated
protein kinase (AMPK), inhibits the activities of AMPK-dependent cystic fibrosis transmembrane conductance regulator (CFTR)
and mammalian target of rapamycin in
Madin–Darby canine kidney renal epithelial
cells, and attenuates cyclic AMP-dependent
growth of Madin–Darby canine kidney cysts
in collagen matrices and of cysts in metanephric organ explants, and cystogenesis in
constitutive and inducible Pkd1-knockout
mice.4 Yoshihara et al. and Blazer-Yost et al.
found that the PPAR‑γ agonist pioglitazone
inhibits renal and hepatic cystogenesis in
PCK rats by possibly complementary mechanisms through the inactivation of mitogen-­
activated protein kinase 3 and mammalian
target of rapa­mycin,5 and inhibition of CFTR
synthesis and cyclic AMP-activated chloride
secretion.6 Because metformin and PPAR‑γ
agonists exert salutary effects by affecting the
same pathways through different mechanisms
(such as phosphoryla­tion and inhibition of
CFTR by metformin and inhibi­tion of CFTR
synthesis by pioglitazone), clinical trials of
metformin and PPAR‑γ agonist combinations in early ADPKD should be considered
to exploit their possible synergism.
Evidence accumulated over the past
two decades points to the importance of
inflamma­t ion in PKD. Karihaloo et al.
hypothesized that macrophage infiltration
contributes to the proliferation of cyst-lining
cells and PKD progression.7 This premise
was based on work showing that macrophages homing to the kidney after ischemia–­
reperfusion undergo a transition from
classically activated, pro­inflammatory cells
to alternatively activated cells that promote
epithelial cell proliferation. The investigators found that Pkd1-null cells secrete
large amounts of the macrophage chemo­
attractant C‑C motif chemokine 2 (also
known as monocyte chemotactic protein 1)
and C‑X‑C motif chemokine 16. Kidneys
from conditional Pkd1-knockout mice
and the Pkd2WS25/– mouse model of PKD2
exhibit a 10-fold increase in the number of
JANUARY 2012 | S45
macrophages (most with an alternatively
activated phenotype and aligned along cyst
walls). Macrophage depletion by intra­
peritoneal liposomal clodronate administra­
tion inhibits epi­thelial cell proliferation and
cyst growth and improves renal function.7
How the loss of polycystin expression leads
to increased cytokine production remains to
be determined. Approaches that inhibit the
expression or action of homing and prolifera­
tion signals provide novel strategies for
treating PKD.
Three studies have shown marked
upregula­tion of the signal transducer and
transcription activator (STAT)3 in patients
with ADPKD and in rodent PKD models.8–10
In the STAT3 signaling pathway, activation of
various cell surface growth factor receptors
and cytokine receptors induces specific tyrosine phosphorylation of the receptors, which
creates docking sites for latent cytoplasmic
STAT3. STAT3 is then phosphorylated at
tyrosine 705 by intrinsic tyrosine kinase activity of the activated growth factor receptors or
by cytokine receptor-associated Janus kinase.
The trigger for phosphorylation and activation of STAT3 in PKD is uncertain. Talbot
et al. showed that phosphorylated STAT3
is highly expressed in cyst-lining cells and
normal-­appearing tubules and inter­stitial
cells in proximity to cysts.8 This finding suggests that diffusible factors such as cytokines
and growth factors are involved in the activation of STAT3. Phosphorylated STAT3 homo­
dimers translocate to the nucleus and bind
to promoter elements of genes that regulate
cell differentiation, proliferation, apoptosis
and angiogenesis. Because STAT3 is critical during development (deletion of STAT3
leads to embryonic lethality), but is dispensable postnatally in conditional knockouts,
targeting STAT3 might be well tolerated in
patients with PKD. Takakura et al. screened
a small-molecule library using a cell-based
functional assay and found that pyrimethamine, a drug used to treat malaria and
toxoplasmosis, inhibits STAT3 signaling.9
Pyrimethamine and S3I‑201, an inhibitor
of STAT3 homodimer complex formation,
suppressed epithelial cell proliferation and
cystogenesis in an inducible Pkd1-knockout
mouse model without toxic effects. In another
study, Leonhard et al. found that curcumin,
a compound with anti-inflammatory and
antiproliferative properties, reduced STAT3
activation, attenuated cell proliferation
and cysto­genesis and delayed renal failure
from 105 to 119 days in an inducible Pkd1knockout model.10 As curcumin has poor
bioavailability, however, novel analogues with
S46 | JANUARY 2012
improved pharmacological profiles might be
more effective than curcumin itself.
In summary, insights into the complex
network of signaling pathways disrupted in
PKD have increasingly led to the identifica­
tion of potential therapies, some of which are
currently used for other indications. Those
compounds best supported by preclinical
studies and with the desired pharmaco­
logical profile should be prioritized for
clinical trials.
Division of Nephrology and Hypertension, Mayo
Clinic College of Medicine, 200 First Street SW,
Rochester, MN 55905, USA (V. E. Torres,
P. C. Harris).
Correspondence to: V. E. Torres
[email protected]
V. E. Torres and P. C. Harris are supported by the NIH
grant DK090728.
Competing interests
V. E. Torres declares an association with the
following company: Otsuka Pharmaceuticals. See
the article online for full details of the relationship.
P. C. Harris declares no competing interests.
Fedeles, S. V. et al. A genetic interaction
network of five genes for human polycystic
kidney and liver diseases defines polycystin-1
as the central determinant of cyst formation.
Nat. Genet. 43, 639–647 (2011).
Reif, G. A. et al. Tolvaptan inhibits ERKdependent cell proliferation, Cl– secretion, and
in vitro cyst growth of human ADPKD cells
stimulated by vasopressin. Am. J. Physiol. Renal
Physiol. 301, F1005–F1013 (2011).
3. Higashihara, E. et al. for the TEMPOFormula
and 156‑05‑002 Study Investigators. Tolvaptan
in autosomal dominant polycystic kidney
disease: three years’ experience. Clin. J. Am.
Soc. Nephrol. 6, 2499–2507 (2011).
4. Takiar, V. et al. Activating AMP-activated protein
kinase (AMPK) slows renal cystogenesis. Proc.
Natl Acad. Sci. USA 108, 2462–2467 (2011).
5. Yoshihara, D. et al. PPAR-γ agonist ameliorates
kidney and liver disease in an orthologous rat
model of human autosomal recessive
polycystic kidney disease. Am. J. Physiol. Renal
Physiol. 300, F465–F474 (2011).
6. Blazer-Yost, B. L. et al. Pioglitazone attenuates
cystic burden in the PCK rodent model of
polycystic kidney disease. PPAR Res. 2010,
274376 (2010).
7. Karihaloo, A. et al. Macrophages promote cyst
growth in polycystic kidney disease. J. Am. Soc.
Nephrol. 22, 1809–1814 (2011).
8. Talbot, J. J. et al. Polycystin-1 regulates STAT
activity by a dual mechanism. Proc. Natl Acad.
Sci. USA 108, 7985–7990 (2011).
9. Takakura, A. et al. Pyrimethamine inhibits adult
polycystic kidney disease by modulating STAT
signaling pathways. Hum. Mol. Genet. 20,
4143–4154 (2011).
10. Leonhard, W. N. et al. Curcumin inhibits
cystogenesis by simultaneous interference of
multiple signaling pathways: in vivo evidence
from a Pkd1-deletion model. Am. J. Physiol.
Renal Physiol. 300, F1193–F1202 (2011).
Biomarkers are transforming our
understanding of AKI
Lakhmir S. Chawla and John A. Kellum
Acute kidney injury (AKI) is a syndrome of decreased renal function that is
associated with an increased risk of death. Studies from 2011, particularly
in the field of AKI biomarkers, have provided important insights into the
diagnosis, prognosis, and treatment of AKI. These advances are now being
brought to the bedside to improve diagnosis and treatment of AKI.
Chawla, L. S. & Kellum, J. A. Nat. Rev. Nephrol. 8, 68–70 (2012); published online 17 January 2012;
Over 2 million people are afflicted by
acute kidney injury (AKI) worldwide. 1
AKI is a serious complication of acute
illness, typically occurring as a result of
underlying conditions such as sepsis, and
is independently associated with death. It
has become increasingly clear that AKI is
associated with the development of endstage renal disease (ESRD) and chronic
kidney disease (CKD).2,3 Individuals fortunate enough to survive a severe episode
of AKI are therefore at risk of develop­ing
CKD-related disease.
Our understanding of AKI has advanced
over the past decade, with a unified definition and standard criteria for staging, a
quantification of renal replacement therapy
(RRT) dosing, and the emergence of biomarkers related to AKI. In 2011, multiple
studies were published that improve our
understanding of AKI through the use of
various biomarkers. Perhaps the most illuminating study published in 2011 was one by
Paragas et al. that examined the expression
of neutrophil-related gelatinase lipocalin
(NGAL) using a luciferase reporter assay.4
Although NGAL has been shown in multiple
large cohort studies to be a biomarker for
early AKI diagnosis and for AKI prognosis,5
NGAL is expressed in multi­ple organs and
is thus not kidney specific. Understanding
how non-kidney sources of NGAL impact
the urinary NGAL signal and indeed
establishing the sources of NGAL in AKI
is essential. Paragas and colleagues4 conducted their novel experiment by creating
a reporter mouse for NGAL. They inserted
a double-fusion reporter gene that encoded
luciferase‑2 and mCherry (Luc2-mC) in
the locus for NGAL (lcn‑2). In so doing, the
investigators could assess the endogenous
NGAL message in real time and assess in
which organs the message was being trans­
cribed. In separate experiments, mice were
subjected to unilateral renal ischemia, bilateral renal ischemia or nephrotoxic injury
with cisplatin. In addition, the investigators
tested whether NGAL-Luc2-mC was activated in pre-renal azotemia and ascertained
the segments of the nephron responsible
for urinary NGAL (uNGAL) production in
AKI. The investi­gators determined that the
principal or exclusive source of uNGAL was
the thick ascending limb and the collecting
ducts of the nephron. In addition, in the prerenal azotemia model, serum creatinine level
rose, but NGAL-Luc2-mC was not activated
and no increase in uNGAL level occurred.
These data show that uNGAL is an appropriate kidney-injury-specific biomarker
despite not being exclusively expressed in
the kidney.
Another study involving NGAL displayed the range of methods by which AKI
biomarkers can assist clinicians caring for
patients with AKI. NGAL has been primarily thought of as a biomarker for the early
diagnosis of AKI,5 but utility of this biomarker and others may be much broader.
Using a large multicenter cohort of patients
with community-acquired pneumonia,
Srisawat and colleagues6 examined plasma
of patients on the first day they experienced
severe AKI (defined as RIFLE‑F by the Risk,
Injury, Failure, Loss and ESRD (RIFLE) criteria). In this study, recovery was defined
as being alive and neither requiring RRT
during hospitalization nor having a persistent RIFLE‑F classification at hospital
discharge. The investigators found that
elevated plasma NGAL (pNGAL) levels
were associated with renal non-recovery.
Although the absolute predictive value of
pNGAL alone was only fair (area under
the receiver operating characteristic curve
0.74), the reclassification of risk of not
■■ The source of urinary neutrophil-related
gelatinase lipocalin (NGAL) seems to be
nearly exclusively from the renal tubule
and does not seem to be released during
pre-renal azotemia in healthy animals4
■■ Plasma NGAL level helps predict which
patients with severe acute kidney injury
(AKI) will recover renal function6
■■ Urine output remains an important
‘biomarker’ of AKI, and predicts death
even in the absence of a rise in serum
creatinine level7
■■ Micro RNAs are a new class of AKI
biomarkers that may prove to be
important new tools in the diagnosis and
treatment of AKI8
recovering renal function was increased
by 17%. These data are most notable for
two reasons. First, because AKI can cause
CKD and ESRD, decisions regarding longterm care (for example, use of dialysis,
vascular access, and follow-up) are often
made in a piecemeal approach. If objective metrics coupled with clinical assessment can improve prognostic accuracy, a
better informed decision can be made for
survivors of AKI. Second, the association
between pNGAL and renal non-recovery
suggests that renal injury is ongoing, and
even if a patient is under­going dialysis,
thera­pies directed at mitiga­t ing ongoing
injury may have a role in AKI treatment.
...some existing biomarkers
such as NGAL will begin to shape
clinical practice...
Within the advancing field of AKI, the
oldest known biomarker is urine output. The
precise utility of oliguria has been controversial, and within the Acute Kidney Injury
Network (AKIN) staging system, there has
been discussion on the precise definition of
oliguria over time (for example, consecutive hours of oliguria versus average urine
output over a period of time). To further
the understanding, Macedo and colleagues7
utilized high-fidelity urimeters to compare
various definitions of oliguria in critically
ill patients. In their study, 317 patients
had speciali­zed high-accuracy urimeters
placed on their urinary catheters in order to
measure their urine output hourly. Serum
creatinine was measured every 12 h, and
definitions of AKI based on urine output
were compared with those based on serum
creatinine. Urine output was classified in
three ways: consecutive hours of oliguria
(oligo6-cons), an average amount of olig­
uria over a fixed block of time (<3 ml/kg in
the period from 6.00 am to 12.00 midday),
or any episode of oliguria over a set period of
time (<3 ml/kg for any 6 h period). Of these
variables, oligo6-cons was the most specific when compared with the AKIN stage I
serum creatinine standard (sensitivity 34%,
specificity 71%). The other two metrics, olig­
uria 6 h fixed (oligo6-fixed) and oliguria 6 h
floating (oligo6-float), were more sensitive,
but less specific (sensitivities 47% and 53%,
and specificities 62% and 54%, respectively).
Of the patients with oliguria for 6 h, 79% of
oligo6-cons advanced to AKIN stage II,
while 64% of oligo6-fixed and 52% of oligo6float advanced to AKIN stage II. Using the
urine output definition of AKI increased
the observed incidence of AKI to 52% as
compared with an AKI incidence of 24%
using serum creatinine definitions alone.
The mortal­ity rate of patients with AKI
defined by urine output alone was comparable to that of patients with AKI defined by
serum creati­nine alone (8.8% versus 10.4%).
Moreover, oliguria was a more sensitive
marker of AKI, and tended to occur earlier
than did change in serum creatinine level.
These data endorse current AKIN definitions that incorporate changes in both urine
output and serum creatinine level.
2011 was also notable for the development of whole new classes of bio­markers for
various diseases including AKI. Whereas
most biomarkers are small peptides,
microRNAs (miRs) are small ribonucleotides that regulate gene expression, and
have been shown to have disease-specific
roles in pre-clinical studies. Lorenzen and
colleagues8 demonstrated in 77 critically ill
patients with AKI and appropriate controls
that patients with AKI had downregulation
of miR‑320 and miR‑16 levels and upregula­
tion of miR‑210 levels. Additionally,
miR‑210 level predicted mortality in a
multivariable analysis that included severity of illness. miR‑210 is a plausible biomarker for AKI as it is involved in the
molecular response to stress. For example,
hypoxia-inducible factor causes induction
of miR‑210.9 miRNAs have potential advantages over plasma proteins as biomarkers.
They are very stable in plasma and might
serve as a noninvasive tool for assessing intra­cellular events. Because miRNAs
have significant gene regulatory capacity,
these molecules may offer an entire new
vantage point from which to understand the
pathophysiology of AKI.
JANUARY 2012 | S47
Also in 2011, investigators sought new
AKI biomarkers that are more directly
related to the injury itself. Munshi and
colleagues10 reported a study where potential biomarker candidates were chosen
from urinary excretion of injury-induced
mRNAs. The investigators posited that the
resulting proteins of these mRNAs might
be good AKI biomarkers and may offer
pathogenic information in addition to
diag­nostic information. Munshi et al. quantified the urinary excretion of the mRNAs
for one of these candi­date proteins, monocyte chemo­attractant protein‑1 (MCP‑1).
The investi­gators conducted multiple preclinical studies wherein various forms of
AKI were induced and MCP‑1 was compared with a known representative AKI
biomarker, NGAL. In the models of AKI,
MCP‑1 protein and mRNA levels increased
more than corresponding increases in
NGAL. Uremia, without kidney injury,
induced the NGAL gene, but not MCP‑1,
which suggests that MCP‑1 may be more
specific for AKI. These findings were
tested in a candidate cohort of Acute
Physiology and Chronic Health Evaluation
(APACHE)-II-matched critically ill
patients with and without AKI. In these
patients, MCP‑1 levels were significantly
higher in those with AKI.
In our view, these studies advanced
the field in 2011 and collectively provide
important information on the current state
of medi­cine pertaining to AKI as well as
suggesting future directions. We believe
that some existing biomarkers such as
NGAL will begin to shape clinical practice
while new biomarkers will continue to be
discovered. This ongoing process of new
discovery and reinvention of existing tools
(including those as primitive as urine flow)
will advance the field further and we will
eventually emerge with a set of tools that
will not only help us diagnose AKI, but will
also help us determine its cause, monitor its
course and predict response to therapy. We
eagerly await this bright future.
Department of Anesthesiology and Critical Care
Medicine, George Washington University
Hospital, 900 23rd Street, NW Room G‑105,
Washington DC 20037, USA (L. S. Chawla).
604 Scaife Hall, Department of Critical Care
Medicine, University of Pittsburgh,
3550 Terrace Street, Pittsburgh, PA 15261,
USA (J. A. Kellum).
Correspondence to: J. A. Kellum
[email protected]
Competing interests
L. S. Chawla declares associations with the following
companies: Abbott, Alere, Astute Medical.
S48 | JANUARY 2012
J. A. Kellum declares associations with the following
companies: Abbott, Alere, Astute Medical, Roche.
See the article online for full details of the
Hoste, E. A. et al. Epidemiology of acute kidney
injury. Contrib. Nephrol. 165, 1–8 (2010).
Chawla, L. S., Amdur, R. L., Amodeo, S.,
Kimmel, P. L. & Palant, C. E. The severity of
acute kidney injury predicts progression to
chronic kidney disease. Kidney Int. 79,
1361–1369 (2011).
Ishani, A. et al. The magnitude of acute serum
creatinine increase after cardiac surgery and
the risk of chronic kidney disease, progression
of kidney disease, and death. Arch. Intern. Med.
171, 226–233 (2011).
Paragas, N. et al. The Ngal reporter mouse
detects the response of the kidney to injury in
real time. Nat. Med. 17, 216–222 (2011).
Shemin, D. & Dworkin, L. D. Neutrophil
gelatinase-associated lipocalin (NGAL) as a
biomarker for early acute kidney injury. Crit.
Care Clin. 27, 379–389 (2011).
6. Srisawat, N. et al. Plasma neutrophil gelatinaseassociated lipocalin predicts recovery from
acute kidney injury following community-acquired
pneumonia. Kidney Int. 80, 545–552 (2011).
7. Macedo, E., Malhotra, R., Bouchard, J.,
Wynn, S. K. & Mehta, R. L. Oliguria is an early
predictor of higher mortality in critically ill
patients. Kidney Int. 80, 760–767 (2011).
8. Lorenzen, J. M. et al. Circulating miR‑210
predicts survival in critically ill patients with
acute kidney injury. Clin. J. Am. Soc. Nephrol. 6,
1540–1546 (2011).
9. Fasanaro, P. et al. MicroRNA‑210 modulates
endothelial cell response to hypoxia and
inhibits the receptor tyrosine kinase ligand
Ephrin‑A3. J. Biol. Chem. 283, 15878–15883
10. Munshi, R. et al. MCP‑1 gene activation marks
acute kidney injury. J. Am. Soc. Nephrol. 22,
165–175 (2011).
Progression, prediction,
populations and possibilities
Adeera Levin
In 2011, studies of chronic kidney disease (CKD) were published in
abundance. The articles selected here represent the growing appreciation
of the importance of CKD as a modifier of outcomes, breakthroughs in
understanding the pathobiology and genetics of specific conditions, and
clinical trials of treatment strategies that offer hope to patients with CKD.
Levin, A. Nat. Rev. Nephrol. 8, 70–72 (2012); published online 6 December 2011;
In 2011, almost a decade after the initial
publication that described a new system
of evaluation, classification, and definition of chronic kidney disease (CKD), this
condition has become of mainstream interest. Articles published in 2011 reflect an
increase in sophistication of analysis, collaboration and real attempts at understanding the impact of CKD on individuals and
health-care systems, the importance of predicting disease progression, and, perhaps
most crucially, the effect of specific therapies on outcomes such as cardio­vascular
disease. Selecting the most influential
papers of 2011 is subject to bias and perspectives of the reviewer, but the selection
here represents the spectrum of studies that
I believe will influence thinking and practice
for the next decade.
Although controversies regarding the
identification and definition of CKD in the
general population still exist, Peralta and
colleagues determined that the addition of
cystatin C to traditional measures of kidney
function (creatinine and albumin­u ria)
better identifies individuals with ‘true’ CKD
and those who are likely to progress to endstage renal disease (ESRD) than using a
one-marker or two-marker approach.1 Much
controversy surrounding CKD staging has
related to individuals with higher values of
estimated glomerular filtration rate (eGFR).
The study by Peralta et al., which used eGFR
derived from cystatin C in a general popula­
tion cohort within the REGARDS study,
confirms that this additional biomarker can
reliably identify those at high risk of adverse
events (death, cardiovascular disease, or
renal replacement therapy). This finding is
extremely important for the large group of
patients who have been identified as having
milder stages of CKD, but for whom outcomes are uncertain. Peralta et al. demonstrate that knowledge of cystatin C values,
used in an eGFR equation, will reclassify
individuals and distinguish important prognostic differences (a threefold increased risk
of death and fourfold increased risk of ESRD
in this study). This simple observation could
profoundly affect patients and care providers
with respect to care-plan modification.
Confirmation that this simple test helps
to appropriately reclassify these patients
is important. Further studies will facilitate
our understanding.
Variability in outcomes between ethnic
groups and within CKD populations with
similar diseases, has been another source
of confusion over the past decades. Two
important breakthroughs were made in
2011 with respect to genetic determinants
of kidney disease. Firstly, the importance
of the apolipoprotein L1 (APOL1) gene in
predicting any form of progressive CKD in
African Americans,2 will have a profound
impact on clinical care (by suggesting alternative treatment), familial donation strategies and our understanding of progressive
disease in this vulnerable population.
Secondly, in another study, the finding that
soluble urokinase-type plasminogen activator receptor (suPAR) is an important propagator of focal segmental glomerulosclerosis
(FSGS), given its ability to activate podocyte
β2 integrin, provides new insights into this
condition.3 In an elegant set of experiments,
Wei et al. demonstrated the importance of
elevated levels of suPAR in patients with
FSGS, and found that lowering of suPAR
levels with plasmapheresis led to remission
of disease.3 Using a series of mouse models,
Wei and collaborators clearly identified a
relationship between suPAR and podocyte
damage. Although the reason for the elevation in suPAR levels remains to be determined, the fact that not all patients with
FSGS have high levels of this protein further
confirms that FSGS is a response to injury
rather than a specific disease. Improving
diagnostics and identifying new therapies
based on a better understanding of the
pathobiology is becoming possible with
discoveries such as those described above.
Overall, 2011 was an
astounding year in terms
of publications focusing on
nondialysis CKD
Predicting progression of CKD and
outcomes of patients with CKD has been
the focus of many publications over the
past 2 years. Increasingly, investigators
have identified aspects of kidney disease
that portend poorer outcomes. A series of
papers have described the ability of simple
laboratory parameters to predict progression to ESRD over 5 years, 4 and highlighted that the rate of eGFR decline needs
■■ Consensus is increasing in the clinical and
scientific community that classification
of chronic kidney disease (CKD) needs
to evolve to a more complex staging
system that includes etiology, albuminuria,
estimates of glomerular filtration rate and
other available parameters1,4,7
■■ More sophisticated methods of
genotyping and phenotyping individuals
have led to an improved ability to predict
outcomes in patients with CKD2,3,8
■■ The SHARP trial determined that a
lipid-lowering strategy safely reduced
atherosclerotic events in patients with
CKD; these findings, which showed that
the benefits were proportional to the lipid
lowering achieved, irrespective of initial
lipid values, have huge implications for
clinical care10
to be considered in prediction equations,5
as well as the need to incorporate urine
albumin excretion and estimates of GFR
into risk prediction models6 in patients
with established CKD. The report of the
consensus conference arranged under the
KDIGO (Kidney Disease Improving Global
Outcomes) organizational banner helped to
consolidate a series of key conceptual issues
relating to prognosis and risk stratification.7
Another important paper was that by
Isakova et al., who demonstrated the importance of fibroblast growth factor 23 (FGF23)
in predicting adverse outcomes in patients
with CKD.8 Levels of FGF23 demonstrated
a stepwise increase in association with
mortal­ity and ESRD. Within the context
of our increasing knowledge of the unique
role of FGF23 in cardiac hypertrophy and
potential therapeutic interventions, this
study has refocused the nephrology community on the utility of newer biomarkers
in risk stratification and in understanding
the underlying biological processes involved
in disease progression.
The importance of well-conducted clinical trials that answer questions of key importance cannot be overstated. Two such papers
were published in 2011, one addressing a
new compound that might delay dialysis in
patients with CKD, and the other addressing
atherosclerotic events in patients with CKD.
A study examining the effect of bardoxolone
(an oral antioxidant, anti-inflammatory
modulator) in patients with diabetes and
CKD, demonstrated an improvement in
eGFR at 24 weeks and 52 weeks in the participants.9 This study is the first to potentially offer some hope to patients with CKD,
although issues related to mechanism of
action of the medication, long-term effects
of increasing eGFR (that is, potential for
accelerated progression), and utility in all
clinical circumstances were not addressed
in this phase II study. However, these findings, which show that therapeutic agents
that might improve kidney function exist,
are a source of optimism for the millions
of patients with CKD.
The SHARP study, an international 5‑year
randomized, placebo-controlled trial of
>9,000 patients with CKD, found that active
lipid-lowering treatment (simva­statin 20 mg
and ezetimibe 10 mg) was effective in reducing atherosclerotic events in patients both
on and off dialysis.10 This study is seminal
in that it is the largest trial ever conducted
in CKD, and enables clini­cians to appreciate the benefit and very low risk profile of
a fixed dose of lipid-lowering medication
for this population. Although the study
was not powered to examine mortality, the
results did demon­strate a 17% reduction in
major atherosclerotic events. Publication of
this study in the Lancet further validates the
importance of these findings and the impact
it is likely to have on general care. Moreover,
the consistent finding that for any lowering
of LDL-cholesterol levels, there was a benefit
to patients with CKD similar to that found
in the general population, adds credence to
the hypothesis that atherosclerotic events
can be ameliorated by targeted therapy in
a wide range of patients. The SHARP study
also reminds us that morbidity and mortality in CKD is more complex than simply
being linked to atherosclerotic events, and
thus highlights that there is more research
to be done.
Overall, 2011 was an astounding year
in terms of publications focusing on non­
dialysis CKD. The majority of these papers
are in general medical journals,1,3,4,8–10 which
is further testimony to the recognition of
the importance of this area of medicine.
The overall message from these studies
is that CKD is important, identi­f iable by
simple means, is variable in its expression
and outcomes, and both progressive disease
and associated comorbidi­ties are amenable
to thera­peutic interventions, which is reassuring for patients and their caregivers.
Ongoing genomic and proteomic studies will
further unravel the mystery of CKD progression and outcomes. The breadth and depth
of nephrology insights and understandings
continues to grow, as is evidenced in this
landmark year. With such growth, we would
anticipate an exciting future for both science
and clinical care in the years to come.
JANUARY 2012 | S49
Division of Nephrology, Department of
Medicine, St Paul’s Hospital, University of
British Columbia, 1081 Burrard Street, Room
6010-A, Vancouver, BC V6Z 1Y6, Canada.
[email protected]
Competing interests
A. Levin declares an association with the following
company: Merck. See the article online for full
details of the relationship.
Peralta, C. A. et al. Detection of chronic kidney
disease with creatinine, cystatin C, and urine
albumin‑to‑creatinine ratio and association
with progression to end-stage renal disease
and mortality. JAMA 305, 1545–1552 (2011).
Friedman, D. J., Kozlitina, J., Genovese, G.,
Jog, P. & Pollak, M.R. Population-based risk
assessment of APOL1 on renal disease. J. Am.
Soc. Nephrol. 22, 2098–2105 (2011).
Wei, C. et al. Circulating urokinase receptor as
a cause of focal segmental
glomerulosclerosis. Nat. Med. 17, 952–960
Tangri, N. et al. A predictive model for
progression of chronic kidney disease to
kidney failure. JAMA 305, 1553–1557 (2011).
Perkins, R. M. et al. GFR decline and mortality
risk among patients with chronic kidney
disease. Clin. J. Am. Soc. Nephrol. 6,
1879–1886 (2011).
6. Astor, B. C. et al. and the Chronic Kidney
Disease Prognosis Consortium. Lower
estimated glomerular filtration rate and higher
albuminuria are associated with mortality and
end-stage renal disease. A collaborative metaanalysis of kidney disease population cohorts.
Kidney Int. 79, 1331–1340 (2011).
7. Levey, A. S. et al. The definition, classification,
and prognosis of chronic kidney disease:
a KDIGO Controversies Conference report.
Kidney Int. 80, 17–28 (2011).
8. Isakova, T. et al. for the Chronic Renal
Insufficiency Cohort (CRIC) Study Group.
Fibroblast growth factor 23 and risks of
mortality and end-stage renal disease in
patients with chronic kidney disease. JAMA
305, 2432–2439 (2011).
9. Pergola, P. E. et al. for the BEAM Study
Investigators. Bardoxolone methyl and kidney
function in CKD with type 2 diabetes. N. Engl. J.
Med. 365, 327–336 (2011).
10. Baigent, C. et al. on behalf of the SHARP
Investigators. The effects of lowering LDL
cholesterol with simvastatin plus ezetimibe in
patients with chronic kidney disease (Study of
Heart and Renal Protection): a randomised
placebo-controlled trial. Lancet 377,
2181–2192 (2011).
Can cardiovascular risk in dialysis
patients be decreased?
Peter Stenvinkel and Peter Bárány
More than 1.4 million patients are on renal replacement therapy
worldwide. Mortality in patients with end-stage renal disease (ESRD) is
as high as that seen in some types of metastatic cancer, and premature
cardiovascular disease is the major killer in ESRD. Several publications in
2011 addressed how interventions can modify cardiovascular risk factors
and improve outcomes.
Stenvinkel, P. & Bárány, P. Nat. Rev. Nephrol. 8, 72–74 (2012); published online 13 December 2011;
Although nephrologists have witnessed
many technical advances in dialysis therapy
over the past few decades, these improvements have translated into only minor
reductions in mortality rate. Another major
concern in nephrology is that several randomized controlled trials (RCTs) have
not been able to demonstrate a significant
effect of targeting various traditional and
nontraditional risk factors on outcomes in
this high-risk patient group. Inflammation
seems to be a major driving force for the
uremic phenotype and the majority of
dialy­sis patients have signs of persistent lowgrade inflammation with repetitive bursts
of increased inflammatory activity. Because
increased levels of circulating inflammatory biomarkers, such as C‑reactive protein
S50 | JANUARY 2012
(CRP), pentraxin 3, and IL‑6, consistently
predict poor outcome in dialysis patients,
inflammation seems to be a logical therapeutic target for putative anti-inflammatory
nutritional and pharmaco­logical interventions in this high-risk patient group. Recent
observational studies suggest that the
presence of persistent inflammation via a
catalytic effect magnifies the risk of poor
outcome via mechanisms related to selfenhancement of the inflammatory cascade
and exacerbation of wasting and vascular calcification processes.1 Results from a
2011 study in a cohort of 225 hemodialysis
patients followed over a period of 13 years
support the catalyst hypothesis: inflamma­
tion was shown to amplify the risk of
death and cardio­vascular events associated
with high asymmetric dimethylarginine
(ADMA) levels.2 Given the multi­factorial
origin of uremic inflammation, treatment
is complex and the nephrologist needs to
consider both factors related to the dialysis
procedure (such as dialysis catheters and
quality of dialysate) and factors unrelated
to the dialysis procedure per se (such as
infectious complica­tions and volume overload). Nevertheless, following the exclusion of such factors, no apparent reason for
persistent low-grade inflammation would
be found in a consider­able proportion of
patients on dialysis. In a small RCT of 22
patients published in 2011, Hung et al. 3
demonstrated that 4 weeks of treatment
with a recombinant human IL‑1 receptor
antagonist (anakinra) significantly reduced
mean CRP level (by 53%) and mean IL‑6
level (by 40%), while mean prealbumin
level increased by 23%. Moreover, the anti-­
cytokine treatment was well tolerated and
safe. Although the small patient number and
the short treatment period limit the value of
this study, it still provides important information to the nephrology community. It is
the first study showing that targeted anticytokine treatment decreases inflamma­
tion parameters in this patient group. Thus,
long-term studies should be initiated to
study the impact of anti-­inflammatory treatment strategies on cardiovascular outcomes,
quality of life, nutritional status and death in
patients on dialysis.
Inflammation seems to be a
major driving force for the uremic
The Study of Heart and Renal Protection
(SHARP), also published in 2011, is the
largest randomized study ever conducted
in nephrology and included 9,270 patients
random­ized to receive either 20 mg simva­
statin plus 10 mg ezetimibe daily or
placebo. 4 One-third of the participants
were dialysis dependent (2,527 on hemodialysis and 496 on peritoneal dialysis)
and median follow-up was 4.9 years. In the
whole cohort, treatment with simvastatin
plus ezetimibe resulted in a 17% reduction in the primary end point of the risk of
major athero­sclerotic events (relative risk
0.83; 95% CI 0.74–0.94). Significant reductions in revascularization procedures and
non­hemorrhagic strokes contributed most
to the effect on the primary end point.
Importantly, treatment with simvastatin
plus ezetimibe was not associ­ated with
any survival benefit or a renal protective
effect. Although the effect on the primary
end point was similar in all subgroups,
the risk reduction was less prominent in
patients on dialysis (relative risk 0.90; 95%
CI 0.75–1.08) than in nondialysis patients
(relative risk 0.78; 95% CI 0.67–0.91). Thus,
the modest effect on cardiovascular risk in
patients on dialysis is of a similar magnitude
to that reported in previous studies of statins
(nonsignificant risk reductions of 8% in the
4D study and 4% in AURORA). Persistent
inflammation and comorbidities may influence the effects of lipid-­lowering treatment
in patients on dialysis. Although the precise
role of these factors and the mechanism/s
through which they interact with statin
treatment are not yet evident, it can be speculated that if persistent inflammation serves
as a catalyst and magnifies risk,1 the effect
of long-term statin treatment may differ
in patients on dialysis with and without
inflammation; these groups should be analyzed separately. Comorbidities, such as diabetes, may also influence the cardiovascular
effects of statin treatment in this patient
group. To determine whether rosuvastatin
might reduce the risk of cardiac events in
731 hemo­dialysis patients with diabetes, a
post hoc analysis of the AURORA cohort
was performed.5 Rosuvastatin treatment
significantly reduced the rates of cardiac
events by 32% among patients with diabetes. However, it is a concern that although
there was no difference in overall stroke
incidence between the rosuvastatin and
the placebo groups, rosuvastatin-treated
patients with diabetes had a higher risk of
hemorrhagic strokes than the placebo group
(relative risk 5.21; 95% CI 1.17–23.27).
The high prevalence of atrial fibrillation
is another major clinical problem in dialysis
units. In a recent study based on 2.5 million
observations of hemodialysis patients,
Winkelmayer et al.6 showed that the prevalence of atrial fibrillation had increased
threefold from 1992 (3.5%) to 2006 (10.7%).
They also showed that mortality was twice
as high among hemodialysis patients with
atrial fibrillation compared to those without
atrial fibrillation. In 2011, the same group
also reported data from 2,313 hemodialysis
patients with new-onset atrial fibrillation.
Whereas patients treated with warfarin had
double the risk of hemorrhagic stroke compared with nonusers, the risk of ischemic
stroke did not differ between users and nonusers of warfarin.7 As warfarin treatment
does not reduce the risk of ischemic stroke
and, in addition to increasing the risk of
■■ Inflammation interacts with other
cardiovascular risk factors in the
uremic milieu, such as asymmetric
dimethylarginine, and increases the risk
of poor outcomes2
■■ Short-term treatment with a recombinant
human IL‑1 receptor antagonist reduces
inflammatory biomarkers in patients on
■■ Lipid-lowering treatment is associated
with beneficial effects on major
atherosclerotic events in the entire
chronic kidney disease population;
the effect is less pronounced in the
subpopulation of dialysis patients4
■■ The prevalence of atrial fibrillation in
North American hemodialysis patients
increased threefold between 1992 and
■■ Whereas warfarin treatment does not
reduce the risk of ischemic stroke, it
increases the risk of hemorrhagic stroke
in older hemodialysis patients with atrial
hemorrhagic stroke, may have a role in both
the progression of vascular calcification and
in the life-threatening condition calciphylaxis, the risks of war­farin treatment seem
to outweigh its potential bene­ficial effects
in dialy­sis patients with atrial fibrilla­tion.
Thus, until an adequately powered RCT
settles the important question as to whether
warfarin reduces the overall risk of stroke
and improves survival it seems advisable not
to prescribe vitamin K antagonists, such as
warfarin, to patients on dialysis with atrial
fibrillation. This finding should also prompt
the initiation of studies to evaluate the safety
and efficacy of novel anti­coagulant treatment strategies in patients with advanced
chronic kidney disease, such as drugs targeting factor IIa (dabigatran) and factor Xa
(rivaroxaban, apixaban and edoxaban)
in the coagulation cascade. Indeed, in a
recent RCT conducted in 18,201 patients
with atrial fibrillation (but without kidney
disease), apixaban was superior to warfarin
in preventing stroke or systemic embolism, caused less bleeding, and resulted in
lower mortality.8
The effects of the intensity of dialysis on
morbidity and mortality was first studied
in the 1981 National Cooperative Dialysis
study, which suggested that a longer hemodialysis duration aiming at a predialysis
blood urea nitrogen level of 20 mmol/l
resulted in less morbidity than a shorter
hemodialysis duration. The results of the
2002 HEMO study 9 showed that aiming
at a higher dialysis dose did not result in
signifi­cant benefits on mortality or morbidity (over standard doses) in patients
undergoing thrice-weekly hemo­dialysis. In
2010, the randomized study by the Frequent
Hemodialysis Network (FHN) studied the
effect of frequent hemo­dialysis versus conventional hemodialysis (six times weekly
versus three times weekly) in 245 patients
followed up for 1 year.10 Frequent hemo­
dialysis was associated with significant
beneficial effects on the risk of an increase
in left ventricular muscle mass and the
risk of worse physical health scores as well
as improved control of hypertension and
hyperphosphatemia. The drawback was
a 71% increase in the risk of requiring
interventions related to vascular access.
Although frequent hemodialysis is a promising modality, the cost–benefit ratio over
longer follow-up periods than in the FHN
study remains to be evaluated.
In conclusion, several studies published in 2011 have provided the nephrology community with new small pieces of
knowledge to add to the complex puzzle
of the increased risk of premature cardio­
vascular death in patients on dialysis.
Based on current knowledge, we conclude
that inflammatory biomarkers and an
assessment of atrial fibrillation should be
included in the risk factor profile monitored by nephrologists. Whereas statin
therapy seems to have a modest, but signifi­
cant, beneficial effect on cardiac events,
vitamin K antagonist treatment cannot be
advocated in dialysis patients with atrial
fibrillation until an RCT has proved its
safety and efficacy.
Division of Renal Medicine, Department of
Clinical Science, Intervention and Technology,
Karolinska University Hospital at Huddinge,
Karolinska Institutet, SE 141 86 Stockholm,
Sweden (P. Stenvinkel, P. Bárány).
Correspondence to: P. Stenvinkel
[email protected]
Competing interests
P. Stenvinkel declares an association with the
following company: Gambro. See the article online
for full details of the relationship. P. Bárány declares
no competing interests.
Carrero, J. J. & Stenvinkel, P. Of persistent
inflammation as a catalyst for other risk factors
in chronic kidney disease. a hypothesis
proposal. Clin. J. Am. Soc. Nephrol. 4, S49–S55
Tripepi, G. et al. Inflammation and asymmetric
dimethylarginine for predicting death and
cardiovascular events in ESRD patients. Clin. J.
Am. Soc. Nephrol. 7, 1714–1721 (2011).
Hung, A. M., Ellis, C. D., Shintani, A., Booker, C.
& Ikizler, T. A. IL‑1β receptor antagonist reduces
JANUARY 2012 | S51
inflammation in hemodialysis patients. J. Am.
Soc. Nephrol. 22, 437–442 (2011).
Baigent, C. et al. The effects of lowering LDL
cholesterol with simvastatin plus ezetimibe in
patients with chronic kidney disease (Study of
Heart and Renal Protection): a randomised
placebo-controlled trial. Lancet 377,
2181–2192 (2011).
Holdaas, H. et al. Rosuvastatin in diabetic
hemodialysis patients. J. Am. Soc. Nephrol. 22,
1335–1341 (2011).
Winkelmayer, W. C., Patrick, A. R., Liu, J.,
Brookhart, M. A. & Setoguchi, S. The increasing
prevalence of atrial fibrillation among
hemodialysis patients. J. Am. Soc. Nephrol. 22,
349–357 (2011).
Winkelmayer, W. C., Liu, J., Setoguchi, S. &
Choudhry, N. K. Effectiveness and safety of
warfarin initiation in older hemodialysis
patients with incident atrial fibrillation. Clin. J.
Am. Soc. Nephrol.
8. Granger, C. B. et al. Apixaban versus warfarin in
patients with atrial fibrillation. N. Engl. J. Med.
365, 981–992 (2011).
9. Eknoyan, G. et al. Effect of dialysis dose and
membrane flux in maintenance hemodialysis.
N. Engl. J. Med. 347, 2010–2019 (2002).
10. Chertow, G. M. et al. In-center hemodialysis
six times per week versus three times per
week. N. Engl. J. Med. 363, 2287–2300
New agents, new ideas and new
Titte R. Srinivas and Bruce Kaplan
The past year was marked by several excellent studies that represent
important therapeutic advances in kidney transplantation or that further
our understanding of the genetic basis of chronic allograft dysfunction,
clinical tolerance and outcomes of kidney transplantation.
Srinivas, T. R. & Kaplan, B. Nat. Rev. Nephrol. 8, 74–75 (2012); published online 20 December 2011;
The year 2011 was marked by the addition of an important alternative to calcineurin inhibitors for immuno­suppression
in kidney transplantation. Calcineurin
inhibitor-based combina­tions are the most
frequently utilized immuno­s uppressive
regimens in kidney transplantation, but
they carry a metabolic burden and may
be associated with both short-term and
possibly progressive diminution in graft
function. Belatacept is a costimulation
blocker that binds to CD80 and CD86
on antigen-­presenting cells and provides
Key advances
■■ 3-year follow-up of BENEFIT demonstrates
that belatacept provides superior renal
function outcomes to ciclosporin3
■■ Sensitized patients treated with a
desensitization regimen have better
survival than patients on dialysis
or sensitized patients who undergo
compatible transplantation4
■■ Discovery of APOL1 variants in African
Americans points to a biological
influence rather than race underlying
transplantation outcomes6
■■ Increased cold ischemia times of
extended criteria donor kidneys are
associated with delayed graft function,
but do not affect graft survival10
S52 | JANUARY 2012
effective immuno­suppression while being
devoid of renal and nonrenal metabolic
adverse effects. 1 A phase III study of
belatacept (BENEFIT) published in 2010
reported superior renal function and an
improved cardiovascular and metabolic
risk profile compared with a ciclosporinbased regimen.2 In this trial, 686 de novo
kidney transplant recipients (of living and
standard criteria deceased donor kidneys)
were randomly assigned to more-intensive
or less-intensive belatacept regimens or to
ciclosporin. All patients received basiliximab induction therapy, mycophenolate
mofetil and corticosteroids. Equivalent
rates of graft and patient survival were
noted despite an increased frequency and
severity of early rejection episodes and
increased frequency of post-transplant
lympho­proliferative disorder associated
with belatacept.2
Vincenti et al. have now reported on the
efficacy and safety of belatacept at 3 years
post-transplantation in the BENEFIT
study.3 Graft survival was equivalent across
all regimens. At year 3 post-­transplantation,
the mean estimated glomerular filtration
rate (eGFR) was 21 ml/min/1.73 m2 higher
in the belatacept groups than in the ciclosporin group. More importantly, from
month 3 through to month 36, the slope
of eGFR in the belatacept groups averaged 1.1 ml/min/1.73 m 2 per year versus
–2.0 ml/min/1.73 m 2 per year with ciclosporin. Neither a significant increase in
acute rejection episodes nor new cases of
post-transplant lymphoproliferative dis­
order after 18 months post-transplantation
were observed.3 Taken together, these findings are indeed gratifying and represent
the emergence of a non-nephrotoxic and
efficacious immunosuppressive regimen
in renal transplantation. This study also
demon­strated a disconnect between rejection rates and graft survival and function,
which either reflects short follow-up or
reasons that prompt reconsideration of
the relationship between classic T‑cellmediated rejection and outcomes. Although
initial results from BENEFIT are promising,
their generalizability to clinical practice at
large will depend on a paradigm shift from
oral maintenance therapy to infusion-based
therapy. As such, whether the promising
initial results are sustainable on logistical
and economic grounds needs further study.
Sensitized transplant candidates have
prolonged waiting times, a reduced
transplanta­tion rate and an increased rate
of death compared with nonsensitized
candidates. Although numerous centers
report successful utilization of desensitization regimens in renal transplantation,
they are plagued by the absence of suitable
control groups. Montgomery et al. compared outcomes in 211 HLA-sensitized live
donor kidney transplant recipients treated
with low-dose intra­v enous immuno­
globulin and plasma­pheresis with that of
two matched control groups drawn from
the United Network for Organ Sharing
kidney transplant waiting list who either
remained on dialysis (dialysis-­only group)
or who were dialyzed or under went
HLA-compatible renal transplantation
(dialysis-or-­t ransplantation group).4 In
the desensitized group, estimates of patient
survival were 90.6%, 85.7% and 80.6% at 1,
3 and 5 years, respectively, versus 91.1%,
67.2% and 51.5% in the dialysis-only
group and 93.1%, 77.0% and 65.6% in the
dialysis-or-transplantation group (overall
P <0.001; differences were not statistically
significant after 1 year).4 Although these
results are encouraging, several unanswered
questions remain. For instance, this report
does not address actual long-term graft
survival or antibody-­mediated rejection in
the popula­tion studied. Furthermore, the
costs of a desensitization regimen followed
by transplantation are considerable and
have not been compared systematically to
the cost of remaining on dialysis. In addition, this study compared patient survival
with living donor transplantation following desensitiza­tion with those accrued on
the waitlist. Whether these encouraging
observa­t ions could extend to deceased
donor transplantation after successful
desensitization remains to be seen.
African Americans are at increased risk
of end-stage renal disease compared with
white individuals. Recent studies suggest an
association between variants of the apolipoprotein L1 gene (APOL1) and nondiabetic
nephropathy among African Americans.5
In a single-center study, kidneys from
African American deceased donors with
two APOL1 risk variants were more likely to
fail (HR 3.48, P = 0.008) than kidneys from
individuals with one or no APOL1 risk variants.6 These associations outweighed the
risk of graft loss associated with African
American ancestry. This study highlights
that transplant outcomes are more likely to
be based on biology than on arbitrary and
often culturally based definitions of race.
Whether or not donor genotype or donor–
recipient genotype interactions influence
transplant outcomes through nonimmunological mechanisms demands confirmation.
More importantly, the application of these
associations to improve matching of donors
and recipients in organ allocation schemes
merits further study.
Successful transplantation in the clinical
arena is inseparable from the pharmacopeia
that facilitates successful transplantation.
Immunosuppressive drugs are associated
with well-known immunological and nonimmunological adverse effects. However, it
is increasingly apparent that a small number
of recipients of renal and liver allografts
continue to have excellent graft function
many years (sometimes decades) after
stopping immunosuppression. Such individuals are described as being operationally
tolerant. A unique B-cell expression signature associated with increased numbers of
naive and transitional B cells in peripheral
blood has been described in operationally tolerant kidney transplant recipients.7
A recent study comparing character­istics
of liver and kidney transplant recipients
who are operationally tolerant showed
KEY ADVANCES IN MEDICINE that the gene-expression signatures of
peripheral blood mononuclear cells in
these individuals was very different, with
no significant detectable overlap between
their immuno­phenotypic profiles.8 By contrast, tolerant and non­tolerant liver recipients had similar peripheral blood B-cell
phenotypic markers.8
...transplant outcomes are
more likely to be based on biology
than on arbitrary ... definitions of
It is important to note that the studies
discussed above represent associations and
do not imply causal relationships. Patients
who maintain excellent graft survival
despite discontinuing immuno­suppression
might differ considerably from those who
reject under similar circumstances; findings drawn from operationally tolerant
individuals could therefore be affected
by survivorship bias. Whether genomic
and immunophenotypic markers associated with prolonged discontinuation of
immuno­suppression and excellent graft
survival could be identified prospectively
and enable programmed discontinuation
of immunosuppression in selected recipients earlier in the course of transplantation
deserves further systematic study.
Extended criteria donor (ECD) kidneys
permit successful transplantation especially among transplant candidates with
poor expected waitlist survival, such as the
elderly and those with diabetes.9 However,
delays in the placement of ECD kidneys are
associated with increased cold ischemia
times (CITs) and a high rate of discard of
such organs. A recent study using data from
the Scientific Registry of Renal Transplant
Recipients examined the effect of increasing CITs on graft survival among recipients of paired kidneys (kidneys from the
same deceased donor transplanted to
two different recipients where one donor
had delayed graft function and the other
did not). 10 Longer CITs were associated
with increasing incidence of delayed graft
function. However, graft survival did not
signifi­cantly differ across the range of CITs
studied. Although this study is limited by
its retrospective nature and is prone to
selection bias, it provides some impetus to
reconsider discard of ECD kidneys with
higher CITs and optimize ECD utilization
and acceptance.
The articles summarized above provide
food for thought and prospects for further
inquiry in coming years. We look forward
to fruitful developments in these and related
areas in clinical transplantation.
Department of Nephrology and Hypertension,
Glickman Urological and Kidney Institute, 9500
Euclid Avenue, Cleveland, OH 44195, USA
(T. R. Srinivas). Division of Nephrology,
University of Arizona School of Medicine,
1501 North Campbell Avenue, Tucson,
AZ 85724, USA (B. Kaplan).
Correspondence to: B. Kaplan
[email protected]
Competing interests
The authors declare no competing interests.
Larsen, C. P. et al. Rational development of
LEA29Y (belatacept), a high-affinity variant of
CTLA4-Ig with potent immunosuppressive
properties. Am. J. Transplant. 5, 443–453
2. Vincenti, F. et al. A phase III study of belataceptbased immunosuppression regimens versus
cyclosporine in renal transplant recipients
(BENEFIT study). Am. J. Transplant. 10,
535–546 (2010).
3. Vincenti, F. et al. Three-year outcomes from
BENEFIT, a randomized, active-controlled,
parallel-group study in adult kidney transplant
recipients. Am. J. Transplant.
4. Montgomery, R. A. et al. Desensitization in HLAincompatible kidney recipients and survival.
N. Engl. J. Med. 365, 318–326 (2011).
5. Friedman, D. J., Kozlitina, J., Genovese, G.,
Jog, P. & Pollak, M. R. Population-based risk
assessment of APOL1 on renal disease. J. Am.
Soc. Nephrol. 22, 2098–2105 (2011).
6. Reeves-Daniel, A. M. et al. The APOL1 gene and
allograft survival after kidney transplantation.
Am. J. Transplant. 11, 1025–1030 (2011).
7. Newell, K. A. et al. for the Immune Tolerance
Network ST507 Study Group. Identification of a
B cell signature associated with renal
transplant tolerance in humans. J. Clin. Invest.
120, 1836–1847 (2010).
8. Lozano, J. J. et al. Comparison of
transcriptional and blood cell-phenotypic
markers between operationally tolerant liver
and kidney recipients. Am. J. Transplant. 11,
1916–1926 (2011).
9. Merion, R. M. et al. Deceased-donor
characteristics and the survival benefit of
kidney transplantation. JAMA 294, 2726–2733
10. Kayler, L. K., Srinivas, T. R. & Schold, J. D.
Influence of CIT-induced DGF on kidney
transplant outcomes. Am. J. Transplant. 11,
2657–2664 (2011).
JANUARY 2012 | S53
Don’t Miss Out!
e-ALERTS give you:
The latest issue’s table of contents
Full access to abstracts
Listings from Naturejobs
Related Nature Publishing Group product information
View the entire series here:
Major advances across the spectrum
of stroke care
Lee H. Schwamm
Several pivotal clinical trials that could have a major impact on the care of patients with stroke were published
in 2011. The studies cover a wide range of stroke-care aspects, including stroke prevention, imaging to select
patients for thrombolysis, therapies for stroke recovery, and stroke registries to improve care quality.
Schwamm, L. H. Nat. Rev. Neurol. 8, 63–64 (2012); published online 10 January 2012; doi:10.1038/nrneurol.2011.225
The current framework for analyzing the
complete spectrum of stroke care is conceptualized in the ‘stroke systems of care’ model1
(Figure 1), as delineated in a policy statement of the American Stroke Association in
2005. The model specifies seven domains of
care—primary prevention; community education; notification and response of emergency
medical services; acute stroke treatment; subacute stroke treatment and secon­dary prevention; rehabili­tation; and con­tinuous quality
improvement ­activities—and emphasizes the
links between them. In 2011, five domains of
the stroke systems of care model have witnessed major breakthroughs that have the
potential to dramati­cally influence the delivery
of stroke care now and in the future, thereby
further eroding the therapeutic nihilism that
still surrounds much of stroke treatment.
Nonvalvular atrial fibrillation is a major
preventable cause of first ever and recurrent
stroke. Warfarin has been the mainstay of
therapy for almost half a century, but it has
a narrow therapeutic range that is difficult to
maintain in many patients, requires frequent
monitoring, and is highly susceptible to drug
and food interactions. The ARISTOTLE trial2
compared the oral direct factor Xa inhibitor
apixaban with warfarin for efficacy in stroke
prevention in over 1,800 patients with atrial
fibrillation and an elevated stroke risk score.
In both groups, just under 15% of patients
had a prior myocardial infarction, and 20%
had a prior stroke, transient ischemic attack
or systemic embolism. The results showed
that apixaban is superior to warfarin in
reducing the risk of stroke or systemic embolism (incidence rates 1.27% versus 1.60% per
year; HR 0.79; P <0.001 for non­inferiority;
P = 0.01 for superiority). For every 1,000
patients treated for 1.8 years, use of apixaban
KEY ADVANCES IN MEDICINE instead of warfarin prevented a stroke in six
patients, major bleeding in 15 patients, and
death in eight patients. The superior efficacy
of apixaban for stroke prevention manifested
early during treatment, seeming to reach a
maximum after about 12 months, whereas
the reductions in major bleeding associated with this drug were evident throughout
the 30 month trial period. Two recent trials
identified alternative drugs that are as effective as warfarin in stroke prevention: dabigatran,3 an oral direct thrombin inhibitor, and
rivaroxaban,4 another factor Xa inhibitor.
These new drugs represent major advances
for stroke prevention, but they offer convenience at a higher cost, and no rapid reversal
agents exist to manage acute, life-threatening
bleeding resulting from these agents. Whether
the benefits of these new drugs are maintained in the clinic, and whether the reduced
medical contact for patients receiving oral
anticoagulation has deleterious effects,
remains to be seen.
Intracranial atherosclerotic stenosis is a
major risk factor for stroke. This risk was
quantified in the WASID trial,5 with approxi­
mately 20% of patients who received war­farin
or aspirin experiencing a stroke in the affected
brain area over 1.8 years. The increased mortality associated with war­farin, and the lack of
any difference between warfarin and aspirin
in stroke prevention, justified evaluation of a
more aggressive intervention than warfarin
or aspirin alone. In the SAMMPRIS trial,6
patients with sympto­matic intra­cranial artery
stenosis of at least 70% received inten­sive
medical therapy (dual antiplatelet therapy for
90 days, lipid-lowering drugs, hyper­tension
control, and lifestyle modification) alone or in
combination with intracranial stenting. The
trial was halted early owing to un­expectedly
Continuous quality
EMS response
Acute treatment
Figure 1 | The ‘stroke systems of care’
model defines a spectrum of seven domains
that cover all aspects of stroke care. 2011 saw
key advances in primary prevention, acute
treatment, secondary prevention, rehabilitation,
and continuous quality improvement activities.
Abbreviation: EMS, emergency medical
services. MRI scan adapted from Asdaghi, N. &
Coutts, S. B. Nat. Rev. Neurol. 7, 6–7 (2011).
high rates of stroke or death associated with
stenting. The SAMMPRIS trial design was
based on estimates of adverse-event rates in
open-label stent registries, which were probably too low. Importantly, the increased rate
of adverse events in the stenting group was
not associated with operator inexperience,
and seems, therefore, to be a patient-related
or device effect. Of equal importance, the
SAMMPRIS trial6 highlights the fact that historical controls or placebo arms of prior trials
may not reflect the impact of modern intensive risk-­reduction strategies on l­owering
rates of stroke.
JANUARY 2012 | S55
Key advances
■■ Apixaban is superior to warfarin in
reducing the risk of stroke or systemic
embolism, and is safer in reducing the
risk of major bleeding or death2
■■ Intracranial stenting produces higher rates
of stroke and death than an intensive
medical management strategy for severe
symptomatic atherosclerotic stenosis6
■■ A standardized computer algorithm for
post-processing of MRI scans may be
superior to individual visual estimation for
selecting patients for thrombolytic therapy8
■■ A combination of fluoxetine and
physical therapy could produce greater
improvements in limb motor function
after stroke than physical therapy alone9
■■ Hospitals are rapidly adopting stroke
quality improvement registries, which
could reduce mortality, cost and
complications of stroke by increasing the
use of evidence-based therapies10
When prevention fails, acute stroke
manage­ment begins. Early reperfusion of
ische­mic tissue is the primary goal, which is
reflec­ted in the mechanism of action of the
only FDA-approved therapy for improving
stroke outcomes: tissue plasminogen activator. The proportion of patients who are
eligible for thrombolysis, however, remains
disappointingly low, owing largely to the
narrow time window for this treatment
and exclusion of patients with mild stroke.
Considerable efforts have been made to
develop imaging biomarkers of the acute
ischemic ‘penumbra’—brain tissue surrounding the lesion core that is vulnerable
but salvageable with early reperfusion—in
an attempt to expand the number of patients
who can receive reperfusion therapy.
Supporters of MRI-based patient selection for thrombolysis were dealt a blow
3 years ago when the DIAS-2 trial7 failed
to meet its primary end point, as it used a
widely accepted definition of the penum­bra
based on a mismatch between the volumes of
ische­mic brain tissue measured on diffusion
versus perfusion MRI. The MRI-mismatch
approach to patient selection could be
improved by the use of a standard­ized computer algorithm for post-processing of MRI
scans, as suggested by a recent pooled analysis8 of the EPITHET and DEFUSE trials.
The study emphasizes the importance of
human error and inter-operator variability
when visual inspection, rather than automated techniques, is used to determine suitability for reperfusion treatment. Ongoing
trials are evaluating alternative strategies of
MRI-based patient selection for intravenous
S56 | JANUARY 2012
reperfusion (MR WITNESS and EXTEND)
or intra-arterial reperfusion (MR RESCUE).
Patients with acute ischemic stroke are
almost always left with a degree of dis­ability
that requires rehabilitative treatment. Early
studies suggested that constraint therapy
(immobilization of the ‘good’ arm to encourage use of the affected arm) and pharmacological stimulants could modulate brain
plasticity after stroke and reduce the residual neurological deficit and subsequent dis­
ability. However, large-scale, well-designed
randomized controlled trials to support these
initial findings have not been performed.
Important evidence for a beneficial
effect of the antidepressant fluoxetine in
stroke recovery was provided in 2011 by
the FLAME trial.9 In this phase II study, all
patients received physiotherapy, and those
who were randomly assigned to the fluoxe­
tine group showed significantly greater
improvement in motor function of the
affected arm and leg at 90 days poststroke
than did those assigned to the placebo
group. Interestingly, differences between
the study groups were still signifi­cant after
adjustment for depression or thrombolytic use, sug­gesting that fluoxetine could
be beneficial across many subgroups. The
findings, which need to be confirmed in
larger trials, hold promise for thousands of
stroke survivors who may benefit from this
and other re­covery strategies to facilitate
brain plasticity.
With mounting pressure on funding and
resources in the current economic down­
turn, efficient health-care delivery is becoming increasingly important. The Ameri­can
Heart Association’s ‘Get with The Guide­
lines®-Stroke’ (GWTG-Stroke) quality
improve­ment registry tracks characteristics,
performance measures, and in-hospital outcomes in nearly 2 million patients admitted
to almost 2,000 US hospitals with stroke or
transient ischemic attack, and provides valuable information on the effectiveness and
safety of care delivery.10 This registry has
played a vital part in the defini­tion, assessment and validation of hospital performance
metrics that have become the standard for
measuring stroke-care quality. The latest
GWTG-Stroke study 10 suggests that although
in-hospital mortality may be lower at registry
sites, the results are largely generalizable to
patients in non-registry hospitals. The measures defined by GWTG-Stroke are among
the first to be incorporated into a new, electronically derived set of quality measures in
the Meaningful Use program, which aims to
shift health records to an electronic system
and define quality metrics in a more affordable, actionable and cost-effective manner.
Strategies that seek to minimize the costly
practice of capturing patient-level clinical
information while maximizing the use of
more cost-effective, administratively derived
health records are likely to form the backbone
of future quality improvement efforts.
Together, the advances described above
reflect progress across the spectrum of stroke
care (Figure 1). From novel drug or device
development to new uses for approved drugs,
every step forward takes us closer to the ultimate goal of better outcomes at lower cost for
patients with stroke and their families.
Department of Neurology-ACC 720, Harvard
Medical School, Massachusetts General
Hospital, 55 Fruit Street, Boston, MA 02114,
[email protected]
Competing interests
The author declares no competing interests.
Schwamm, L. H. et al. Recommendations for
the establishment of stroke systems of care:
recommendations from the American Stroke
Association’s Task Force on the Development
of Stroke Systems. Circulation 111,
1078–1091 (2005).
2. Granger, C. B. et al. Apixaban versus warfarin in
patients with atrial fibrillation. N. Engl. J. Med.
365, 981–992 (2011).
3. Diener, H.‑C. et al. Dabigatran compared with
warfarin in patients with atrial fibrillation and
previous transient ischemic attack or stroke.
Lancet Neurol. 9, 1157–1163 (2010).
4. Patel, M. R. et al. Rivaroxaban versus warfarin
in nonvalvular atrial fibrillation. N. Engl. J. Med.
365, 883–891 (2011).
5. Chimowitz, M. I. et al. Comparison of warfarin
and aspirin for symptomatic intracranial arterial
stenosis. N. Engl. J. Med. 352, 1305–1316
6. Chimowitz, M. I. et al. Stenting versus
aggressive medical therapy for intracranial
arterial stenosis. N. Engl. J. Med. 365,
993–1003 (2011).
7. Hacke, W. et al. Intravenous desmoteplase in
patients with acute ischaemic stroke selected
by MRI perfusion–diffusion weighted imaging or
perfusion CT (DIAS‑2): a prospective,
randomised, double-blind, placebo-controlled
study. Lancet Neurol. 8, 141–150 (2009).
8. Lansberg, M. G. et al. RAPID automated patient
selection for reperfusion therapy: a pooled
analysis of the Echoplanar Imaging
Thrombolytic Evaluation Trial (EPITHET) and the
Diffusion and Perfusion Imaging Evaluation for
Understanding Stroke Evolution (DEFUSE)
study. Stroke 42, 1608–1614 (2011).
9. Chollet, F. et al. Fluoxetine for motor recovery
after acute ischaemic stroke (FLAME):
a randomised placebo-controlled trial. Lancet
Neurol. 10, 123–130 (2011).
10. Reeves, M. J. et al. Representativeness of the
Get With The Guidelines-Stroke Registry:
comparison of patient and hospital
characteristics among medicare beneficiaries
hospitalized with ischemic stroke. Stroke
Translating new research findings
into clinical practice
Christine Klein and Dimitri Krainc
The discovery of mutations that contribute to movement disorders
has facilitated the identification of converging pathways and novel
therapeutic targets. Successful translation of these research findings into
clinical practice will require identification of early markers of disease
progression, and recent research indicates that progress is being made in
this area.
Klein, C. & Krainc, D. Nat. Rev. Neurol. 8, 65–66 (2012); published online 20 December 2011;
The prevalence of Parkinson disease (PD)
and other neurodegenerative disorders
is rapidly increasing worldwide and is
expected to double by 2030,1 yet not a single
disease-modifying treatment exists for this
disorder. Thus, translation of research findings into all aspects of clinical practice—that
is, diagnosis, monitoring and treatment—in
movement disorders is a high priority. One
of the major challenges of translational
research in movement disorders has been
the apparent lack of specific molecular tar­
gets as platforms for drug development.
How­e ver, as highlighted by a number of
important studies published in 2011, the
emergence of various forms of genetic PD
has offered us an exciting opportunity to
identify such targets.
Identification and monitoring
of the earliest disease stages
… is of great importance
Most mutations that have been linked
to PD and related synucleinopathies seem to
converge on lysosomal and mito­chondrial
pathways, suggesting that disruptions of
these pathways have key roles in disease
pathogenesis. The importance of lysosomal
function in synucleinopathies was highlighted in a recent study by Mazzulli and
colleagues that examined Gaucher disease,
a rare lysosomal storage disorder caused by
mutations in the glucocerebro­sidase (GBA1)
gene.2 Patients with Gaucher disease and
their relatives have an increased incidence of
PD, and patients with idiopathic PD have an
elevated prevalence of mutations in GBA1.
Interestingly, accumulated α‑synuclein
interferes with trafficking of glucocerebro­
sidase from the endo­p lasmic reticulum
to the Golgi apparatus, which in turn leads to
decreased glucocerebrosidase activity and
KEY ADVANCES IN MEDICINE more accumulation of α‑synuclein. The
bidirectional effects of α‑synuclein and
glucocerebro­sidase form a positive feedback
loop that, beyond a certain threshold, leads
to self-propagating disease.
The findings suggest that this molecular
pathway applies also to patients with idiopathic PD or other synucleinopathies who
carry a normal GBA1 gene. The results indicate that therapeutic targeting of mutated
or normal gluco­cerebrosidase to lysosomes
would be expected to prevent or diminish
the formation of toxic α‑synuclein oligomers
and break the vicious circle of α‑synuclein
aggregation and toxicity. Importantly, these
findings were confirmed in human neurons
derived from induced pluripotent stem
cells,2 which represent a major advance for
the modeling of neurodegenerative and
other diseases.3
The importance of lysosomal pathways
in PD was further highlighted in two backto-back publications by Vilarino-Guell
et al.4 and Zimprich et al.5 that described
a dominantly inherited mutation in the
vacuo­lar protein sorting 35 (VPS35) gene as
a novel, albeit rare, cause of late-onset PD.
As a component of the retromer complex,
VPS35 sorts cargo from endosomes back to
the trans-Golgi network, and disruptions in
this pathway are expected to affect lysosomal
function. Intriguingly, both studies detected
the same VPS35 mutation (Asp620Asn), in
Swiss and Austrian kindreds, respectively,
and the same mutation was also recently
identified in a German pedigree.6 When the
analyses were extended to additional families from various ethnic backgrounds, both
initial studies confirmed the Asp620Asn
mutation in independent cases and found
two additional variants in the VPS35 gene.4,5
On the basis of these studies, and previous
data showing that the retromer regulates
endosomal sorting of amyloid precursor
protein in models of Alzheimer disease, it
will be interesting to examine the role of
VPS35 in the pathogenesis of PD. These
studies also highlight the importance of
next-generation sequencing for the dis­
covery of genetic factors—many of which
are likely to be rare—that cause or ­contribute
to complex movement disorders.7
Besides improving our understanding of
disease biology, the identification of monogenic forms of movement disorders has
opened up a new window into the pre­motor
stage of neurodegenerative conditions—the
historical frontrunner being Huntington
disease—whereby individuals at risk can
be studied prospectively. Identification
and monitoring of the earliest disease
stages, during which treatment and neuro­
protection are expected to be most effective,
is of great importance. Selection of carefully
characterized and homogeneous cohorts of
patients or at-risk individuals will be vital
for the development of new drugs for PD
and other movement disorders. This task
would be greatly facilitated by employing
adequate biomarkers for diagnosis of the
disease, and for monitoring its course and
potential effects of treatment.
In a study published in PLoS ONE,
Yanamandra et al. hypothesized that autoimmune reactivity towards specific proteins and self-assembled complexes that
are involved in disease pathology may
serve as sensitive biomarkers of neuro­
degeneration.8 The researchers measured
Key advances
■■ α-Synuclein and glucocerebrosidase
form a bidirectional pathogenic loop,
providing an important pathophysiological
mechanism in the development of
Parkinson disease (PD)2
■■ Next-generation sequencing has led to
the discovery of a new gene (VPS35)
associated with late-onset PD; this
finding implicates retrograde transport in
PD pathogenesis4,5
■■ α-Synuclein-reactive antibodies may
serve as a new biomarker for PD,
especially in its early phase8
■■ Induced pluripotent stem cell-derived
neurons represent a patient-specific
cellular model with great potential for
the study of disease biology2
■■ The first double-blind, sham-surgerycontrolled, randomized trial of viral gene
transfer of glutamic acid decarboxylase
resulted in improved motor scores in
patients with PD, and the treatment
was not associated with severe
adverse events9
JANUARY 2012 | S57
levels of auto­antibodies against α‑synuclein,
the main component of Lewy bodies, in the
serum of patients with PD in early and late
stages of the disease. Significantly higher
auto­antibody levels were found in patients
than in controls, but the antibody response
decreased as the condition became more
advanced. The authors interpreted this
find­ing in the context of a protective role for
auto­immunity in PD that seems to be stronger in the early phases of the disease, and
may be of value as a biomarker, especially at
this crucial stage of PD pathogenesis.
…the treatment of PD has
been limited to symptomatic
drug therapy and deep brain
In the absence of neuroprotective therapies, the treatment of PD has been limited
to symptomatic drug therapy and deep
brain stimulation. To explore the potential of gene therapy, LeWitt and colleagues
performed the first double-blind, shamsurgery-­c ontrolled, randomized trial of
in vivo gene transfer in patients with PD.9
The investigators used adeno-associated
viral vector (AAV2) for transduction of
the gene encoding glutamic acid decarboxylase (GAD), an enzyme that catalyzes the decarb­oxylation of glutamate to
γ‑aminobutyric acid and is expected to
increase the inhibitory tone in the basal
ganglia. The trial involved 55 patients aged
30–75 years with levodopa-responsive
PD, who were randomly assigned to sham
surgery (n = 23) or AAV2–GAD infusions
(n = 22) at baseline.
At the 6‑month study end point, 21 and
16 patients were assessed in the shamtreated and gene-therapy groups, respectively. The mean Unified Parkinson’s
Disease Rat­ing Scale (UPDRS) motor score
was found to have decreased by 8.1 points
(23.1%) in the AAV2–GAD group and
by 4.7 points (7.0%) in the sham group. 9
Over­all, the study is encouraging, since the
group dif­ference was highly significant and
no serious adverse events were observed in
either group. Caveats of this study include
rela­tively small group sizes, a heterogeneous
patient population in terms of age of onset,
more-severe dis­e ase in the sham group,
limited dis­criminatory power of the UPDRS
(maximum 108 points) in the lower range,
and the relatively short follow-up period.
In conclusion, these studies highlight the
importance of identifying the molecular
S58 | JANUARY 2012
underpinnings of disease pathways to
uncover therapeutic targets. The discovery
of genetic mutations in these pathways has
sug­gested converging mechanisms in the
patho­genesis of movement disorders. In
order to rapidly translate research advances
into the clinic, it is critically important to
iden­t ify markers of disease progression,
especially in individuals who are at risk of
developing a neurodegenerative disease.
Department of Neurology, University of Lübeck,
Ratzeburger Allee 160, 23538 Lübeck,
Germany (C. Klein). Department of Neurology,
Massachusetts General Hospital, Harvard
Medical School, 114 16th Street, Room 2007,
Charlestown, MA 02129, USA (D. Krainc).
Correspondence to: C. Klein
[email protected]
C. Klein is funded by a career development award
from the Hermann and Lilly Schilling Foundation.
D. Krainc is supported by the National Institute of
Neurological Disorders and Stroke and the
CHDI Foundation.
Competing interests
The authors declare no competing interests.
Dorsey, E. R. et al. Projected number of people
with Parkinson disease in the most populous
nations, 2005 through 2030. Neurology 68,
384–386 (2007).
Mazzulli, J. R. et al. Gaucher disease
glucocerebrosidase and α‑synuclein form a
bidirectional pathogenic loop in
synucleinopathies. Cell 146, 37–52 (2011).
Kiskinis, E. & Eggan, K. Progress toward the
clinical application of patient-specific
pluripotent stem cells. J. Clin. Invest. 120,
51–59 (2011).
Vilarino-Guell, C. et al. VPS35 mutations in
Parkinson disease. Am. J. Hum. Genet. 89,
162–167 (2011).
Zimprich, A. et al. A mutation in VPS35,
encoding a subunit of the retromer complex,
causes late-onset Parkinson disease. Am.
J. Hum. Genet. 89, 168–175 (2011).
Kumar, K. W. et al. Frequency of the D620N
mutation in VPS35 in Parkinson disease. Arch.
Neurol. (in press).
Lill, C. M. & Bertram, L. Complex genetics of
neurodegenerative diseases. Semin. Neurol.
(in press).
Yanamandra, K. et al. α‑Synuclein reactive
antibodies as diagnostic biomarkers in blood
sera of Parkinson’s disease patients. PLoS ONE
6, e18513 (2011).
LeWitt, P. A. et al. AAV2–GAD gene therapy for
advanced Parkinson’s disease: a double-blind,
sham-surgery controlled, randomised trial.
Lancet Neurol. 10, 309–319 (2011).
Advances in therapy, imaging
and risk factors in MS
Bianca Weinstock–Guttman and Murali Ramanathan
Multiple sclerosis research in 2011 produced a combination of new
therapeutic developments and innovative findings. Teriflunomide showed
beneficial effects in a phase III trial, quantification methods for MRI
lesions that should improve monitoring of disease progression were
devised, and a link between high cholesterol and low vitamin D emerged.
Weinstock–Guttman, B. & Ramanathan, M. Nat. Rev. Neurol. 8, 66–68 (2012); published online 10 January 2012;
For multiple sclerosis (MS), the research
milestones of 2011 were a combination of
new therapeutic developments and innovative findings that might help to explain the
heterogeneity in both disease progression
and responses to therapy in patients with
MS. The new oral therapy teriflunomide
showed beneficial effects in a large phase III
clinical trial,1 which provides hope for better
and more-convenient control of MS in the
future. The identification of new methods
to quantify cortical lesions on MRI will
improve our ability to monitor disease progression,2 and studies have identified raised
cholesterol and reduced vitamin D levels as
detrimental factors in patients with MS.3,4
From a therapeutic standpoint, 2011
will be remembered as a defining year
for MS. Following a successful clinical
trial, reported in 2010, 5 fingolimod—an
oral d
­ isease-modifying therapy for MS—
received regulatory approval and became
available to patients in over 50 countries
including the USA, the European Union
and, most recently, Japan. The results from
a phase III clinical trial of teriflunomide
(another oral therapy for MS) were published in October 2011 in The New England
Journal of Medicine.1 Teriflunomide is the
active metabolite of leflunomide, which
is already approved for the treatment of
patients with rheumatoid arthritis, and it
acts by reversi­bly inhibiting dihydroorate
dehydrogenase, a key mitochondrial
enzyme involved in pyrimidine synthesis,
to block the proliferation of rapidly dividing lymphocytes that are thought to drive
inflammation in MS.
The trial reported by O’Connor et al.1
involved 1,088 patients with active relapsing–
remitting MS (RRMS)—defined as one or
more relapse in the previous year or two
or more relapses in the previous 2 years—
ranging in age from 18–55 years. Patients
were randomly assigned to receive either
teriflunomide (7 mg or 14 mg), or placebo,
once daily for 108 weeks. The primary end
point was the annualized relapse rate, and
the secon­dary end point was confirmed progression of dis­ability for at least 12 weeks.
Compared with placebo, teriflunomide
reduced the annualized relapse rate by
31.2% and 31.5% in the 7 mg and 14 mg dose
groups, respectively (P <0.001). The proportion of patients with confirmed disability
progression was 27.3% with placebo, 21.7%
with 7 mg teriflunomide (P = 0.08), and
20.2% with 14 mg teriflunomide (P = 0.03).
Teriflunomide therapy reduced the risk of
disability progression by 23.7% (P = 0.008)
for 7 mg and 29.8% (P = 0.03) for 14 mg.
From a therapeutic standpoint,
2011 will be remembered as a
defining year for MS
Both teriflunomide doses were sup­
erior to placebo on multiple MRI end
points, including change in lesion volume
from base­line, number of gadolinium
contrast-enhancing lesions, and unique
active lesions. How­ever, no differences in
brain atrophy were observed between the
teriflunomide-treated and placebo-treated
groups. Adverse events, such as diarrhea,
nausea and hair thin­n ing, were more
common in t­ eriflunomide-treated patients
than in patients who received placebo, and
the incidence of elevated ala­nine aminotransferase levels (a sign of liver damage)
was higher in patients who received teriflunomide at 7 mg and 14 mg (54.0% and
57.3%, respectively) than in those who
received placebo (35.9%). The incidence
of serious infections was similar across
the two treatment groups and the placebo
group. The beneficial therapeutic effects
and safety profile seen in teriflunomidetreated patients suggests that this new oral
drug should be considered in the evolving
therapeutic a­ rmamentarium for MS.
KEY ADVANCES IN MEDICINE The Expanded Disability Status Score
(EDSS) is often used as an outcome measure­
ment in clinical trials; however, the poor
sensitivity of this scoring system limits its
usefulness. The idea that lesions in patients
with MS occur not only in the white matter,
but also in cerebral gray matter, especially
in the cerebral cortex, is now well-accepted.6
Reporting in PLoS ONE, Sormani et al.2 used
double inversion recovery imaging—an MRI
technique that enables selective imaging of
gray matter—to detect cortical lesions in
patients with RRMS. Cortical lesions could
be seen in patients at the earliest clinical
stages of MS, and lesion burden correlated
positively with the severity of physical and
cognitive impairments.
Sormani and colleagues showed that the
zero-inflated Poisson distribution accurately modeled the distribution of new
cortical lesions that developed over 12 and
24 months.2 The researchers also estimated
the sample size needed for clinical trials
that use emergence of new cortical lesions
as the primary outcome measure. A clinical trial with an assumed treatment effect
of 50%, with 90% power and a significance
level of 5%, would require a sample size that
ranged from 72 to 200 patients per treatment group for a 1‑year trial and from 48
to 110 patients for a 2‑year trial. Assuming
a treatment effect of 30%, the estimated
sample size was 212 to 630 patients per arm
for a 1‑year trial and 150 to 320 for a 2‑year
trial. These results suggest that clinical trials
with the number of new cortical lesions as
the primary outcome are feasible. Cortical
lesion measurements may become a useful
marker of disease progression in patients
with MS, especially in the evaluation of the
neuroprotective and regenerative effects of
new therapeutic agents.
2011 has seen an increase in research
into the roles of environmental risk factors,
genetic and environmental interactions, and
comorbidities in determining the severity of
disease in patients with MS. High cholesterol
is a well-established risk factor for cardio­
vascular disease, but the role of lipids as a
risk factor for MS has not been extensively
investigated. In a study from our group,
published in October 2011 in The Journal
of Neuroinflammation, we used clinical and
MRI metrics to identify lipid profiles that
were associated with disease progression.3
492 patients with MS (age 47.1 ± 10.8 years;
disease duration 12.8 ± 10.1 years), who had
baseline and follow up EDSS assessments
after a mean period of 2.2 ± 1.0 years, were
included in the study. Quantitative MRI
Key advances
■■ Teriflunomide—a new oral
immunomodulatory therapy—showed
significant benefit in reducing relapses
and slowing disease progression in a
large phase III clinical trial in patients with
multiple sclerosis (MS)1
■■ New MRI methods that used double
inversion recovery imaging were shown
to detect cortical lesions, which may
be useful as measurements of disease
progression in clinical trials2
■■ Disease progression in MS was linked to
lipid profiles; high cholesterol was shown
to negatively affect vitamin D levels in
patients with the disease3,4
findings at baseline were available for 210
patients. Worsening of disability was associ­
ated with higher baseline LDL (P = 0.006)
and total cholesterol levels (P = 0.001), and
patients with higher triglycerides showed
a trend towards worse EDSS outcomes
(P = 0.025). However, no association was
observed between baseline HDL levels and
clinical outcome.
A similar association between high LDL
and disease worsening was found when we
used the MS severity scale. However, patients
with high HDL levels had lower contrastenhancing lesion volumes than patients with
low HDL (P <0.001) and high total chol­
esterol was associated with a trend for lower
brain parenchymal fraction (P = 0.033).
These data suggest that high cholesterol
and triglycerides have a negative influence
on disease course, and that high HDL levels
have a favorable influence on acute MRI
inflammatory activity in patients with MS.
Cortical lesion measurements
may become a useful marker of
disease progression in patients
with MS…
High levels of serum cholesterol and lipids
may have both direct and indirect effects on
MS disease progression. Lipoproteins could
directly enhance inflammation in patients
with MS, and as 7‑dehydrocholesterol is a
substrate for the endogenous synthesis of
vitamin D in the skin, high cholesterol may
have indirect effects on vitamin D. Indeed,
we found that patients with a high ratio
of cho­lesterol to HDL had reduced levels of
vita­min D.4 Lifestyle changes to improve
serum lipid profiles, such as adoption of a
healthier diet and regular exercise, may be
beneficial for patients with MS.
JANUARY 2012 | S59
In summary, research in 2011 conveyed
new findings to help us to understand, treat
and prevent disease progression in patients
with MS. Better imaging of cortical lesions
could provide insights into the patho­biology
of MS and may yield new outcome measures
for MS clinical trials. Advances in understanding the environmental factors that are
associated with disease progression in MS
could facilitate the development of better
therapies and personalized management for
patients with this disabling disease.
Jacobs Neurological Institute, Buffalo General
Hospital, University at Buffalo, State University
of New York, Building E, 2nd Floor, 100 High
Street, Buffalo, NY 14203, USA
(B. Weinstock‑Guttman). Department of
Pharmaceutical Sciences and Neurology, State
University of New York, 427 Cooke Hall, Buffalo,
NY 14260, USA (M. Ramanathan).
Correspondence to: B. Weinstock–Guttman
[email protected]
The authors would like to thank the Department of
Defense, the Jog for The Jake Foundation, National
Multiple Sclerosis Society, National Science
Foundation and NIH for providing financial support
for their research activities.
Competing interests
B. Weinstock–Guttman declares associations with
the following companies and organizations: Acorda,
Biogen Idec, Cyberonics, EMD Serono, Novartis,
Pfizer, Teva Neuroscience. M. Ramanathan declares
associations with the following companies and
organizations: Allergan, the American Association of
Pharmaceutical Scientists, Biogen Idec, EMD
Serono, Netezza, Novartis, Pfizer. See the article
online for full details of the relationships.
O’Connor, P. et al. Randomized trial of oral
teriflunomide for relapsing multiple sclerosis.
N. Engl. J. Med. 365, 1293–1303 (2011).
Sormani, M. P. et al. Modeling the distribution
of new MRI cortical lesions in multiple
sclerosis longitudinal studies. PLoS ONE 6,
e26712 (2011).
Weinstock-Guttman, B. et al. Serum lipid
profiles are associated with disability
and MRI outcomes in multiple sclerosis.
J. Neuroinflammation 8, 127 (2011).
Weinstock-Guttman, B., Zivadinov, R. &
Ramanathan, M. Inter-dependence of vitamin
D levels with serum lipid profiles in multiple
sclerosis. J. Neurol. Sci. 311, 86–91 (2011).
Kappos, L. et al. A placebo-controlled trial of
oral fingolimod in relapsing multiple
sclerosis. N. Engl. J. Med. 362, 387–401
Calabrese, M., Filippi, M. & Gallo, P. Cortical
lesions in multiple sclerosis. Nat. Rev. Neurol.
6, 438–444 (2010).
Microbleeds in dementia
—singing a different ARIA
Philip Scheltens and Jeroen D. C. Goos
In 2011, researchers used imaging techniques to investigate
brain microbleeds in patients with dementia and highlighted how lobar
microbleeds could be used as a marker for amyloid pathology and for
predicting mortality. New guidelines on the inclusion and exclusion
of participants with microbleeds in anti-amyloid clinical trials were
also published.
Scheltens, P. & Goos, J. D. C. Nat. Rev. Neurol. 8, 68–70 (2012); published online 10 January 2012;
Following initial reports of small dotlike lesions on gradient-echo MRI in the
brains of patients with dementia (Figure 1),
many investigators set out to describe
the prevalence and incidence of these
lesions—later designated as micro­bleeds
or m
­ icrohemorrhages—in both healthy
people and individuals with this condition.
Prevalence rates varied from 10% in healthy
elderly individuals to 60% in patients with
vascular dementia.1 Interest in these lesions
peaked when the first cases of incident
micro­bleeds along with increased signal
inten­sity on fluid-attenuated inversion recovery imaging, thought to represent vaso­genic
edema, were reported in patients receiving
S60 | JANUARY 2012
amyloid-lowering therapy. During the 2010
Inter­national Conference on Alzheimer
Disease (AD), turmoil ensued over a cautionary letter from the FDA, which suggested
drastic cut-offs in randomized clinical trials
of amyloid-lowering drugs, both in terms of
excluding patients with a single microbleed
and terminating the participation of patients
who developed a new microbleed during
the study. In response, a series of important papers regarding the detection, prevalence and clinical relevance of microbleeds
in patients with dementia-related disease
appeared in 2011, includ­ing a consensus
statement from an inter­national working
group that introduced new terminology.
At the beginning of 2011, Cordonnier
and van der Flier reviewed the available
litera­ture on brain microbleeds in patients
with AD. 1 The authors suggested that
these lesions were associated with amyloid
pathology and may have a crucial role in the
pathophysiology of AD. Microbleeds were
proposed to represent a link between the
amyloid cascade hypo­thesis and the vascular ­hypothesis—both popular explanatory
models for the pathogenesis of AD. Further­
more, the location of the microbleeds was
suggested to indicate their underlying etio­
logy: lobar microbleeds would presumably
be associated with cerebral amyloid angio­
pathy (CAA), whereas microbleeds in deep
brain regions would be associated with
hypertensive vasculopathy and increased
risk of vascular complications. The clinical implications of microbleeds in patients
with dementia were also stressed. Besides
an association with cognition, micro­bleeds
have been linked to mortality, especially in
cases of multiple lesions, as described by
Henneman et al. in a 2009 study.2 These
investigators did not, however, assess the
cause of death in their patients.
Throughout 2011—within months of
publication of the review by Cordonnier
and van der Flier 1—reports were published
on several studies that have substantially
extended our knowledge on microbleeds
in patients with dementia-associated diseases. The suggestion that micro­b leeds
were closely linked to amyloid pathology
was supported by the findings of Yates et al.
for the Australian Imaging, Bio­m arkers
and Lifestyle Study of Ageing Research
Group.3 They found that even in healthy
con­trols, lobar microbleeds (detected using
3T ­susceptibility-weighted imaging [SWI])
were associated with higher amyloid burden,
as seen on 11C-Pittsburgh compound B (PiB)
PET imaging. Moreover, PiB-positive scans
were more prevalent in participants with
multiple lobar microbleeds (86%) than
in those with only one lobar microbleed
(67%). In agreement with these findings,
results from a study by Goos et al.,4 in which
cerebro­spinal fluid amyloid biomarkers were
used to assess amyloid burden, confirmed
the relationship between lobar microbleeds
and amyloid pathology.
Altmann–Schneider and colleagues
studied the relationship between microbleeds and mortality (with assessment of
the cause of death) in a population of 435
elderly people with pre-existing vascular
dis­ease.5 Individuals with more than one
microbleed had a sixfold increase in the risk
of stroke-related death compared with those
with no lesions. The location of the lesions
also affected the risk of poor clinical outcomes: compared with individuals without
any lesions, patients with nonlobar microbleeds had a twofold increase in the risk of
cardiovascular death, and indivi­duals with
probable CAA type (lobar) microbleeds had
a sevenfold increase in the risk of strokerelated death. These findings support the
hypothesis that microbleeds have separate
etiologies depending on their location in
the brain,1 and may provide an indication
for the use of anticoagulant treatment in
patients with these lesions—in particular,
those with lobar microbleeds.
The importance of detecting microbleeds
before and during clinical trials was underlined in an extensive review on amyloidrelated imaging abnormali­ties (ARIA) by
Sperling et al. for the Alzheimer’s Associ­
ation Research Roundtable Work­g roup.6
These imaging abnormalities, seen on MRI,
are thought to represent a spectrum of ‘leaky
vessels’ that occur following anti-­amyloid
immunization therapy. The working hypo­
thesis was that vasogenic edema (now called
ARIA-E) caused leakage only of protein­
aceous fluids from the vessels, whereas
microbleeding—under the new umbrella
term ARIA-H (for h
­ emorrhage)—caused
more-extensive leakiness of the vessels
that allowed blood cells to cross the blood–
brain barrier. For patients in clinical trials,
the presence of microbleeds at baseline
may be a risk factor for developing ARIA,
Key advances
■■ Lobar microbleeds in patients with
dementia were found to be associated
with high amyloid burden3
■■ Microbleeds were shown to have
different etiologies and associated risk
of mortality on the basis of their location
in the brain5
■■ New terminology for amyloid-related
imaging abnormalties (ARIA) and a
new cut-off point for microbleeds in
participants of amyloid-modifying clinical
trials were introduced6
■■ Susceptibility-weighted imaging showed
enhanced sensitivity for detecting
microbleeds compared with gradientecho imaging, and may help in future
for identifying associations between
microbleeds and clinical outcomes7
■■ Studies that compared imaging with
histopathology of microbleeds suggest
that further refinement of imaging
techniques is required to accurately
detect these lesions8
KEY ADVANCES IN MEDICINE as the number of lobar lesions is assumed
to ­correlate with the presence and severity
of CAA.
Given the uncertainty regarding the risk
of ARIA and concerns about CAA severity,
Sperling et al. recommended a cut-off value
for the exclusion of participants in trials of
amyloid-modifying therapies for AD at four
microbleeds.6 This new guideline allows for
variability in imaging measurements and
reflects the uncertainty regarding the clinical
relevance of small numbers of microbleeds.
The authors further stated that occurrence of
new asymptomatic microbleeds in patients
during trials should not auto­matically disqualify them from receiving further treatment; however, owing to a lack of data,
no exact cut-off for dis­qualification could
be given. As the authors stressed, counting of microbleeds is not an exact science,
and the uncertainty is further complicated
by the varying sensitivities of different MRI
techniques for detecting these lesions.
Reporting in Stroke in May 2011, Goos
et al.7 compared conventional gradient-echo
imag­ing with SWI to detect microbleeds in
140 patients from a memory clinic, and
also to determine whether microbleeds
were associated with patient and clinical characteristics. As expected, use of the
more-advanced SWI technique enabled
identi­fication of patients with microbleeds
with a greater sensitivity than could be
achieved with gradient-echo imaging (40%
versus 23%). SWI also detected a higher
num­b er of microbleeds per patient than
did gradient-echo imaging. However, the
corre­lation between lesion numbers, clinical
out­comes and other radiological outcomes
was limited. The clinical relevance of microbleeds will probably depend more on their
location and size, than on the total number
of lesions per se. The authors concluded,
therefore, that although new imaging techniques can show a higher number of lesions,
conventional imaging methods can already
detect the majority of clinically relevant
lesions. New imaging methods might help
in identifying any associations between
lesions and clinical and radiological outcomes; however, these new techniques are
in urgent need of validation.
De Reuck et al. 8 made an interesting
attempt to validate MRI findings with
patho­logy, as reported in Cerebrovascular
Disorders. The researchers investigated
20 post­mortem brains from patients with
AD with different cerebrovascular lesions.
Images of 45 large sections of the cerebral
hemi­spheres, brainstem and cere­b ellum
Figure 1 | Brain microbleeds on MRI.
Numerous lobar microbleeds with sparing of
the basal ganglia and thalamus, suggestive
of severe cerebral amyloid angiopathy in a
71-year-old patient with dementia with
Lewy bodies. Image obtained using
susceptibility-weighted imaging at 3T.
obtained using 7.0T T2*-weighted MRI
were paired with images showing histo­
logical detection of hematomas and microbleeds. In the cortico-­subcortical regions,
the sensitivity, specificity, and positive and
negative predictive values of T2* imag­
ing to detect microbleeds were excellent.
How­e ver, analysis of MRI alone resulted
in an over­e stimation of microbleeds in
the stria­tum due to the presence of iron
de­p osits that were, in fact, not related to
real hemor­rhages. Furthermore, 31% of T2*
hypo­signals in the deep white matter were
shown to be vessels filled with post­mortem
thrombi. Judging from these findings,
more studies are needed before we can fully
understand the correlations between MRI
and pathology in patients with AD.
From this selection of studies published
in 2011, we can conclude that lobar micro­
bleeds are a marker for underlying amyloid
patho­logy, are associated with strokerelated mortality, and should be adequately
investi­g ated in patients participating in
anti-­amyloid clinical trials. Optimizaton
of imaging methods to detect microbleeds will be of the utmost importance,
and emerging techni­ques will need to be
evaluated and calibrated using clinical and
pathological correlates.
VU University Medical Center, Alzheimer Center,
Department of Neurology, De Boelelaan, 1118
1081 HZ Amsterdam, The Netherlands
(P. Scheltens, J. D. C. Goos).
Correspondence to: P. Scheltens
[email protected]
JANUARY 2012 | S61
Competing interests
The authors declare no competing interests.
Cordonnier, C. & van der Flier, W. M. Brain
microbleeds and Alzheimer’s disease: innocent
observation or key player? Brain 134, 335–344
Henneman, W. J. et al. MRI biomarkers of
vascular damage and atrophy predicting
mortality in a memory clinic population. Stroke
40, 492–498 (2009).
Yates, P. A. et al. Cerebral microhemorrhage
and brain β-amyloid in aging and Alzheimer
disease. Neurology 77, 48–54 (2011).
Goos, J. D. C. et al. Microbleeds relate to
altered amyloid-β metabolism in Alzheimer
disease. Neurobiol. Aging
Altmann–Schneider, I. et al. Cerebral
microbleeds are predictive of mortality in the
elderly. Stroke 42, 638–644 (2011).
Sperling, R. A. et al. Amyloid-related imaging
abnormalities in amyloid-modifying therapeutic
trials: recommendations from the Alzheimer’s
Association Research Roundtable Workgroup.
Alzheimers Dement. 7, 367–385 (2011).
Goos, J. D. C. et al. Clinical relevance of
improved microbleed detection by
susceptibility-weighted magnetic resonance
imaging. Stroke 42, 1894–1900 (2011).
De Reuck, J. et al. Comparison of 7.0‑T T2*magnetic resonance imaging of cerebral bleeds
in post-mortem brain sections of Alzheimer
patients with their neuropathological
correlates. Cerebrovasc. Dis. 31, 511–517
Insights into epilepsy treatments
and biomarkers
Fernando Cendes
Research published in 2011 identified important factors related to
serious adverse effects of antiepileptic drugs and sudden unexpected
death in epilepsy, along with a potential new treatment and a promising
marker of epileptogenesis. Further advances in these areas are urgently
needed to improve the lives of people with epilepsy.
Cendes, F. Nat. Rev. Neurol. 8, 70–71 (2012); published online 10 January 2012;
Epilepsy affects people of all ages, is
highly prevalent, and can have a good
out­c ome with appropriate anti­e pileptic
drugs (AEDs). However, seizures are
often refractory to clinical treatment, with
­detrimental—and even life-threatening—
consequences. 1 Unpredictable events,
such as serious adverse effects of AEDs
and sudden unexpected death in epilepsy
(SUDEP), make this condition even harder
to manage. The fact that epilepsy still carries
a great stigma, with many people hiding
their condition, hinders public awareness
and, in addition, makes it more difficult for
the condition to be recog­nized as a serious
public health problem, resulting in reduced
availability of research resources compared
with other common diseases. Despite such
obstacles, however, considerable advances
are being made on a number of fronts, as
illustrated by several key papers published
during 2011.
Sudden unexpected death is 20 times
more frequent in people with epilepsy
than in the general population, but the risk
factors identified for SUDEP have not been
consistent across the literature. A recent
study by Hesdorffer et al. has helped to identify groups of people with epilepsy who are
S62 | JANUARY 2012
The currently available AEDs
do not prevent or cure epilepsy,
and are merely antiseizure
at particular risk of SUDEP.2 In this pooled
analysis of four case–control studies, earlyonset refractory symptomatic epilepsy, frequent generalized tonic–clonic seizures, and
use of AED polytherapy were all found to
be significant risk factors for SUDEP. Such
studies are extremely important for drawing
attention to the need to achieve complete
seizure control in people with epilepsy.
The currently available AEDs do not
prevent or cure epilepsy, and are merely
antiseizure medications. This fact underscores our lack of knowledge about the
exact mechanisms of seizure genesis and
modes of action of AEDs. In this context,
a paper by Jeon et al.3 was another important publication in 2011. The researchers studied the effects of cell-free extract
derived from human adipose stem cells
(ASCs) on the acute and chronic phases of
the pilocarpine epilepsy model in mice.4
Pretreatment with the extract did not affect
seizure susceptibility to pilocarpine, but it
did reduce the number of EEG spikes in the
acute phase, and subsequently diminished
sponta­neous recurrent seizures (SRS) in
the chronic epileptic stage.3 Furthermore,
continuing treatment in the chronic epileptic stage suppressed or inhibited SRS and
had a positive effect on animal behavioral
tests. In addition, animals treated with the
extract had less damage to blood–brain
barrier integrity than did untreated controls.
Interestingly, heat-treated ASC extract had
no beneficial effect, suggesting the importance of cytosolic proteins in the positive
effects of this agent.
Studies have suggested that stem or
precursor cells may rescue degenerating
neurons by modulating the host environment via a chaperone-like mechanism, or
by altering immune-like functions including inflammatory responses.5 Thus, some
cell-based therapies might rely more on a
‘bystander’ mechanism involving secretion of soluble factors—such as cytokines
and chemokines—than on replacement of
damaged neurons by neurogenesis.3 The
ASC extract used by Jeon et al. may act by
modulating inflammatory events during
During 2011, important contributions
were also made to our understanding of
serious adverse effects of AEDs, including risks of allergic reactions and terato­
genesis.6,7 Carbamazepine, one of the most
important AEDs, causes various forms of
hypersensitivity reaction, ranging from
maculopapular exanthema to severe
and potentially lethal conditions such as
Stevens–Johnson syndrome and toxic epidermal necrolysis (SJS–TEN). In a key
publication, McCormack et al. 6 showed
that the presence of the HLA‑A*3101
allele is associated with carbamazepineinduced hypersensitivity reactions among
Europeans. The overall prevalence of carba­
mazepine hypersensitivity is 5.0%; presence
of the HLA‑A*3101 allele increased the
risk to 26.0%, whereas absence of the allele
reduced the risk to 3.8%. Another HLA
allele, HLA‑B*1502, is strongly correlated
with carbamazepine-induced SJS–TEN in
Asian populations but not in Europeans.8
Another serious adverse effect of AEDs is
the increased risk of malformations in the
offspring of women who use these medications during pregnancy. Clear evidence is
available that some AEDs, such as valproic
acid, pose major risks of terato­genesis, and
the perception is that some of the newer
AEDs, including lamotrigine, could be safer.
Key advances
■■ Early-onset refractory symptomatic
epilepsy, frequent generalized tonic–
clonic seizures, and use of antiepileptic
drug (AED) polytherapy are all significant
risk factors for sudden unexpected death
in epilepsy2
■■ A cell-free extract from human adipose
stem cells has antiseizure and
antiepileptogenic effects in mice3
■■ The HLA‑A*3101 allele is associated with
carbamazepine-induced hypersensitivity
reactions among Europeans6
■■ The risk of malformation in offspring of
woman taking AEDs depends not only
on the type of medication, but also on
the AED dose and the parental history of
congenital malformation7
■■ Interictal high-frequency oscillations
measured with scalp electrodes can
reliably delineate the seizure-onset zone
in patients with focal epilepsy9
In a recent important publication, Tomson
et al.7 compared the relative teratogenicity
of four common AEDs—carbamazepine,
pheno­b arbital, valproic acid and lamo­
trigine—on the basis of data from the EURAP
epilepsy and pregnancy registry. The authors
showed that the risk of major congenital malformations increased in a dose-dependent
manner with all four assessed AEDs. Particu­
larly high malformation rates were observed
with valproic acid doses >1,500 mg per day.7
The treatment associ­ated with the lowest rate
of malforma­tions—lamotrigine at a dose of
<300 mg per day—served as a reference for
internal comparisons. Compared with this
reference, the risk of major congenital malformations was significantly elevated with
all doses of valproic acid and phenobarbital,
and with the highest doses of carb­amazepine
(>1,000 mg per day). Valproic acid doses of
<700 mg per day were associated with a malformation rate in a similar range to those with
carba­mazepine at <1,000 mg per day, pheno­
barbital at <150 mg per day, and lamotrigine
at >300 mg per day.7 Therefore, lamotrigine at
high doses is no safer than valproic acid
at lower doses. A parental history of major
congenital malformations was indepen­
dently associated with a fourfold greater risk
of teratogenesis.
Despite a continuous increase in the
range of AED options, as many as 40% of
KEY ADVANCES IN MEDICINE people with focal epilepsy will not achieve
adequate seizure control with AEDs, and are
considered to have drug-resistant epilepsy.1
Surgical treatment can produce seizure
freedom or substantial improvement in
many of these patients. However, a significant proportion of patients cannot undergo
surgery due to lack of adequate localization of the seizure-onset zone (SOZ). This
situation underscores the need for reliable
bio­markers that may help to identify and
predict the severity of epilepsy and the
location of the epileptogenic tissue.
EEG has great potential for investigating the presence or severity of epilepsy
(epilepto­g enicity) and its development
(epilepto­genesis) in vivo and in vitro, owing
to the capacity to utilize both macro­
electrodes and microelectrodes, and to
record normal and abnormal neuronal
firing with excellent time resolution.9,10 As
reported in 2011, Andrade-Valença et al.9
investigated the possibility of noninvasive
detection of interictal high-frequency oscillations (HFOs) via scalp EEG recordings
for more-precise delineation of the SOZ in
patients with focal epilepsy.9 Recording of
HFOs with scalp electrodes was previously
thought to be virtually impossible.10
Surgical treatment can
produce seizure freedom or
substantial improvement in many
… patients
Andrade-Valença et al.9 analyzed the rates
of gamma (40–80 Hz) and ripple (>80 Hz)
oscillations, as well as their co-occurrence
with spikes, and the number of channels
with HFOs inside and outside the SOZ.
The accuracy with which the SOZ could
be identified was 43% for spikes, 70% for
gamma oscillations, and 81% for ripples.
This neuro­p hysiological method could
become an important scalp biomarker for the
In conclusion, 2011 has seen advances
in our understanding of the prevention and
treatment of epilepsy and the avoidance of
life-threatening adverse effects of AEDs,
along with the development of a potential
new biomarker. Investigation of HFOs is of
great importance, since these oscillations
can reveal fundamental mechanisms of
epilepto­genesis and epileptogenicity, and also
have possible clinical value.9,10 The exciting
finding that ASC extracts may be effective
in modulating epileptogenesis in addition to
their antiseizure effects is likely to open new
lines of continuing research.3 The studies
involving risk factors of SUDEP2 and AEDinduced life-threatening hypersensitivity 6 are
major breakthroughs with important practical applications. The evidence that the risk
of major congenital malformations is influenced not only by the type of AED, but also
by dose and other variables, should be taken
into account in the management of epilepsy
in women of childbearing potential.7
Department of Neurology, FCM, University of
Campinas (UNICAMP), Rua Vital Brasil, 251,
Cidade Universitária, Campinas, SP 13.083‑888,
[email protected]
Competing interests
The author declares no competing interests.
Kwan, P. et al. Definition of drug resistant
epilepsy: consensus proposal by the ad hoc
Task Force of the ILAE Commission on
Therapeutic Strategies. Epilepsia 51,
1069–1077 (2010).
2. Hesdorffer, D. C. et al. Combined analysis of
risk factors for SUDEP. Epilepsia 52,
1150–1159 (2011).
3. Jeon, D. et al. A cell-free extract from human
adipose stem cells protects mice against
epilepsy. Epilepsia 52, 1617–1626 (2011).
4. Cavalheiro, E. A. et al. Long-term effects of
pilocarpine in rats: structural damage of the
brain triggers kindling and spontaneous
recurrent seizures. Epilepsia 32, 778–782
5. Vezzani, A., French, J., Bartfai, T. & Baram, T. Z.
The role of inflammation in epilepsy. Nat. Rev.
Neurol. 7, 31–40 (2011).
6. McCormack, M. et al. HLA‑A*3101 and
carbamazepine-induced hypersensitivity
reactions in Europeans. N. Engl. J. Med. 364,
1134–1143 (2011).
7. Tomson, T. et al. Dose-dependent risk of
malformations with antiepileptic drugs:
an analysis of data from the EURAP epilepsy
and pregnancy registry. Lancet Neurol. 10,
609–617 (2011).
8. Chen, P. et al. Carbamazepine-induced toxic
effects and HLA‑B*1502 screening in Taiwan.
N. Engl. J. Med. 364, 1126–1133 (2011).
9. Andrade-Valença, L. P., Dubeau, F., Mari, F.,
Zelmann, R. & Gotman, J. Interictal scalp fast
oscillations as a marker of the seizure onset
zone. Neurology 77, 524–531 (2011).
10. Cendes, F. & Engel, J. Jr. Extending
applications for high-frequency oscillations:
the ripple effect. Neurology 77, 518–519
JANUARY 2012 | S63
As a student or postdoc you can save up to 50% off a one year, print and online,
personal subscription to any of the titles below.
Nature Reviews Cardiology
Nature Reviews Nephrology
Nature Reviews Clinical Oncology
Nature Reviews Neurology
Nature Reviews Endocrinology
Nature Reviews Rheumatology
Nature Reviews Gastroenterology and Hepatology
Nature Reviews Urology
Don’t Miss out!
View the entire series here:
RA IN 2011
Advances in diagnosis, treatment and definition
of remission
Gerd R. Burmester
Evidence presented in 2011 suggests that rheumatoid arthritis might comprise two separate diseases—each
with different etiological underpinnings—and that kinase inhibitors could soon be added to the therapeutic
armamentarium. Together with new definitions of remission, these advances could aid the development of
personalized, treat-to-target strategies.
Burmester, G. R. Nat. Rev. Rheumatol. 8, 65–66 (2012); published online 10 January 2012; doi:10.1038/nrrheum.2011.201
Here, I discuss the progress that has been
made in RA research over the past year, focusing on studies that have increased our understanding of the genetic basis of the disease,
that have demonstrated the efficacy of new
treatment modalities, and that have more
clearly defined our current goal—the state
of remission.
The development of laboratory tests for the
detection of anti-citrullinated protein anti­
bodies (ACPAs) has been a major advance in
the diagnosis of RA, and has been recog­nized
as such through inclusion of these assays in
the most up-to-date American College of
Rheumatology (ACR)–European League
Against Rheumatism (EULAR) disease
classi­f ication criteria.1 However, why the
immune system of patients with RA errone­
ously identifies endogenous citrullinated
proteins as foreign or as molecular danger
signals—sometimes mounting an enormous
response directed at these antigens—remains
an unsolved riddle.
Despite our incomplete understanding of
ACPAs, they have become a cornerstone in
the diagnosis of RA. In 2011, Padyukov et al.2
went one step further when they questioned
whether the presence or absence of ACPAs
could be used to define two distinct subsets
of RA or even two genetically-­distinct diseases. The investigators attempted to answer
this question by performing a genome-wide
associ­ation study to identify differences
in risk allele frequency between patients
with and without ACPAs. Analysis of single
nucleotide polymorphisms (SNP) in ACPAnegative patients with RA revealed that no
SNP tested achieved genome-wide significance in comparison with control indivi­
duals.2 However, in a case–case association
■■ Anti-citrullinated protein antibody (ACPA)
status might define two different subtypes
of rheumatoid arthritis (RA) with different
underlying etiologies2
■■ Kinase inhibitors continue to show
promise in the treatment of RA, particularly
tofacitinib,5 a janus-associated kinase
■■ The publication of remission criteria for RA8
will facilitate treat-to-target strategies and
standardize the reporting of trial outcomes
study between these ACPA-negative indivi­
duals with RA and ACPA-positive patients
with RA, marked differences were identified in the HLA region.2 Notably, only a few
SNPs were shared between ACPA-negative
and ACPA-positive disease, and these shared
alleles provide only a minor contribution to
the overall genetic risk of developing RA.2
These findings suggest that no obvious
common denominator determines overall
genetic susceptibility to RA when ACPAnega­tive and ACPA-positive forms are considered together, but different risk alleles
probably underlie development of each of
these disease subsets.
In contrast to ACPA-negative RA, ACPApositive RA was associated with variation
in genes that have been linked to other
auto­immune diseases, such as type 1 diabetes and systemic lupus erythematosus.2,3
Further­more, most of the genes associated
with ACPA-positive RA, including PTPN22,
CD40 and STAT4, are implicated in inflammatory pathways. The odds ratios for some
of these genetic associations with RA were
low; nevertheless these links could point to
important mechanistic similarities and/or
differences between RA and other inflammatory diseases. Increased understanding
of the etiology of RA derived from such
data could lead to the identification of new
th­erapeutic targets.3
That RA does indeed comprise ‘two diseases’, one ACPA-negative and one ACPApositive, now seems reasonable to conclude.
This discovery will have important implications for the future management of this disorder, particularly for diagnosis and therapy.
However, many important questions remain
unanswered; chiefly, what value do ACPAs
hold in the prediction of subsequent response
to treatment? Prognostic value has been
demon­strated for B-cell-depletion therapy,4
but ACPAs seem to provide less or no information pertaining to the success of treatment
of RA using cytokine blockade.
Once a patient has been diagnosed with
RA, whether ACPA-negative or ACPApositive, the next consideration is treatment
of the disease. An extensive armamenta­
rium of therapeutics is currently available
for the treatment of RA; however, consider­
able unmet medical need still exists as
some patients fail to respond to treatment,
whereas many others suffer relapses after
initially demonstrating clinical improvement.
Many failures have been encountered with
small molecule drugs targeted at proteins
involved in signal transduction, both with
regard to efficacy and toxicity. Never­theless,
some persistent researchers in industry and
academia have continued to develop new
agents, notably targeting tyrosine kinases.
Comprehensive drug development programs are underway, the most advanced
of which seems to involve tofacitinib—a
Janus-associated kinase (JAK) inhibitor.
JANUARY 2012 | S65
Treatment population
Treatment options
Treatment goal
Conventional DMARDs
Cytokine inhibition
B-cell depletion
Remission (defined by TJC,
SJC, PtG and CRP [mg/dl]
all ≤1, or SDAI ≤3.3)
Inhibition of costimulation
Inhibition of tyrosine kinases
Figure 1 | Treatment flow in RA. After diagnosis, it might be important to further stratify a patient
with RA according to ACPA status. A variety of treatment options then exist for RA. Tyrosine
kinases represent promising new agents that might be efficacious in the treatment of RA. The
ultimate aim of treatment is to induce a state of remission. Abbreviations: ACPA, anti-citrullinated
protein antibodies; CRP, C-reactive protein; PtG, patient global assessment score; RA, rheumatoid
arthritis; SDAI, simplified disease activity index; SJC, swollen joint count; TJC, tender joint count.
Tofacitinib binds to and inhibits the important intracellular enzymes JAK1, JAK2 and
JAK3, which are involved in immune cell activation, production of proinflammatory cytokines and cytokine signaling. A major step
towards the introduction of tofacitinib to the
clinic was made in 2011 with the publi­cation
of the results of a phase IIb, 24-week, doubleblind, randomized controlled trial of this
compound by Fleischmann et al.5 The patient
populations enrolled in this study comprised
indivi­duals with RA and an inadequate
response to DMARDs, who were assigned to
receive monotherapy of various dosages of
tofacitinib, the TNF inhibitor adalimumab or
placebo. Treatment with tofacitinib resulted
in a rapid clinical response, with ACR 20%
improvement criteria (ACR20) response at
week 12 ranging from 59.2% (5 mg) to 71.9%
(15 mg), and the response was maintained at
24 weeks.5 By contrast, ACR20 was achieved
by 22% of the placebo group and an adali­
mumab response rate of 35.9% was recorded
—notably, the latter drug was also used
as a monotherapy without metho­trexate.5
Markedly improved ACR50 and ACR70
responses as well as disease activity score in
28 joints (DAS28) remission were observed
for tofacitinib, compared with placebo.5 The
adverse events associated with tofacitinib
treatment included urinary tract infections,
diarrhea and anemia.5 These data confirmed
the efficacy of tofacitinib, and have led to the
initiation of a large phase III program for this
drug in patients with RA.
Thus, the therapeutic options available to
rheumatologists could soon be increased by
the addition of new orally administered small
molecules, especially when one con­siders
that fostamatinib, another tyrosine kinase
inhibitor that targets spleen tyrosine kinase
(Syk), has also demonstrated con­siderable
S66 | JANUARY 2012
efficacy in RA.6 Naturally, important questions remain unanswered, particularly with
regard to long term safety in the ‘real world’,
where these agents will be used in combination with a multitude of other drugs that treat
comorbid diseases. Moreover, it remains to be
seen whether doctors and patients will prefer
60 tablets per month over a syringe that can
be self-­administered once a week, or even
once per month.
The introduction of new treatment modalities, such as tofacitinib and fostamatinib,
could help us to reach our current goal in
RA therapy, namely a state of remission,
particularly if a treat-to-target approach is
used.7 However, what does ‘remission’ really
mean? Last year, a group of American and
Euro­pean experts came together to explore
this important question and to generate a
stringent definition of this state of minimal
disease activity.8
On the basis of the evidence from large
trials that analyzed the development of
radiologically evident joint destruction
or loss of function, the panel representing
ACR–EULAR came up with a definition for
remission,8 in which a number of different
individual disease activity measures are considered together. Using this approach, remission is said to be achieved when the number
of tender and swollen joints, the level of CRP
(mg/dl) and the patient global assessment
score (0–10 scale) are all ≤1.8 An alternative
definition is an SDAI (simplified disease
activity index) of not more than 3.3.8 In clinical practice and in trials, these definitions
have been applied successfully.9,10 However,
to achieve a patient global assessment of ≤1
will be a challenge. Although the goals set
by ACR–EULAR are demanding, the 2011
remission criteria define clear targets for the
management of patients with RA and will
probably change our approach to treating this
debilitating disease, especially immediately
after onset.
As a result of key breakthroughs made in
2011, we are now able to define RA much
more effectively, we could soon have new
thera­peutics at our disposal and we have clear
goals to aim at when treating patients with
this previously devastating disease (Figure 1).
Hope­fully, the progress made last year will
improve patient care and lead to further
­evolution of the field in 2012 and beyond.
Department of Rheumatology and Clinical
Immunology, Charité-Universitätsmedizin
Berlin, Charitéplatz 1, 10117 Berlin, Germany.
[email protected]
Competing interests
The author declares associations with the following
companies: Abbott and Pfizer. See the article online
for full details of the relationships.
Aletaha, D. et al. 2010 rheumatoid arthritis
classification criteria: an American College
of Rheumatology/European League Against
Rheumatism collaborative initiative. Ann.
Rheum. Dis. 69, 1580–1588 (2010).
2. Padyukov, L. et al. Epidemiological Investigation
of Rheumatoid Arthritis (EIRA) study group. A
genome-wide association study suggests
contrasting associations in ACPA-positive
versus ACPA-negative rheumatoid arthritis.
Ann. Rheum. Dis. 70, 259–265 (2011).
3. Stahl, E. A. et al. Genome-wide association
study meta-analysis identifies seven new
rheumatoid arthritis risk loci. Nat. Genet.
42, 508–514 (2010).
4. Chatzidionysiou, K. et al. Highest clinical
effectiveness of rituximab in autoantibodypositive patients with rheumatoid arthritis and
in those for whom no more than one previous
TNF antagonist has failed: pooled data from
10 European registries. Ann. Rheum. Dis.
70, 1575–1580 (2011).
5. Fleischmann, R. et al. Phase 2B dose-ranging
study of the oral JAK inhibitor tofacitinib
(CP690,550) or adalimumab monotherapy
versus placebo in patients with active
rheumatoid arthritis with an inadequate
response to DMARDs. Arthritis Rheum.
6. Weinblatt, M. E. et al. An oral spleen tyrosine
kinase (Syk) inhibitor for rheumatoid arthritis.
N. Engl. J. Med. 363, 1303–1312 (2010).
7. Smolen, J. S. et al. T2T Expert Committee.
Treating rheumatoid arthritis to target:
recommendations of an international task
force. Ann. Rheum. Dis. 69, 631–637 (2010).
8. Felson, D. T. et al. American College of
Rheumatology/European League Against
Rheumatism provisional definition of remission
in rheumatoid arthritis for clinical trials. Ann.
Rheum. Dis. 70, 404–413 (2011).
9. Shahouri, S. H. et al. Remission of rheumatoid
arthritis in clinical practice. Application of the
American College of Rheumatology/European
League Against Rheumatism 2011 remission
criteria. Arthritis Rheum. 63, 3204–3215 (2011).
10. Klarenbeek, N. B. et al. Association with joint
damage and physical functioning of nine
composite indices and the 2011 ACR/EULAR
remission criteria in rheumatoid arthritis.
Ann. Rheum. Dis. 70, 1815–1821 (2011).
JIA IN 2011
New takes on categorization
and treatment
Alberto Martini
In 2011, new treatment recommendations for juvenile idiopathic arthritis
(JIA) were proposed, inroads were made towards understanding the
heterogeneity of this disease, and data were presented demonstrating
the potential efficacy of DMARD combination therapies for JIA treatment.
These advances hold promise for improved management of JIA in 2012
and beyond.
Martini, A. Nat. Rev. Rheumatol. 8, 67–68 (2012); published online 10 January 2012;
2011 has witnessed the publication of a
number of important pieces of information
that increase our understanding of juvenile idiopathic arthritis (JIA) hetero­geneity,
which guide treatment decisions and pave the
way for future studies regarding the potential efficacy of combination therapies with
inexpensive DMARDs. Here, I dis­cuss these
advances and their potential to improve the
management of JIA throughout the world.
JIA is not a defined disease, but rather
an exclusion diagnosis encompassing all
forms of arthritis with symptoms that onset
before the age of 16 years, persist for more
than 6 weeks, and for which the cause is
unknown. In the absence of defined pathogenic underpinnings, attempts have been
made to classify this collection of hetero­
geneous disorders into homogeneous,
mutually exclusive subgroups on the basis
of clinical and laboratory features.1 Some of
the official JIA categories developed by the
International League Against Rheumatism
(ILAR)1 seem to represent distinct disease
entities (systemic, polyarticular rheumatoid factor (RF) positive, enthesitis-related
arthritis and oligoarticular), whereas others
seem to define heterogeneous conditions
(polyarticular RF‑negative, psoriatic).
In Western countries most individuals
with JIA classified as oligoarticular belong to
a well-defined subset of patients with disease
that is charac­terized by several common features: asymmetric arthritis; early onset of
symptoms (before 6 years of age); female
predominance; the presence of antinuclear
anti­bodies (ANA); high risk of developing
chronic iridocyclitis; and associ­ation with
specific HLA alleles. However, on the basis of
a 2003 literature review,2 it was suggested that
patients with the same disease—ANA positive, early-onset oligo­articular JIA—could be
included into several different JIA categories:
oligo­articular, polyarticular RF‑negative
KEY ADVANCES IN MEDICINE or psoriatic. Therefore, patient age at onset
of symptoms, the presence of symmetric or
asymmetric arthritis, and ANA positivity
were proposed as potentially more suitable
criteria for disease classification than the
number of joints involved or the presence
of psoriasis.2
2011 has witnessed important
new advances in the understanding
and management of JIA...
A key paper published by Ravelli et al.3 in
2011 provides strong evidence to support
this hypothesis, particularly the potential use of ANA status in JIA categoriza­
tion. In a retro­spective study that enrolled
more than 900 patients with JIA, this group
compared the features of ANA-positive
and ANA-negative individuals. Impor­
tantly, the authors provi­ded a definition of
ANA ­positivity: antibody titers of ≥1:160
detected in ≥2 tests performed at least
3 months apart.3 Many other publications
have neglected to define ANA positivity,
which could provide a source of confusion.
Ravelli and colleagues3 found that ANApositive patients, despite being classified
under different ILAR-defined JIA categories,
were similar in terms of age at dis­ease onset,
female-to-male ratio, and fre­quency of
asymmetric arthritis and irido­cyclitis, independent of the number of joints involved or
the presence of psoriasis. Perti­nent in this
respect are the findings of Barnes and coworkers.4 These investi­gators studied gene
expression in peripheral blood mononuclear
cells of 104 patients with recent-onset JIA
(39 with oligo­articular JIA, 45 with polyarticular RF‑negative disease and 20 with syste­
mic JIA), as well as in cells from 56 healthy
controls. Their results reveal that a B‑cell
gene expression signature charac­terizes
patients with early-onset arthritis (≤6 years
of age), indepen­dent of the number of joints
involved.4 Of note, a different cluster of genes
related to cellular immunity and myeloid
cell lineages was expressed at higher levels
in patients with late-onset oligo­articular JIA
compared with individuals with early-onset
dis­ease.4 Thus, these data suggest that age at
onset of symptoms can also be of relevance
in unravel­ing disease heterogeneity among
patients with oligoarticular JIA.
Taken together, the findings I have des­
cribed strongly substantiate the hypo­thesis
that age at onset of disease and ANA positivity could be used to identify a homo­genous
subset of patients with JIA—patients who are
currently included in multiple different JIA
categories on the basis of the ILAR classification criteria. However, future studies need
to investigate whether, as might be expected,
patients with a B‑cell signature and early
disease onset are also ANA positive.
Over the past decade, dramatic improvements in the treatment of JIA have been
made. This progress has been made possible not only by the introduction of biologic
agents to the clinic, but also by the implementation of the so called ‘pediatric rule’5
by the FDA and the European Medicines
Agency (EMA). This rule states that when
the pharmaceutical industry attempts to register a new drug for any given adult dis­ease,
data must also be provided on the safety and
efficacy of that drug in ­children—provided
a pediatric form of the disease exist. This
stipulation has opened up the opportunity
for several randomized controlled trials
with new biologic therapies to be performed
in patients with JIA.
The rapid advances made in the treatment of JIA have presented new thera­peutic
options and made the decision-making processes more complex for rheumatologists.
Therefore, a major step forward was provided
last year by the publication of the American
College of Rheumatology (ACR)-endorsed
2011 recommendations for the treatment of
Key advances
■■ Age at disease onset and antinuclear
antibody status seem to define a
distinct category of juvenile idiopathic
arthritis (JIA)3
■■ New treatment recommendations for JIA
are an important tool,6 but will soon need
to be updated
■■ Combination DMARD therapy might be
more efficacious than methotrexate
alone,10 which could influence treatment
of JIA worldwide
JANUARY 2012 | S67
JIA.6 They represent the first official treatment recommendations for JIA to be based
on objective, validated methods, and have
already been reviewed in this journal.7 The
term ‘recommendations’ was used, instead of
‘guidelines’, to emphasize their nonprescriptive nature; the intended aim of the publi­
cation was to support, but not to dictate,
the individual physician’s decision. The
recommendations are based on a comprehensive list of clinical scenarios, in which
hypothetical patients are categorized on the
basis of different combinations of key clinical
parameters rele­vant to the decision-making
process, such as disease activity and potential
prognostic features.
The ACR recommendations are a very
useful tool, which can guide the treatment of
patients with JIA, but will soon require updating given the rapid evolution of this field.
Indeed, tocilizumab—an antibody against
the IL-6 receptor that has now been approved
for the treatment of systemic JIA by both the
FDA and the EMA—is not considered in
the recommendations. More­over, trials are
currently being performed in patients with
systemic JIA that aim to investigate the efficacy of two IL‑1 inhibitors, canakinumab (a
monoclonal anti-IL‑1 antibody) and rilonacept (the recombinant extracellular domain of
the human IL‑1 receptor); the results of these
studies will soon be available, and will presumably provide important new information
re­garding treatment of this disease.
Although the introduction of biologic
therapies has represented a major advance
in the treatment of JIA, the fact remains
that the high costs associated with provision of these agents can be afforded only by
the health­care systems of a limited number
of countries.8 Around the world, most children with JIA have to rely on much less
expensive drugs, such as methotrexate.
Therefore, data regarding the efficacy of
combination therapies comprising methotrexate and other synthetic DMARDs in
compari­son with metho­trexate alone would
be viewed with interest by rheuma­tologists
worldwide. However, in contrast to rheumatoid arthritis—a dis­ease in which many trials
have supported the over­all superiority of
combination DMARD therapy over metho­
trexate monotherapy9—data on combi­nation
DMARD therapy are lacking for JIA.
Important findings supporting the potential efficacy of combination DMARD therapy
in JIA were presented last year by Tynjälä
et al.10 The authors randomly assigned 59
patients with early poly­a rticular JIA to
three treatment cohorts: infliximab plus
S68 | JANUARY 2012
methotrexate; methotrexate alone; or metho­
trexate, sulfa­salazine and hydroxychloroquine in combination. The results of this
trial showed that infliximab plus methotrexate was su­perior to combination therapy and
strikingly superior to methotrexate alone. An
interest­ing aspect of this study was, therefore,
that synthetic DMARDs in combi­nation
seemed to be superior to metho­t rexate
treatment alone, although the difference
in effi­cacy was not statistically significant.
The study was, however, underpowered to
show a signifi­cant difference and the trend
in favor of combination therapy was consistent for all the endpoints tested. Thus, these
results strongly suggest that combination
therapies with inexpensive synthetic drugs
could prove superior to methotrexate alone
in future studies with appropriate sample
sizes. The funding of these potentially very
relevant investigator-initiated randomized
studies should represent a future priority for
in­ternational public funding bodies.
2011 has witnessed important new advances
in the understanding and manage­ment of JIA,
but much remains to be done. In the future,
more accurate classification of patients with
JIA will be possible and more effec­tive treatments will become avail­able, owing to the
integration of clinical data with the results of
research into dis­ease pathogenesis.
University of Genoa, Pediatria II e
Reumatologia, Istituto G. Gaslini, Largo G.
Gaslini 5, 16147 Genoa, Italy.
[email protected]
Competing interests
The author declares no competing interests.
Petty, R. E. et al. International League of
Associations for Rheumatology classification
of juvenile idiopathic arthritis: second revision,
Edmonton, 2001. J. Rheumatol. 31, 390–392
2. Martini, A. Are the number of joints involved
or the presence of psoriasis still useful tools
to identify homogeneous disease entities in
juvenile idiopathic arthritis? J. Rheumatol.
30, 1900–1903 (2003).
3. Ravelli, A. et al. Antinuclear antibody-positive
patients should be grouped as a separate
category in the classification of juvenile
idiopathic arthritis. Arthritis Rheum. 63,
267–275 (2011).
4. Barnes, M. G. et al. Biologic similarities based
on age at onset in oligoarticular and
polyarticular subtypes of juvenile idiopathic
arthritis. Arthritis Rheum. 62, 3249–3258
5. Hirschfeld, S. & Saint-Raymond, A. Pediatric
regulatory initiatives. Handb. Exp. Pharmacol.
205, 245–268 (2011).
6. Beukelman, T. et al. 2011 American College
of Rheumatology recommendations for the
treatment of juvenile idiopathic arthritis:
initiation and safety monitoring of therapeutic
agents for the treatment of arthritis and
systemic features. Arthritis Care Res.
(Hoboken) 63, 465–482 (2011).
7. Hashkes, P. J. Pediatric rheumatology:
strengths and challenges of a new guide for
treating JIA. Nat. Rev. Rheumatol. 7, 377–378
8. Sawhney, S. & Magalhães, C. S. Paediatric
rheumatology—a global perspective. Best
Pract. Res. Clin. Rheumatol. 20, 201–221
9. Ma, M. H., Kingsley, G. H. & Scott, D. L.
A systematic comparison of combination
DMARD therapy and tumour necrosis inhibitor
therapy with methotrexate in patients with early
rheumatoid arthritis. Rheumatology (Oxford)
49, 91–98 (2010).
10. Tynjälä, P. et al. Aggressive combination drug
therapy in very early polyarticular juvenile
idiopathic arthritis (ACUTE-JIA): a multicentre
randomised open-label clinical trial. Ann.
Rheum. Dis. 70, 1605–1612 (2011).
SLE IN 2011
Deciphering the role of NETs
and networks in SLE
Thomas Dörner
From neutrophil extracellular traps to genetic networks that underlie the
disease and new targeted therapies, important advances in 2011 improve
our understanding of the pathogenesis of systemic lupus erythematosus
and mark the beginning of our ability to treat it effectively.
Dörner, T. Nat. Rev. Rheumatol. 8, 68–70 (2012); published online 10 January 2012;
2011 was a year of achievements in systemic
lupus erythematosus (SLE)—for advances
in basic and clinical research, and the translation of these findings into daily clinical
practice. After a decade of failures of some
biologic therapies in SLE, the recog­nition of
particular cellular and cytokine pathways as
involved in the pathogenesis has initiated a
promising period of progress in this complex
disease. Many excellent contributions were
published in 2011, but a few studies deserve
emphasis for providing new perspectives.
Inflammatory response
Abnormal activation of adaptive
immunity with memory maintenance
Abnormal innate
immune activation
Subphenotype 1
■ HLA-DRB1*0301
■ Lupus nephritis
Subphenotype 2
■ Cumulative genetic risk
factor and anti-dsDNA
■ Young age at onset,
hematological disorders
and absence of oral ulcers
Subphenotype 3
■ Lack of known SLE
susceptibility genes
■ Malar or discoid
rash, photosensitivity,
serositis and
neurologic disorders
Plasmacytoid DC
HNP or
T cell
UV light
(Viral infections)
Lung manifestations
Vascular occlusions
B cell
Immune complexes
Complement consumption
Genetic risk
Plasma cell
Myeloid DC
Figure 1 | From NETs to disease networks in SLE. A proposed pathogenic model of SLE incorporating key findings from 2011. Firstly, three genetic
subphenotypes of SLE have been identified (left panel), which might be under different epigenetic control. Next, understanding of abnormal innate
and adaptive immunity in SLE (middle panel) has increased, in particular with the finding that NETosis of neutrophils leads to IFN‑α production by
pDCs. Finally, belimumab, which targets BAFF (a key cytokine in the adaptive immune response in SLE), has shown clinical efficacy. Abbreviations:
APC, antigen-presenting cell; TH1 cell, type 1 T helper cell; NET, neutrophil extracellular trap; DC, dendritic cell; SLE, systemic lupus erythematosus.
Despite the accepted roles of auto­
reactive lymphocytes in adaptive immunity, re­searchers are only just beginning to
understand the important early events of
immune activation within innate immunity.
Chronic activation of plasmacytoid dendritic cells (pDCs) leads to increased IFN‑α
production, which stimulates autoreactive
lympho­c ytes and reduces the activation
threshold of auto­reactive B cells. Crucially,
several recent papers have addressed neutrophil function in SLE. Increased neutrophil
turnover corre­lates with interferon levels,
suggesting a link between neutrophil activa­
tion and chronic pDC activation in SLE.1
Moreover, self nucleic acids (normally nonimmunogenic) incorporated within immune
complexes can trigger Toll-like receptor
(TLR)72 and TLR9 activation.3 In 2004, a
cell-death process distinct from necrosis and
apoptosis was identified,4 in which immunogenic self DNA–antimicrobial peptide complexes are released by dying neutrophils. This
process, ‘NETosis’, involves extrusion into the
extracellular space by activated neutro­phils of
large amounts of nuclear DNA, in the form
of web-like structures called neutrophil extracellular traps (NETs).4 NETs are abundantly
released by neutrophils in patients with SLE.
A key 2011 paper by Lande et al.5 has
addressed a number of puzzling issues of
KEY ADVANCES IN MEDICINE NETosis in SLE. The researchers found that
sera from patients with SLE (in contrast with
control sera) contained immune complexes
of self DNA and antimicrobial cationic peptides LL37 (also known as cathelicidin) and
human neutrophil peptide (HNP). These
immunogenic complexes were protected
from nuclease degradation, and interaction
with Fcγ receptor IIa (FcγRIIa, encoded by
FCGR2A) on pDCs led to their internaliza­
tion and, ultimately, to pDC activation.
More­over, autoantibodies against LL37 and
HNP enhanced NET formation and facilitated further pDC activation.5 These data
sup­port the notion that dysregulation of this
neutrophil-dependent pathway drives chronic
pDC activation and autoimmunity in SLE. An
important finding was that the composition of
the immune complexes—in particular LL37–
double-stranded (ds)DNA and LL37–HNP–
dsDNA complexes—determined their fate:
FcRγIIa-mediated endo­cytosis. Thus, pDCs
have a hitherto unknown ‘hidden’ receptor
specificity that is not essentially related to
speci­f ic immuno­receptor binding. These
findings provide insights into early immune
perturbations in SLE and deepen our understanding of SLE patho­genesis, as well as
challenging a key concept in SLE: the selective specificity of the auto­immune response
against certain self antigens.6
SLE is complex and heterogeneous.
Rheuma­tologists have learned that finetuning diagnostic approaches and therapeutic decisions can substantially influence
patient outcomes (exemplified by differentiating anti­phospholipid syndrome from SLE).
Genetic loci identified as risk loci for SLE susceptibility (HLA-DRB1, FCGR2A, PTPN22,
IRF5 and STAT4, among others) are mostly
related to immunological pathways (such
as antigen presentation and IFN signaling).
Nevertheless, whether patterns of multiple
risk alleles in patients with SLE constitute specific genotypes that relate to particular clinical manifestations has not been t­horoughly
studied; the direct genetic and epigenetic networks that underlie SLE pathogenesis remain
unclear (Figure 1).
In the second 2011 paper highlighted here,
Taylor et al.7 undertook a compre­hensive
analysis of 22 genetic variants implicated
in SLE pathogenesis. The most statistically
significant individual risk alleles associated with SLE were HLA-DR3-IRF5 and
FCGR2A-PXK (consistent with the findings
by Lande et al.,5 which demonstrated the
importance of FcγRIIa in pDC activation).
Importantly, in addition to the characterization of SLE susceptibility loci, the researchers
studied their relationship with disease charac­
teristics. An interesting categorization of SLE
JANUARY 2012 | S69
Key advances
■■ Neutrophil extracellular traps induce
innate immunity early in systemic lupus
erythematosus (SLE) and possibly
activate the adaptive immune response5
■■ Genetic studies identify certain SLE
subtypes and simultaneously indicate
distinct roles of epigenetic modulation7
■■ Approval of belimumab—the first
targeted therapy in SLE—translates
insights into abnormalities of the
immune system in SLE into the clinic9
subphenotypes emerged, on the basis of the
strength of association of each manifestation
with known genetic susceptibility loci, and
a genetic risk score. Thus, subphenotype 1
is closely associated with HLA-DRB1*0301
(DR3) and lupus nephritis, with ITGAM
alleles protecting against arthritis; subphenotype 2 is closely associated with cumulative
genetic risk factors and is characterized by
anti-dsDNA autoantibodies, immunological
abnormalities, young age at onset, hematological disorders and absence of oral ulcers;
and subphenotype 3—not associated with
any known SLE risk loci (perhaps reflecting
susceptibility to environmental or epigenetic
modulation)—is characterized by malar or
discoid rash, photosensitivity, serositis and
neurological disorders.
These interesting data highlight two major
aspects of SLE genetics: first, the existence
of distinct phenotypes based on different
ge­n­etics, each associated with particular
morbid­ity and mortality risks; the other,
that epi­genetic modification probably differs
between the hypothetical subphenotypes.
However, the data cannot be extrapolated to
different ethnic backgrounds; the participants
were almost exclusively of European ancestry.
Finally, in 2011 Navarra et al.9 reported the
phase III trial of belimumab, a new cytokinetargeted therapy for SLE—aimed, one could
say, at disrupting the pathogenic ‘net­work’
of the disease. A number of cytokine perturbations have previously been iden­tified
in patients with SLE, mainly comprising
increased levels of B‑cell activating factor
(BAFF; TNF superfamily member 13B,
also known as BLyS) and type I interferons.
Belimumab, which blocks soluble BAFF, is
a welcome step in improving SLE therapy
—now approved by the FDA and Euro­
pean Medical Agency, it is the first targeted
therapy directed at B cells in SLE, in a disease
known to involve hyper­active B lympho­cytes.
Importantly, the 2011 BLISS‑52 trial9 did
not study the efficacy of beli­mumab in only
severe SLE; its success in a hetero­geneous
S70 | JANUARY 2012
SLE population needed a large effects size
to differentiate the results from placebo.
Nonetheless, the drug was of insufficient
potency to show a strong effect early in
disease. That response to beli­mumab might
differ between the SLE sub­phenotypes discussed above is probable. As the frequency of
ANA-positivity was 92–95% among phase III
participants,9 compared with 66.7–74.3%
in a previous phase II trial,10 the benefit
of belimumab for ANA-negative patients
remains unclear. Furthermore, identifying
the specific profile of belimumab respon­ders
could improve understanding of the dis­
tinct immuno­pathologies of SLE sub­pheno­
types and lead to more effective therapeutic
de­cisions by tailoring therapy to each patient.
Overall, from NETs to disease networks,
studies published in 2011 collectively open
new avenues of investigation in SLE, including early innate activation paths, delineation
of subsets (with genetic profiles probably
complemented by epigenetic modifications) and the identification of therapeutic
responders to belimumab. Thus, basic and
clinical research support re-evaluation SLE
in terms of better defined subtypes.
Department of Medicine, Rheumatology and
Clinical Immunology, Charite Center 12, Charite
Universitätsmedizin Berlin and Deutsches
Rheumaforschungszentrum, Chariteplatz 01,
10098 Berlin, Germany.
[email protected]
Competing interests
The author declares no competing interests.
Craft, J. Dissecting the immune cell mayhem
that drives lupus pathogenesis. Sci. Transl.
Med. 3, 73ps9 (2011).
2. Rubtsov, A. V. Toll-like receptor 7 (TLR7)-driven
accumulation of a novel CD11c+ B‑cell
population is important for the development of
autoimmunity. Blood 118, 1305–1315 (2011).
3. Leadbetter, E. A. et al. Chromatin-IgG
complexes activate B cells by dual engagement
of IgM and Toll-like receptors. Nature 416,
603–607 (2002).
4. Brinkmann, V. et al. Neutrophil extracellular
traps kill bacteria. Science 303, 1532–1535
5. Lande, R. et al. Neutrophils activate
plasmacytoid dendritic cells by releasing
self‑DNA‑peptide complexes in systemic lupus
erythematosus. Sci. Transl. Med. 3, 73ra19
6. Dörner, T., Giesecke, C. & Lipsky, P. E.
Mechanisms of B cell autoimmunity in SLE.
Arthritis Res. Ther. 13, 243 (2011).
7. Taylor, K. E. et al. Risk alleles for systemic
lupus erythematosus in a large case–control
collection and associations with clinical
subphenotypes. PLoS Genetics 7, e1001311
8. Lindh, E. et al. AIRE regulates T‑cell‑independent
B‑cell responses through BAFF. Proc. Natl Acad.
Sci. USA 105, 18466–18471 (2008).
9. Navarra, S. V. et al. Efficacy and safety of
belimumab in patients with active systemic
lupus erythematosus: a randomised, placebocontrolled, phase 3 trial. Lancet 377, 721–731
10. Wallace, D. J. et al. A phase II, randomized,
double-blind, placebo-controlled, dose-ranging
study of belimumab in patients with active
systemic lupus erythematosus. Arthritis
Rheum. 61, 1168–1178 (2009).
OA IN 2011
Age-related OA—a concept
emerging from infancy?
Thomas Aigner and Wiltrud Richter
That primary osteoarthritis (OA) is an age-related disorder is undoubted,
but how aging contributes to OA is poorly understood. New insights
from 2011 offer potential explanations, novel models for study, and the
suggestion that a deeper understanding of what ‘aging’ actually is might
pave the way to everlasting joints.
Aigner, T. & Richter, W. Nat. Rev. Rheumatol. 8, 70–72 (2012); published online 10 January 2012;
Aging is the most prominent risk factor for
the initiation and progression of pri­mary
OA. Many explanations for this phe-no­
menon have been suggested over time, 1
without conclusion. In fact—although
the concepts underlying them might be
largely correct—the details of these opposing theories might have obscured the
true nature of what ‘aging’ is in relation
to OA.
The most longstanding theory of the
pathogenesis of primary OA (if one takes
the articular cartilage as the starting point)
involves the cumulative effects of continuous mechanical wear and tear on the
articu­lar cartilage matrix during a lifetime
of joint use (Box 1). A second theory is
related to well known age-related intrinsic changes in the extracellular matrix of
articular carti­lage, such as collagen network
stiffen­ing and the formation of advanced
glycation end products. A third, frequent
explanation for OA cartilage degeneration
is the mere loss of viable cells (due to apoptosis or any other mechanism of cell death),
as cellular replenish­ment does not occur in
articular cartilage.
We and others have suggested that OA
is a consequence of (premature) aging of
chondrocytes in joint tissues (cell senescence theory, Box 1). Although not dead,
senescent cells nevertheless lose their capac­
ity to sustain the cartilage matrix, which
can trigger an osteoarthritic degeneration
pathway.2–4 As we discuss below, chondrocyte behavior seems to change with increasing age: the cells respond less well to growth
factors, and anabolic activity declines. In
2011, Loeser and colleagues5 went some
way towards unpicking what OA‑related
cellular aging actually looks like at the
level of gene expression. Their interest­ing
functional genomic study targeted gene
expression alterations in a surgical meniscus destabilization mouse model of OA,
and examined test animals of different
ages. As well as disease-related changes, the
investigators also observed substantial differences in the injury-induced patterns of
expression between ‘young’ and ‘old’ mice,
in chondrocytes, meniscal cells and other
joint tissues.5 Genes involved in matrix production, immunity and defense were among
those affected, but our current knowledge of
chondrocyte cell biology—whether young
or aging—is not detailed enough to explain
the differences observed. Nonetheless, this
finding clearly shows that cartilage cells
react differently depending on their aging
status, a phenomenon also documented for
mesenchymal progenitor cells.6
Articular cartilage is especially vulnerable
to the effects of cellular aging and senescence. At skeletal maturity, sub­chondral
cal­c ification isolates the tissue from the
vas­cular system such that no new cells can
enter the articular cartilage. Chondrocytes
become postmitotic—they cease proliferation and are thus among the oldest cells of
the human body, able to survive for decades
without replacement. Furthermore, they
are constrained within the cartilage matrix,
unable to move. Even when chondro­cytes
start to divide again, as occurs in OA carti­
lage, cells from one site are, there­fore, unable
to replace damaged or absent cells else­
where, except for those literally adjacent.
One consequence of this postmitotic status
is that oxidatively damaged molecules (proteins, DNA, lipids) accumulate in articular
KEY ADVANCES IN MEDICINE Box 1 | Aging theories in the pathogenesis of OA
Continuous loading theory
Osteoarthritis (OA) results from the cumulative effects of continuous mechanical wear and tear
on cartilage; thus ‘aging’ is simply the accumulation of loading cycles over a lifetime of joint use.
Matrix pathobiochemistry theory
OA results from age-related changes in the extracellular matrix of articular cartilage:
■■ Collagen network stiffening due to increased covalent cross-linking
■■ Altered expression of aggrecan molecules that account for the enormous swelling capacity
of articular cartilage and, thus, its resistance to compression
■■ The formation of advanced glycation end products affects matrix integrity and chondrocyte
Progressive (apoptotic) cell loss theory
OA results from the loss of viable cells, due to apoptosis or any other mechanism of cell death,
at the beginning of and during the disease process. As chondrocytes are solely responsible
for replenishing the articular matrix, substantial loss of these cells results in the loss and
degeneration of the surrounding matrix.
Mitochondrial degeneration theory
OA results from the continuous degeneration of chondrocyte mitochondria that eventually leads
to energy failure as well as high oxidative stress in the cells.
Cell (pre)senescence theory
OA results form the inability of the cells to maintain tissue homeostasis (in particular after
insults of a mechanical or inflammatory nature) due to cellular degeneration and/or aging
caused in part by the accumulation of oxidatively-damaged molecules.
chondrocytes with aging, as these molecules
are not completely re-­synthesized within
the cells, as would occur during cell division. This accumulation of damaged molecules impairs the ability of chondrocytes
to remain functional and maintain tissue
homeostasis. More importantly, the sene­
scent cellular phenotype alters how cells
react to stimuli. Indeed, our conception of
the laws of cellular organization and function in relation to age is evolving rapidly in
the 21st century.7
Thus, the most important implications of
this emerging aging concept in OA development might be that the biology of cellular
degeneration is part of the basis for the initiation and progression of OA. Furthermore,
this knowledge emphasizes that whatever
treatment of OA we envisage, we must take
into account that we are dealing with aged
and senescent or pre-senescent cells that
no longer have the abilities of their juvenile counterparts to counteract mechanical, inflammatory, and/or other insults to
the tissue. This previously insufficiently
appreciated requirement might be one core
explanation why many promising therapeutic approaches, established and tested
in ‘juvenile’ cell lines and animal models,
have subsequently failed in aged human OA
cartilage in vivo.
Is the only way to investigate these agerelated effects to wait until individuals
age—which is rather a long time in the
con­text of most experiments—or can we
find quicker, informative model systems?
Key advances
■■ Gene expression changes in joint tissue
of older mice hint at how age and injury
interact in OA5
■■ Oxidative stress ages human chondrocytes
in vitro,8 which might provide a model for
studying age-related OA development
■■ Cartilage senescence is delayed by growth
inhibition,10 indicating that cellular ‘aging’
might be uncoupled from time itself
Taking chondrocytes from the replaced
joints of patients with OA is of limited
practical use; isolating diseased chondrocytes is a difficult endeavor with low
yields. More troublesome still is the suspicion that one might only sample the more
healthy subpopulation of cells and not the
really diseased ones, which would explain
the low yields of isolated cells per gram
of tissue (in fact, most OA chondrocytes
are lost during the isolation procedure).
Quicker in vitro model systems are clearly
required. Poten­tially help­ful, therefore, is
the 2011 report by Brandl and colleagues8
of an in vitro model of aging of chondrocytes. Chondrocytes are unique in their
avascular setting, and the effect of oxidative stress on these cells is not well characterized. Brandl et al.8 found that oxidative
stress accelerated cellular aging in chondrocytes—and can thus be used to artificially induce ‘age’ in cell culture—and that
senescent chondrocytes are less resistant
to oxidative stress. Similar strategies might
be adapted to in vivo models, evading the
JANUARY 2012 | S71
usually unfeasible requirement of waiting
for the natural aging of test animals.
Another potentially interesting in vivo
study system was also advanced in 2011,
and has furthered our understanding of
aging in cartilage. Baron and colleagues
have, 9 for several years now, studied
senescence in fetal growth plate cartilage,
previously finding it to be delayed as a consequence of hypothyroidism; once the deficiency is resolved, new, ‘catch-up’ growth
results. In 2011, the group showed that
cartilage senescence is delayed by general
growth inhi­bition;10 thus, the onset of senescence is not governed by time, but rather by
growth. These findings are further support
for the notion that a time-­dependent wear
process might be insufficient to explain the
onset of OA. Nevertheless, whether growthplate cartilage reflects substantial aspects
of OA chondrocyte senescence and, thus,
can contribute substantially to OA and
chondrocyte aging research remains to
be examined.
OA and OA research is increasingly leav­
ing behind the simple age-related fatalis­
tic concept of chronic wear and tear, that
is, matrix and cell degeneration. The con­
cept of aging as a phenomenon with its
own biology needs to be understood by the
OA community as a new, innovative and
excit­ing area of chondrocyte and carti­lage
research. Though recent advances leave one
with the impression that our understanding in this regard remains in its infancy,
the emerging data might help to push OA
research from a niche of orthopedic interests to a highly topical research area: what
human aging means, in terms of biological processes, and what we can do about
it. OA research might merge with the everexpanding search for the Holy Grail, or to
speak with a German tongue, the ‘Jung­
brunnen’ (fountain of youth). In the case
of OA research and patient management,
the result would be ever-functioning,
painless joints.
Institute of Pathology, Medical Center Coburg,
Ketschendorferstraße 33, 96450 Coburg,
Germany (T. Aigner). Research Centre for
Experimental Orthopedics, Heidelberg
University Hospital, Schlierbacher Landstraße
200a, 69118 Heidelberg, Germany (W. Richter).
Correspondence to: T. Aigner
[email protected]
We acknowledge the support of the DFG for our work
(grant numbers Ai 20/7‑1,2 and Ri 707/7-1).
Competing interests
The authors declare no competing interests.
S72 | JANUARY 2012
Aigner, T., Rose, J., Martin, J. & Buckwalter, J.
Aging theories of primary osteoarthritis:
from epidemiology to molecular biology.
Rejuvenation Res. 7, 134–145 (2004).
Aigner, T., Söder, S., Gebhard, P. M.,
McAlinden, A. & Haag, J. Mechanisms
of disease: role of chondrocytes in the
pathogenesis of osteoarthritis—structure,
chaos and senescence. Nat. Clin. Pract.
Rheumatol. 3, 391–399 (2007).
Martin, J. A., Ellerbroek, S. M. &
Buckwalter, J. A. Age-related decline in
chondrocyte response to insulin-like growth
factor‑I: the role of growth factor binding
proteins. J. Orthop. Res. 15, 491–498
Loeser, R. F. Aging and osteoarthritis: the role
of chondrocyte senescence and aging changes
in the cartilage matrix. Osteoarthritis Cartilage
17, 971–979 (2009).
Loeser, R. F. et al. Microarray analysis reveals agerelated differences in gene expression during the
development of osteoarthritis in mice. Arthritis
6. Dexheimer, V., Mueller, S., Braatz, F. &
Richter, W. Reduced reactivation from
dormancy but maintained lineage choice
of human mesenchymal stem cells with donor
age. PLoS ONE 6, e22980 (2011).
7. Kirkwood, T. B. Understanding the odd science
of aging. Cell 120, 437–447 (2005).
8. Brandl, A. et al. Oxidative stress induces
senescence in chondrocytes. J. Orthop. Res.
29, 1114–1120 (2011).
9. Lui, J. C., Nilsson, O. & Baron, J. Growth plate
senescence and catch-up growth. Endocr. Dev.
21, 23–29 (2011).
10. Forcinito, P. et al. Growth-inhibiting conditions
slow growth plate senescence. J. Endocrinol.
208, 59–67 (2011).
SSC IN 2011
From mechanisms to medicines
Luc Mouthon
Findings from ongoing studies of imatinib in systemic sclerosis (SSc)
were eagerly awaited in 2011, but results from these clinical trials have
so far been disappointing. However, progress in the understanding of the
mechanisms that underlie SSc pathogenesis could provide clues to novel
targets for 2012 and beyond.
Mouthon, L. Nat. Rev. Rheumatol. 8, 72–74 (2012); published online 10 January 2012; doi.10.1038/
Systemic sclerosis (SSc) is a connective
tissue disorder characterized by excessive collagen deposition in the dermis and
inter­nal organs, vascular hyper-reactivity
and vascular obliteration. Tissue fibrosis
results from the increased release of extracellular matrix from aberrantly activated
fibroblasts.1 This accumulating extra­cellular
matrix disrupts the physiological tissue
structure, leading to organ dys­f unction,
which contributes to the high morbidity
and increased mortality in patients with
SSc. However, the mechanisms under­
lying pathological fibroblast activation are
incompletely understood. Transforming
growth factor β (TGF‑β) seems, however,
to be a master regulator of both physio­
logical and pathological matrix remodeling,
and might be responsible for maintaining
the activated fibroblast phenotype in SSc.1
Although a number of therapeutic trials
have been conducted with antifibrotic
agents (including d‑penicillamine, relaxin,
IFN‑α and IFN‑γ) in patients with SSc,
none had a major therapeutic effect. Here,
I will highlight the results of recent clinical studies of imatinib (a new antifibrotic
agent) in patients with SSc and the latest
advances in understanding the mechanisms
of fibrosis in SSc. 2011 has been a year of
progress, but will new mechanistic insights
lead to new medicines for SSc?
Understanding of fibrosis in SSc has
already led to trials of new therapies that
might control the disease. An example
being imatinib mesylate, a tyrosine kinase
inhibi­tor already approved for the treatment of chronic myelogenous leukemia.
Imatinib binds to c‑Abl (an important
down­stream signaling molecule of TGF‑β)
and blocks its tyrosine kinase activity,2 as
well as interfering with platelet-derived
growth factor signaling. In an in vitro
model of bleomycin-­induced pulmonary
fibrosis, c‑Abl inhibition by imatinib prevented TGF‑β-induced extracellular matrix
component gene expression, transformation and proliferation of fibroblasts.3 Similar
results were seen in mouse models of dermal
fibrosis in SSc.4 These interesting results led
a number of investigators to launch clinical
studies to examine the effectiveness of imatinib as an SSc therapy, the findings of which
were eagerly anticipated. Of the six trials
testing the efficacy of imatinib registered
on, four have been completed; findings from three of these studies
were published in 2011, while the results of
the fourth are not yet available. Moreover,
two other studies of different tyrosine kinase
inhibitors are ongoing in patients with SSc:
one open-label phase IIa study using nilotinib and another open-label phase I/II study
using dasatinib.
…will new mechanistic
insights lead to new medicines
for SSc?
Results from the imatinib trials published
in 2011 were somewhat disappointing in
terms of the tolerance and efficacy of the
drug. Spiera et al.5 conducted a phase IIa,
open-label, single-arm clinical trial in 30
patients with diffuse SSc who were treated
with imatinib (400 mg per day) over
12 months. Mean Rodnan skin score (MRSS)
decreased by 6.6 points (or by 22.4%) at
12 months (P = 0.001).5 Pulmonary function was also tested, with forced vital capa­
city (FVC) improving by 6.4% predicted
(P = 0.008), whereas the diffusion capacity
remained stable.5 Six patients withdrew from
the study, two because of non­compliance,
four because of adverse events (including
one who died after 11 months).5 In total, 171
adverse events of various grades were identified, the most common being edema, which
was observed in 80% of patients. 24 serious
adverse events were identified, two of which
were attributed to the study medication.5 The
authors concluded that imatinib has acceptable safety and toler­ability, and suggest it has
potential efficacy for cutaneous and pulmonary manifestations of SSc.5 However, other
studies did not replicate these findings.
In their phase I/IIa pilot study, Khanna
and colleagues6 recruited 20 patients with
SSc-related interstitial lung disease (ILD)
who had FVC <85% predicted, dyspnea
on exertion and presence of ground glass
appearance on high-resolution CT of the
lung. Patients received imatinib therapy (up
to 600 mg per day) for 1 year. 12 patients
completed the study, seven discontinued
due to adverse events and one was lost to
follow-up. Although treatment with imatinib showed a trend towards an improvement of FVC (by 1.74%; P >0.05) and MRSS
(of 3.9 units; P <0.001), use of high-dose
daily (600 mg per day) imatinib was associated with a large number of adverse events
(including edema and fatigue) in this
study.6 Furthermore, Pope et al.7 performed
a 6‑month double-blinded pilot study of
imatinib in patients with active diffuse SSc
KEY ADVANCES IN MEDICINE who had at least one of: worsening skin
score (measured by the MRSS); tendon
fric­tion rubs; and/or increased erythrocyte
sedi­mentation rate due to the active dis­
ease. The researchers planned to recruit 20
patients to their study and stratify them by
cur­rent use of methotrexate. However, after
enrolling only 10 patients (nine on imatinib,
one placebo), the authors found poor tolerability and high numbers of adverse events
with imatinib (200 mg twice a day) and
enrollment was discontinued.7 Most adverse
effects (such as fluid retention, weakness,
nausea and vomiting) occurred within the
first week of treatment and even reintroduction at a lower dose (200 mg daily) was
poorly tolerated.7 Two participants were
hospitalized due to these effects (one with
syncope, and the other with nausea, edema
and fatigue). The authors concluded that
imatinib was poorly tolerated and the study
was unfortunately insufficiently powered to
give conclusions on drug efficacy.7
Taken together, the results of the above
studies provide evidence that imatinib at
doses of 400–600 mg per day is poorly tolerated in patients with SSc. No conclusion can,
however, be made on drug efficacy because
of the heterogeneity, limited power and
nonrandomized nature of the above studies,
although a trend towards an improvement of
FVC and MRSS was observed in imatinibtreated patients. Because of poor tolerance,
whether testing of imatinib for SSc will
progress to large prospective randomized
trials is unknown, as is whether imatinib
warrants further study.
Although the initial imatinib trial results
were unsatisfying, recent advances in the
understanding of fibrotic mechanisms have
enabled the identification of a wide range
of potential biologic agents with which to
disrupt the development of fibrosis in SSc.1
These approaches include neutralizing the
principal extracellular signals that activate
Key advances
■■ The results of three studies show that
imatinib is poorly tolerated in patients
with SSc and do not enable clinicians
to draw conclusions on drug efficacy5–7
■■ Tissue factor—the primary in vivo
initiator of coagulation—interacts with
endothelin‑1 signaling in the activation
of myofibroblasts from patients with SSc8
■■ 5-HT–5-HT2B receptor signaling links
vascular damage and platelet activation
to tissue remodeling in SSc and could
be a novel therapeutic target to treat
fibrotic diseases9
fibroblasts and small molecules to disrupt
intracellular signaling pathways linked to
fibrosis.1 Progress in understanding fibrosis in SSc has been ongoing in 2011 and I
will highlight the extrinsic coagulation
system8 and platelet activation9 as potential
new targets for reducing the risk of fibrotic
complications in SSc.
In 2011, Chrysanthopoulou et al.8 found
that myofibroblasts cultured from the colon
of patients with SSc and gastrointestinal
symptoms expressed increased levels of
tissue factor, the primary in vivo initia­tor
of the coagulation system. Upregula­tion of
tissue factor led to thrombin generation
and subsequent production of connective
tissue growth factor (CTGF) and collagen
via protease-activated receptor 1 signaling.8
Further­more, thrombin induces endo­thelin 1
receptor A expression, while endo­thelin‑1
signaling is implicated in the enhanced
expression of tissue factor in myo­fibroblasts
from patients with SSc. 8 These findings
suggest that blocking the extrinsic coagulation system could be a potentially novel
approach to reduce the risk of fibrotic
compli­cations in SSc. Moreover, the essential
role of the tissue factor—­thrombin—CTGF
pathway in the fibrotic manifestations of SSc
has been demonstrated, as well as an interplay between the coagulation cascade and
endothelin‑1 signaling in the activation of
myofibroblasts in SSc.8
Results from the imatinib
trials published in 2011 were
somewhat disappointing…
Existing evidence indicates that platelets have a role in SSc. The levels of circulating platelet aggregates and of several
platelet-derived molecules (such as
β‑thrombomodulin and platelet-derived
growth factor) are elevated in patients with
SSc.10 However, the role of platelet activation
in the pathogenesis of SSc and other fibro­
tic diseases has not yet been established.
In 2011, Dees and colleagues9 showed that
serotonin (5-­hydroxytryptamine [5-HT])
stored in platelets strongly induces synthesis of extracellular matrix components
in inter­stitial fibroblasts via activation of
5‑HT2B receptors in a TGF‑β–dependent
manner. Dermal fibrosis was reduced in
5‑Htr2b–/– mice using both inducible and
genetic models of fibrosis.9 Pharmacological
inactivation of 5‑HT2B receptors also effectively prevented the onset of experimental
JANUARY 2012 | S73
fibrosis and ameliorated established fibrosis.9 Consistent with these findings, mice
deficient in tryptophan hydroxylase 1—the
rate-limiting enzyme for 5‑HT production
outside the central nervous system—had
reduced experimental skin fibrosis.9 These
findings, together with recent results from
5‑HT receptor antagonists in experi­mental
pulmonary or liver fibrosis, suggest that
5‑HT–5-HT2B receptor signaling links
vascular damage and platelet activation to
tissue remodeling, and highlight the 5‑HT2B
receptor as a novel therapeutic target to treat
fibrotic diseases.
In summary, blocking extrinsic coagulation system and/or platelet activation might
represent promising future strategies for SSc
treatment. From mechanisms to new medicines, research in 2011 has made tenta­tive
steps to providing an effective targeted
therapy for SSc. Although the efficacy of
imatinib has not yet been confirmed,
mecha­n istic insights into the develop­
ment of fibrosis in SSc have provided vital
clues for the development of new drugs for
this challenging disease and trials of new
therapies are ongoing.
Université Paris Descartes, Faculté de
Médecine, Service de Médecine Interne,
Hôpital Cochin, 27 rue du Faubourg Saint
Jacques, 75014 Paris, France.
[email protected]
Competing interests
The author declares associations with the following
companies: Actelion, GlaxoSmithKline, Lilly, Pfizer.
See the article online for full details of the
Bhattacharyya, S., Wei, J. & Varga, J.
Understanding fibrosis in systemic sclerosis:
shifting paradigms, emerging opportunities.
Nat. Rev. Rheumatol.
Distler, J. H. & Distler, O. Imatinib as a novel
therapeutic approach for fibrotic disorders.
Rheumatology (Oxford) 48, 2–4 (2009).
Aono, Y. et al. Imatinib as a novel antifibrotic
agent in bleomycin-induced pulmonary fibrosis
in mice. Am. J. Respir. Crit. Care Med. 171,
1279–1285 (2005).
Distler, J. H. et al. Imatinib mesylate reduces
production of extracellular matrix and
prevents development of experimental dermal
fibrosis. Arthritis Rheum. 56, 311–322
Spiera, R. F. et al. Imatinib mesylate
(Gleevec) in the treatment of diffuse
cutaneous systemic sclerosis: results of a
1‑year, phase IIa, single-arm, open-label
clinical trial. Ann. Rheum. Dis. 70, 1003–1009
Khanna, D. et al. A one-year, phase I/IIa,
open-label pilot trial of imatinib mesylate in the
treatment of systemic sclerosis-associated
active interstitial lung disease. Arthritis Rheum.
63, 3540–3546 (2011).
S74 | JANUARY 2012
Pope, J. et al. Imatinib in active diffuse
cutaneous systemic sclerosis: results of a
six-month, randomized, double-blind, placebocontrolled, proof‑of‑concept pilot study at a
single center. Arthritis Rheum. 63, 3547–3551
Chrysanthopoulou, A. et al. Tissue factorthrombin signaling enhances the fibrotic
activity of myofibroblasts in systemic sclerosis
through up-regulation of endothelin receptor A.
Arthritis Rheum. 63, 3586–3597 (2011).
9. Dees, C. et al. Platelet-derived serotonin links
vascular disease and tissue fibrosis. J. Exp.
Med. 208, 961–972 (2011).
10. Postlethwaite, A. E. & Chiang, T. M. Platelet
contributions to the pathogenesis of systemic
sclerosis. Curr. Opin. Rheumatol. 19, 574–579
The renaissance of granulomatous
inflammation in AAV
Stephan D. Gadola and Wolfgang L. Gross
In 2011, the year that subtypes of ANCA-associated vasculitis (AAV) were
officially renamed according to key pathological characteristics, important
progress was made not only in differentiating these subtypes, but also in
understanding—and treating—their eponymous manifestations.
Gadola, S. D. & Gross, W. L. Nat. Rev. Rheumatol. 8, 74–76 (2012); published online 10 January 2012;
corrected online 24 January 2012; doi:10.1038/nrrheum.2011.218
Granulomatosis with polyangiitis (GPA, formerly Wegener´s granulomatosis), ChurgStrauss syndrome (CSS) and microscopic
polyangiitis (MPA) constitute the antineutrophil cytoplasmic antibody (ANCA)associated vasculitides (AAV), forms of
primary small-vessel vasculitis associated
with ANCA seropositivity. Owing to the
rarity of their disease, patients with dif­ferent
AAV are often pooled in clinical trials. Dis­
ease evolution, organ involvement, prog­
nosis, and other parameters , however, dif­fer
substantially between GPA, CSS and MPA,
although the mechanistic bases for these
differences remain largely unexplored. To
unravel them, effective classification of
patients with AAV is essential, beginning
with defining—and naming—the subtypes.
2011 saw the launch of an international
effort among vasculitis experts, including
rheumatologists, nephrologists, pathologists
and others, to replace honorific eponyms of
Key advances
■■ A new artificial neural network helps
to differentiate subtypes of ANCAassociated vasculitis, including
granulomatosis with polyangiitis (GPA)2
■■ The effects of rituximab on granulomatous
and vasculitic manifestations of GPA
offer hope for patients with refractory
disease, plus insights into the pathology
of granulomatous lesions7
■■ Autoantibodies with transmembrane
protein targets are found—and
generated—in GPA granulomas, but are
undetectable in plasma9
diseases with names that highlight diseasespecific characteristics.1 GPA is in fact the test
case for this initiative; its new name underlines the clinical and pathological similarities
between GPA and MPA with regard to smallcaliber vessel involvement, while emphasizing the importance of granulomatous
inflammation as a distinctive feature of GPA
(Figure 1a). Indeed, since the introduction
of effective immunosuppressive protocols
that control vasculitic manifestations of GPA
(and other AAV), clinicians have increasingly
faced the challenges posed by granulomatous
manifestations (for which effective treatments have lagged behind). Differentiating
GPA and MPA, however, is sometimes difficult (biopsy proof of GPA is required, and
established diagnostic criteria are lacking),
especially when inter­disciplinary care cannot
be provided. Surrogate markers of granulomatous inflammation have been proposed to
dis­tinguish GPA, but are, as yet, unvalidated.
Linder and colleagues2 stepped into this
breach in 2011, applying an artificial neural
net­work (ANN) approach to generate and
validate improved criteria for distinguishing
GPA and MPA. The ability of ANNs to find
patterns in complex data sets has previously
been used to differen­tiate GPA and CSS.3
In training the ANN, Linder et al.2 identified four potentially rele­vant parameters
—involvement of the nose, sinuses, or ears,
and presence of pulmonary nodules—to help
in differen­tiating between GPA and MPA.
Indeed, when these cri­teria were applied as
the only input neurons, the ANN correctly
differentiated cases of GPA and MPA with an
accuracy of 94.3%,2 a substantial improvement over conventional classification criteria.
Importantly, the authors validated the ANN
in two independent cohorts of patients with
GPA (n = 46) and MPA (n = 21) from a single
center.2 This study, therefore, confirms that
certain surrogate markers can be used in differential diagnosis of GPA, while also indica­
ting that ANNs might be useful to distinguish
AAV subtypes in clinical trials.
ANN-type approaches to the classification
of AAV might be enhanced by the inclu­
sion of demographic and genetic markers.
It is now known, for example, that MPA is
much more common than GPA in Japan,
whereas GPA is the predominant AAV in
the UK.4 Further­more, a genetic associ­ation
study in German and British patients with
AAV found HLA-DPB1*0401 to be a strong
risk factor for GPA, but not MPA or CSS,5
suggest­i ng that HLA-DPB1*0401 might
contri­bute to the granulomatous aspect of
GPA. The Euro­pean Vasculitis Study Group
has defined two pheno­types of GPA—the
rare, local­ized phenotype that presents with
granuloma­tous tissue inflammation but
without clinical signs of vasculitis, and the
more frequent general­ized phenotype of
GPA, which features granulo­matous tissue
inflammation alongside systemic vasculitis.
Localized disease was thought to be an essentially harmless variant of GPA, but recent
studies have shown that granulomatous
manifestations can indeed cause significant
morbidity and even death. As these manifestations tend to be more refractory to treatment than vasculitic aspects of GPA, the need
to understand and treat them is pressing.6,7
Now, Holle and colleagues7 have com­pared
the efficacy of the B‑cell depleting anti­body
rituximab for the treatment of granulo­ma­
tous and vasculitic manifestations of GPA.
Their study, in 59 patients with refractory
GPA, included a high proportion of patients
with orbi­tal masses and pachymeningitis.7
Com­plete or partial remission was achieved
for 90% of refractory vasculitic but only
58% of granulomatous manifestations, and,
interest­ingly, rituximab was more effec­tive
for pulmonary masses (83.4% responded,
with 16.7% achieving complete remission)
compared with either orbital masses (44.4%
responded but none achieved complete
remission) or pachymeningitis (50.0% and
8.3%, respectively). The rationale for using
rituximab in AAV is to eliminate ANCAproducing B cells, and is based on a large body
of evidence indicating a key role for ANCA
in AAV small-vessel inflammation. Data
published in 20108 added to this rationale by
Giant cells
Fibrocyte rich
pannus-like tissue
Epitheloid cells
Destruction of
bone and cartilage
Figure 1 | Clinical burden of granulomatous manifestations in GPA. a | Left panel: MRI showing
retro-orbital ‘granuloma’ (arrowhead) leading to vision loss; middle panel: photograph and CT scan
showing severe bone and cartilage destruction, with saddle nose, loss of inner nose, and fistula
between the orbit and cavum nasi (arrowheads); right panel: large septal defect (circled) with
granulomatous ‘pannus’ (arrowhead). b | Histopathology of GPA. Left panel: sinonasal mucosa
(adjacent to area of geographic necrosis, not shown) showing giant cells (arrowheads), ELS
(circled) and eroded bone (B) with palisading fibrocytes (crosses). Right panel: key pathological
features. B‑cell-rich ELS resemble germinal centers, and B cells isolated from them can produce
autoantibodies.9 Abbreviations: ELS, ectopic lymphoid structures; GPA, granulomatosis with
polyangiitis; PMN, polymorphonuclear leukocytes. Permission to use her image was obtained from
the patient depicted. Photograph and CT scan reproduced from Aries, P. M. & Both, M. N. Engl.
J. Med. 352, 392 (2005), © Massachusetts Medical Society; histopathology slide reproduced
from Mueller, A. et al. Rheumatology (Oxford) 47, 1111–1113 (2008), with permission.
showing that ANCAs stimulate neutro­phils
to release B‑cell activating factor (BAFF, also
known as BLyS and TNFSF13B), which might
perpetu­ate ANCA production by increas­ing
the survival of auto­reactive B cells. Never­
theless, ANCA and B-cell involvement in
the granuloma­tous in­f lammation of GPA
remains largely unexplored.
Although consistent with a possible partial
role for B cells in granulomatous inflammation, the findings of Holle et al.7 indicate that
B‑cell-independent mechanisms are at the
heart of granuloma formation. In fact,
the granulomatous lesions of GPA har­bor
B‑cell-rich germinal center-like lymp­
hoid structures, and have long been suspected to be the very sites where pathogenic
­auto­antibody-secreting B cells arise. In a tour
de force, Thurner and colleagues9 provided
in 2011 the first evidence for the existence
of auto­antibody-secreting B cells in GPA
granulomas. Using laser micro­dissection,
they isolated single B cells from GPA granulo­
mata, cloned their immunoglobulin genes,
reconstituted the corresponding antibodies
and, finally, tested their binding to human
proteins. ‘Disappointingly’ none bound
to proteinase 3 (one of the principle auto­
antigen targets of ANCAs), perhaps owing
to the small number of antibodies tested.
Two transmembrane proteins, the lysosomal protein TMEM9B and cell-surface
expressed TM4SF2, were, however, identified as antigens for these auto­antibodies.9
Interestingly, two different anti­bodies from
the same granuloma recognized TMEM9B,
whereas no circulating antibodies against the
protein could be detected in the corresponding patient plasma. Thus, this study9 not only
helps to elucidate the role of the granuloma
in autoantibody formation (Figure 1b), but
also makes a strong point that the search for
new autoantibodies in inflammatory diseases
should encompass tertiary lymphoid structures in inflamed tissues, and not be restricted
to plasma. Furthermore, the findings clearly
suggest that ANCA-secreting B cells might
well be present in granulomatous tissue even
when they are not detectable in plasma (that
is, in ANCA-negative patients with GPA).
JANUARY 2012 | S75
Elucidating the role of B cells in granulomatous tissue injury would not only underpin the rationale for using B‑cell depletion
in granulomatous complications of GPA,
but also inspire new B‑cell directed drugs.
Ulti­mately, however, we will need to understand why B‑cell-rich germinal center-like
structures arise within GPA lesions, and how
B cells are (dys)regulated by the local innate
immune response. Promisingly, several genes
involved in innate immunity have recently
been implicated in GPA.10 Only with more
such studies will we begin to get to grips
with the complex pathophysiology of this
fascinating disease—and the true meaning
of its name…
Department of Rheumatology, University
Hospital of Southampton NHS Foundation
Trust, University of Southampton, Southampton
SO16 6YD, UK (S. D. Gadola). Department of
Rheumatology, University Hospital of
Schleswig-Holstein, Campus Lübeck, D‑24576
Bad Bramstedt, Germany (W. L. Gross).
Correspondence to: W. L. Gross
[email protected]
Competing interests
The authors declare no competing interests.
Falk, R. J. et al. Granulomatosis with
polyangiitis (Wegener’s): an alternative name
for Wegener’s granulomatosis. Arthritis Rheum.
63, 863–864 (2011).
Linder, R. et al. Differentiation between
Wegener’s granulomatosis and microscopic
polyangiitis by an artificial neural network and
by traditional methods. J. Rheumatol 38,
1039–1047 (2011).
Schmitt, W. H., Linder, R., Reinhold-Keller, E.
& Gross, W. L. Improved differentiation between
Churg-Strauss syndrome and Wegener’s
granulomatosis by an artificial neural network.
Arthritis Rheum. 44, 1887–1896 (2001).
Fujimoto, S. et al. Comparison of the
epidemiology of anti-neutrophil cytoplasmic
antibody-associated vasculitis between Japan
and the UK. Rheumatology (Oxford) 50,
1916–1920 (2011).
Arning, L. et al. Are there specific genetic
risk factors for the different forms of ANCAassociated vasculitis? Ann. Rheum. Dis.
70, 707–708 (2010).
6. Holle, J. U. et al. Prospective long-term
follow-up of patients with localised Wegener’s
granulomatosis: does it occur as persistent
disease stage? Ann. Rheum. Dis. 69,
1934–1939 (2010).
7. Holle, J. U. et al. Rituximab for refractory
granulomatosis with polyangiitis (Wegener’s
granulomatosis): comparison of efficacy
in granulomatous versus vasculitic
manifestations. Ann. Rheum. Dis. http://
8. Holden, N. J. et al. ANCA-stimulated neutrophils
release BLyS and promote B cell survival:
a clinically relevant cellular process. Ann.
Rheum. Dis. 70, 2229–2233 (2010).
9. Thurner, L. et al. Wegener’s granuloma harbour B
lymphocytes with specificities against a
proinflammatory transmembrane protein and
a tetraspanin. J. Autoimmun. 36, 87–90 (2011).
10. Laudien, N. et al. Molecular signatures of a
disturbed nasal barrier function in the primary
tissue of Wegener’s granulomatosis. Mucosal
Immunol. 4, 564–573 (2011).
Don’t miss out!
Sign up to free e-ALERTS and be notified when new articles are published
in Nature Reviews Rheumatology. Research Highlights, News & Views, Reviews,
and Perspectives—you won’t miss a thing!
S76 | JANUARY 2012
Redefining the therapeutic landscape for CRPC
Carmel Pezaro and Gerhardt Attard
2011 was a breakthrough year for the treatment of castration-resistant prostate cancer. The encouraging results
of two large clinical trials were reported, as well as data identifying a number of promising new therapeutic
targets. Bone-modulating agents continued to show potential for the prevention of skeletal events.
A number of significant breakthroughs
in the field of advanced prostate cancer
were reported this year, both clinically and
in translational and scientific research.
Recently reported analy­ses of the phase III
studies of abiraterone acetate and MDV3100
provide irrefutable evidence that the androgen receptor (AR) is a critical target in
castration-­resistant prostate cancer (CRPC).1
These agents entered clinical development
5–6 years ago and proceeded to large (>1,000
patients) phase III trials in docetaxel-treated
patients.2 MDV3100 is a novel potent AR
antagonist, and abiraterone acetate specifically inhibits CYP17A1, a key enzyme in the
testosterone bio­synthesis pathway that converts adrenal and potentially intratumoral
steroid precursors into androgens.
Interim analysis of the abiraterone acetate
trial revealed that abiraterone acetate
plus prednisone improved survival by
3.9 months compared to placebo plus prednisone (HR 0.65; P <0.001).1 Final analysis
after 775 events showed an overall survival benefit of 4.6 months for abiraterone
acetate (HR 0.74; P <0.0001).3 The drug was
well tolerated, with most reported adverse
effects relating to accumulation of steroid
precursors with potent mineralocorticoid
properties upstream of CYP17A1. 55% of
patients who received abiraterone acetate
experienced toxicity due to mineralo­
corticoid excess, and although few required
a dose reduction, 3.8% of patients required
increased doses of prednisone to suppress
adrenocorticotropic hormone. 1.6% of
patients treated with abiraterone acetate
and 1.8% of those who received placebo
were administered the mineralocorticoidreceptor antagonist eplerenone to manage
adverse effects.1
The outcome of interim analysis of the
MDV3100 phase III AFFIRM study was
Pezaro, C. & Attard, G. Nat. Rev. Urol. 9, 63–64 (2012); published online 17 January 2012; doi:10.1038/nrurol.2011.235
recently made public in a press release
from the manufacturers. Administration
of MDV3100 to docetaxel-treated patients
improved survival by 4.8 months compared
to placebo (HR 0.63; P <0.0001). Unblinding
of the study in order to offer MDV3100
treatment to patients in the placebo group
is expected in the near future.
Both abiraterone acetate and MDV3100
are currently undergoing evalua­t ion in
chemotherapy-naive patients ( identifiers NCT00887198 and
NCT01212991, respectively) and predictably, multiple new AR antagonists and
CYP17A1 inhibitors are under clinical
investigation.2 How best to select between
new thera­p ies for individual patients
is now the subject of further research.
Combined treatment with AR antagonists
and CYP17A1 inhibitors may offer greater
benefit but requires clinical evaluation.
Unfortunately, resistance to drugs targeting the AR inevitably occurs. In the majority of patients, progression on abiraterone
acetate or MDV3100 is associated with a
rise in PSA (the product of an AR‑regulated
gene), suggesting reactivation of steroid
signaling.2 Bidirectional crosstalk between
the serine/threonine protein kinase AKT
and the AR has been suggested as a potential mechanism of this resistance. In two
studies published back-to-back earlier
this year genetically engineered mice with
prostate-­conditional homo­zygous loss of
PTEN (phosphatase and tensin homolog)
were used to show that pharmacological
blockade of the AR is associated with activation of AKT.4,5 One of the mechanisms
responsible was the loss of AKT inhibition
by the AKT-inactivating PH domain and
leucine rich repeat protein phosphatases
(PHLPP1 and PHLPP2). 4,5 Additionally,
pharmacological inhibition of the PI3K
pathway in these models caused activation of receptor tyrosine kinases HER2
and HER3, which in turn led to activation of the AR. Moreover, a separate study
reported that the AR can inhibit PTEN
transcription in cultured prostate cancer
cells, which leads to activation of AKT
signaling.6 These findings are particularly
provocative because aberrations that result
in activation of PI3K–AKT signaling are
common in prostate cancer, and drugs
targeting the PI3K–AKT cascade or HER
signaling are in clinical development.5
Data reported over the past 6 years have
prompted reclassification of prostate cancer
into molecularly defined subtypes, based
on, for example, the presence of underlying
hormone-driven oncogene rearrangements,
most commonly involving ETS genes such
as ERG and ETV1.7 Hormone-dependent
overexpression of ERG and ETV1 proteins
JANUARY 2012 | S77
Key advances
■■ Abiraterone acetate can now be
considered standard of care for patients
with CRPC who progress on docetaxel1
■■ Dual blockade of the PI3K–AKT pathway
and the androgen receptor has emerged
as an important potential strategy for
reversing castration resistance5
■■ Preclinical validation of targeting of
PARP in ETS-positive cancers introduces
a potential new biologically rational
approach for treating advanced prostate
■■ Denosumab is superior to zoledronic acid
in delaying skeletal-related events in men
with bone metastasis9
■■ Radium223 chloride improves survival
in patients with advanced CRPC and
symptomatic bone metastases10
owing to ETS gene rearrangements has
been directly implicated in tumor invasion and metastasis in prostate cancer cell
lines; therefore, the downregulation of these
proteins could help explain the antitumor
activity of therapeutics that disrupt AR signaling in nonselected patients. Accordingly,
it has been suggested that direct targeting of ETS proteins could avoid the toxi­
city and AR‑related drug resistance of
hormone therapies.
In a study published this year, Brenner
et al.7 screened for proteins that interact
with ERG. They found that trans­cription
regulated by both ERG and ETV1 was
mediated by a protein complex comprising
PARP1 (poly ADP-ribose polymerase 1)
and DNAPKcs (DNA-dependent protein
kinase, catalytic subunit), and that disruption of this complex using PARP1 inhibitors
suppressed growth of ETS-driven cancers.
Moreover, it has also been hypothesized
that some sporadic prostate cancers (possibly those involving loss of PTEN) contain
DNA repair defects, so PARP1 inhibition
could function as an effective therapy either
directly via synthetic lethality or indirectly
through inhibition of the PARP1–ETS
complex.8 Studies of PARP inhibitors, both
alone and in combination with AR‑targeting
drugs, are now planned.
Targeting of bone metastases has long
been a priority in CRPC, because of both
its frequent occurrence and associated
morbidity. The phase III double-blind
study of denosumab, a humanized monoclonal antibody against receptor activator of nuclear factor κB ligand (RANKL;
a stimulator of osteoclastic bone resorption) versus zoledronic acid was published
earlier this year. 9 Although denosumab
S78 | JANUARY 2012
treatment resulted in a 3.6-month advantage in the composite end point of time to
a skeletal-related event (SRE), there was no
alteration in overall survival or disease progression. The clinically important adverse
event of jaw osteonecrosis was rare, occurring in 22 patients in the denosumab group
and 12 men who received zoledronate.
These results suggest that denosumab is
more effective at delaying SREs than the
current standard. Denosumab also offers
advantages in terms of its subcutaneous
administration and a lesser requirement
for renal monitoring; however, cost may
limit its widespread use.
The potential survival benefit of effective
bone targeting was proven in the recently
presented phase III ALSYMPCA trial,
testing the intravenously delivered alphaemitter radium223 chloride, which incorporates in place of calcium in replicating
bone causing double-strand DNA damage
in adjacent cells. ALSYMPCA compared
radium 223 chloride with best supportive
care in men with symptomatic bone-only
meta­stases.10 The trial was unblinded after
the preplanned interim analysis met efficacy criteria, demonstrating a 2.8-month
improvement in the primary survival
end point in the radium223 chloride group
(HR 0.695; P = 0.00185). Secondary efficacy end points of time to SRE, time to
PSA progression, and extent and duration of alkaline phosphatase response also
favored the experimental arm. The rate of
hemato­logical toxicity was low, but there
was an increased rate of grade 1–2 diarrhea observed in patients who received
active treatment. Radium223 chloride could
therefore prove to be a highly effective
treatment with low morbid­ity for men with
bone metastases.
In summary, significant milestones in
prostate cancer research in 2011 included
announcement of the results of two positive phase III trials (ALSYMPCA and
AFFIRM); granting of marketing approval
for abiraterone acetate by the FDA and the
EMA; and approval of cabazitaxel, already
approved by the FDA in 2010, by the EMA.
Furthermore, increased use of the autologous vaccine therapy sipuleucel‑T for
the treatment of predominantly chemo­
therapy-naive patients was reported in the
USA. Over the next 12 months physicians
could therefore have at their disposal five
new agents (abiraterone acetate, MDV3100,
cabazitaxel, sipuleucel-T and radium 223
chloride) in addition to docetaxel that
prolong the life of patients with CRPC, and
a second bone-modulating agent in addition to bisphosphonates (denosumab).
Additionally, an unprecedented number of
clinical trials are currently accruing patients
with prostate cancer to test promising novel
therapies. The therapeutic landscape of
CRPC has certainly been redefined over the
past 2 years. It is now anticipated that significant further advances will require biologically rational targeting of mol­ecularly
defined subgroups.
Section of Medicine, The Institute of Cancer
Research and the Royal Marsden NHS
Foundation Trust, Downs Road, Sutton, Surrey
SM2 5PT, UK (C. Pezaro, G. Attard).
Correspondence to: G. Attard
[email protected]
Competing interests
G. Attard declares associations with the following
companies/organizations: AstraZeneca, The
Institute of Cancer Research, Ipsen, Janssen-Cilag
and Sanofi-Aventis. See the article online for full
details of the relationships. C. Pezaro declares no
competing interests.
de Bono, J. S. et al. Abiraterone and increased
survival in metastatic prostate cancer. N. Engl.
J. Med. 364, 1995–2005 (2011).
2. Attard, G. & de Bono, J. S. Translating scientific
advancement into clinical benefit for castrationresistant prostate cancer patients. Clin. Cancer
Res. 17, 3867–3675 (2011).
3. Scher, H. I. et al. Evaluation of circulating tumor
cell enumeration as an efficacy response
biomarker of overall survival in metastatic
castration-resistant prostate cancer (mCRPC):
Planned final analysis of COU-AA-301, a
randomized, double-blind, placebo-controlled,
phase III study of abiraterone acetate plus lowdose prednisone post docetaxel [abstract].
J. Clin. Oncol. 29, LBA4517 (2011).
4. Mulholland, D. J. et al. Cell autonomous role of
PTEN in regulating castration-resistant prostate
cancer growth. Cancer Cell 19, 792–804
5. Carver, B. S. et al. Reciprocal feedback
regulation of PI3K and androgen receptor
signaling in PTEN-deficient prostate cancer.
Cancer Cell 19, 575–586 (2011).
6. Wang, Y. et al. Differential regulation of PTEN
expression by androgen receptor in prostate
and breast cancers. Oncogene 30, 4327–4338
7. Brenner, J. C. et al. Mechanistic rationale for
inhibition of poly(ADP-ribose) polymerase in
ETS gene fusion-positive prostate cancer.
Cancer Cell 19, 664–678 (2011).
8. de Bono, J., Sandhu, S. & Attard, G. Beyond
hormone therapy for prostate cancer with PARP
inhibitors. Cancer Cell 19, 573–574 (2011).
9. Fizazi, K. et al. Denosumab versus zoledronic
acid for treatment of bone metastases in men
with castration-resistant prostate cancer:
a randomised, double-blind study. Lancet 377,
813–822 (2011).
10. Parker, C. et al. Overall survival benefit of
radium-223 chloride (Alpharadan) in the
treatment of patients with symptomatic bone
metastasis in castration-resistant prostate
cancer: a phase III randomised trial
(ALSYMPCA) [abstract LBA1]. Presented at the
16th ECCO–36th ESMO Annual Congress.
The dawn of personalized medicine
Thomas W. Flaig and Dan Theodorescu
Long awaited data from the clinical investigation of bladder cancer in
both the neoadjuvant and adjuvant settings were released in 2011,
setting the stage for the next generation of work in this area. The findings
of a number of studies provide the first steps towards a personalized
approach to this disease.
Flaig, T. W. & Theodorescu, D. Nat. Rev. Urol. 9, 65–66 (2012); published online 20 December 2011;
2011 was a year of progress in the field of
bladder cancer. Efforts to set the context
for more definitive, practice-changing
studies of targeted agents in bladder cancer
were reported. Updated analysis of a large
European phase III trial of neoadjuvant
chemo­t herapy confirmed a significant
survival advantage, and mutations that
define potential drivers of disease, as well
as a gene signature that predicts lymph node
involvement before cystectomy, are now
available. Taken together, these advances
make signifi­cant strides in our understanding of bladder cancer and begin to pour
the founda­t ion for a more personalized
approach to the assessment and treatment
of affected patients.
Currently, no targeted therapies are utilized in the routine clinical care of advanced
bladder cancer, and this is largely due to
lack of study rather than negative results.
Two notable studies published in 2011
addressed this dearth of evidence. The
importance of angiogenesis in oncogenesis
is widely accepted and several antiangiogenic agents have been approved for cancer
treatment in recent years. Preclinical evidence indicates that the vascular end­othelial
growth factor (VEGF) axis is important in
urothelial carcinoma, but there has been
little clinical investigation of vascular inhibition in patients with urothelial cancer.
This year, the Hoosier Oncology Group
published a single-arm, phase II study of
a 21-day cisplatin–­gemcitabine regimen
for advanced urothelial carcinoma with
addition of the VEGF-directed antibody
bevacizumab, in 43 patients.1 The overall
radiographic response rate was 72% (19%
were complete responses), with a median
overall survival of 19.1 months. Both of
these outcomes are encouraging and suggest
improvement over historical series, in
which overall survival has ranged from 14
to 18 months. Pulmonary embolism or deep
vein thrombosis was observed in 21% of
KEY ADVANCES IN MEDICINE the study cohort, prompting a dose reduction of gemcitabine midway through the
trial. The Cancer and Leukemia Group B is
sponsoring a randomized, phase III study
of this regimen in patients with advanced
uro­t helial carcinoma (Clinical
identifier: NCT00942331).
Adding to our insight of targeted therapy,
a new understanding of the mechanism of
pulmonary metastasis in bladder cancer was
described in 2011. Endothelin‑1 (ET‑1) is a
vasoconstrictor, expression levels of which
correlate positively with muscle invasion
and negatively with bladder cancer-­specific
survival. In new preclinical findings, ET‑1
and its receptor were determined to be
neces­s ary for lung metastasis in bladder
cancer, and this process was shown to be
dependent on macrophage activity in the
lung.2 Importantly, pharmacologic inhibition of the ET‑1 axis, using orally bioavailable ET‑1 receptor inhibitors, prevented the
development of lung metastases, but had
little effect on established primary or metastatic tumors. Attempting to translate these
findings into a clinical context suggests that
the use of ET‑1 receptor inhibitors will be
most efficacious in the adjuvant setting, not
in the metastatic setting. This information
may guide the future development of ET‑1
inhibitors for bladder cancer, given their
availability and tolerability for chronic use.
An important update to the large
European phase III neoadjuvant chemotherapy trial was reported in 2011.3 This trial
included patients with T2 grade 3, T3 or T4a
bladder cancer (N0 M0) planning to undergo
radical cystectomy or definitive radiation
therapy. A total of 976 patients were randomized to either three cycles of a 21-day cisplatin, methotrexate and vinblastine (CMV)
chemotherapy regimen or no chemotherapy
before undergoing cyst­ectomy or radiation
therapy. The initial reports of this study in
1999 revealed a trend toward chemotherapy
benefit that did not reach significance. In the
© Orlando Florin Rosu |
updated report, there was a 16% reduction in
the risk of death and an absolute increase in
the 10-year survival (from 30% to 36%) with
CMV chemotherapy. Of note, the reduction in the risk of death was 26% in patients
treated with cyst­ectomy, which is compar­able
to that reported in the neo­adjuvant findings
with metho­trexate, vinblastine, doxorubicin, and cisplatin (MVAC) chemotherapy
reported by SWOG.4 Importantly, there is
now evidence of a survival advantage with
the use of neoadjuvant chemotherapy in two
randomized, phase III studies, so a discussion of neo­adjuvant chemotherapy should
take place in all fit patients with muscle-­
invasive disease preparing for definitive local
bladder cancer therapy.
Despite these positive findings, which are
based on pathological risk stratification, it
is clear that the majority of patients treated
with neoadjuvant chemotherapy do not
benefit. A robust prognostic test is needed
that would facilitate the use of chemotherapy
in those at highest risk of recurrence, with
a predictive test to select the therapy most
likely to elicit a response in each patient.
We did see some progress in the search for
better prognostic markers for bladder cancer
in 2011. In one new study, gene expression
was assessed in three separate cystectomy
cohorts and evaluated in the context of
pathologic lymph node status at the time
of surgery.5 Two cohorts (including 90 and
66 patients) were used as training sets and a
20-gene signature was developed, which was
validated in a larger group (n = 185). Patients
in the valida­tion cohort were prospectively
collected in an earlier clinical trial and analyzed retrospectively. The 20-gene expression model yielded an RR for the discovery
of lymph node metastases of 1.74 in the
desig­nated ‘high-risk’ group compared to
an RR of 0.70 in ‘low-risk’ patients, and was
independent of stage and lymphovascular
JANUARY 2012 | S79
Key advances
■■ Encouraging results have been obtained
from studies of targeted therapies in
bladder cancer,1 laying the groundwork for
future investigation in this area
■■ The reporting of a second randomized
phase III trial of neoadjuvant
chemotherapy3 confirms a survival
advantage for patients with muscleinvasive bladder cancer
■■ A 20-gene expression model has been
developed5 that can identify patients,
before cystectomy, at high risk of lymph
node involvement
■■ Evidence suggests a number of chromatin
remodeling genes are involved in bladder
cancer,6 representing a novel area for
future investigation
■■ The immunohistochemical assessment of
p53 is not a useful marker of prognosis or
predictive of response to chemotherapy9
invasion. Conceivably, with further validation, patients at high risk of node-positive
disease could be identified using this algorithm and then strongly encouraged to
undergo neoadjuvant chemotherapy before
definitive treatment.
Whole-exome sequencing of paired
tumoral and peripheral blood samples
was recently performed in a small set of
patients with bladder cancer, in an attempt
to identify prognostic biomarkers. 6 The
initial findings from this cohort were then
tested in a larger cohort of 88 patients.
Several previously defined mutations were
observed (in TP53, RB1 and HRAS), as well
as a number of novel mutations. Of these,
UTX was the most commonly mutated
gene, identified in 21% of tested individuals,
with 17 nonsense, 5 missense and 1 synonymous mutation noted when combining the
discovery and validation screens. Of particular note, eight of the newly identified
mutated genes were related to chromatin
remodeling, suggesting a potential area for
bladder cancer investigation. Interestingly,
mutations in chromatin remodeling genes
have been commonly found in several other
cancer types, suggesting a fundamental
contribution to carcinogenesis.
Results of retrospective studies have suggested that p53 is both a marker of prog­
nosis7 and potentially predictive of response
to cytotoxic chemotherapy.8 However, the
results of a much anticipated phase III trial
testing p53 as a biomarker in patients after
radical cystectomy were somewhat dis­
appointing.9 Notably, these patients had early
stage disease (pT1–2 N0 M0), and based
on current pathologic risk stratification
S80 | JANUARY 2012
they would frequently not receive adjuvant
chemo­therapy. Patients whose tumors were
negative for p53 overexpression—with <10%
nuclear reactivity on immuno­histochemical
analysis—were observed, while those with
p53-positive tumors were offered randomization to MVAC chemo­therapy or observation. A total of 521 patients were registered,
272 were positive for p53, and 42% of these
consented to random­ization for chemo­
therapy. The findings of the trial were negative: the 5‑year rate of recurrence was 20%
regardless of p53 status, and p53-positive
patients randomized to chemo­therapy had
a similar recurrence rate to p53-positive
patients who did not receive chemotherapy.
In other words, the immunohistochemical
assessment of p53 is not useful in this setting.
This study highlights both the challenge of
successfully completing large-scale clinical
trials in bladder cancer, with a trial accrual
period of more than 8 years despite the participation of many academic institutions, and
also the ongoing need for robust prognostic
biomarkers for urothelial carcinoma.
In summary, building on the findings
described here should result in a better
understanding of the underlying science
of bladder cancer and possible clinical
applications in the years to come. A future
is approaching in which we can rationally
and scientifically apply many of the targeted agents that are currently in development, based on a personalized assessment
of patients with bladder cancer.
University of Colorado Comprehensive Cancer
Center, 1665 Aurora Court, Aurora, CO 80045,
USA (T. W. Flaig, D. Theodorescu).
Correspondence to: D. Theodorescu
[email protected]
Competing interests
D. Theodorescu declares an association with the
following organization: University of Colorado. See
the article online for full details of the relationship.
T. W. Flaig declares no competing interests.
Hahn, N. M. et al. Phase II trial of cisplatin,
gemcitabine, and bevacizumab as first-line
therapy for metastatic urothelial carcinoma:
Hoosier Oncology Group GU 04–75. J. Clin.
Oncol. 29, 1525–1530 (2011).
Said, N., Smith, S., Sanchez-Carbayo, M. &
Theodorescu, D. Tumor endothelin‑1 enhances
metastatic colonization of the lung in mouse
xenograft models of bladder cancer. J. Clin.
Invest. 121, 132–147 (2011).
Griffiths, G., Hall, R., Sylvester, R., Raghavan, D.
& Parmar, M. K. International phase III trial
assessing neoadjuvant cisplatin, methotrexate,
and vinblastine chemotherapy for muscleinvasive bladder cancer: long-term results of
the BA06 30894 trial. J. Clin. Oncol. 29,
2171–2177 (2011).
Grossman, H. B. et al. Neoadjuvant
chemotherapy plus cystectomy compared with
cystectomy alone for locally advanced bladder
cancer. N. Engl. J. Med. 349, 859–866 (2003).
Smith, S. C. et al. A 20-gene model for
molecular nodal staging of bladder cancer:
development and prospective assessment.
Lancet Oncol. 12, 137–143 (2011).
Gui, Y. et al. Frequent mutations of chromatin
remodeling genes in transitional cell carcinoma
of the bladder. Nat. Genet. 43, 875–878
Esrig, D. et al. Accumulation of nuclear p53 and
tumor progression in bladder cancer. N. Engl. J.
Med. 331, 1259–1264 (1994).
Ferreira, C. G., Tolis, C. & Giaccone, G. p53 and
chemosensitivity. Ann. Oncol. 10, 1011–1021
Stadler, W. M. et al. Phase III study of
molecularly targeted adjuvant therapy in locally
advanced urothelial cancer of the bladder
based on p53 status. J. Clin. Oncol. 29,
3443–3449 (2011).
Advances in epidemiology,
pathophysiology and treatment
Eric Chung and Gerald B. Brock
Research into sexual dysfunction is expanding rapidly; in 2011 alone
more than 600 articles on sexual dysfunction were listed in PubMed.
Increased understanding of pathophysiology, epidemiology, clinical
diagnosis and therapeutic options has propelled clinicians and
researchers alike to be more innovative and holistic in their approach to
the management of sexual dysfunction.
Chung, E. & Brock, G. B. Nat. Rev. Urol. 9, 66–68 (2012); published online 10 January 2012;
Although evidence has previously indicated that erectile dysfunction (ED) might
resolve spontaneously,1 the natural course
of ED had not been addressed in detail
before 2011. Recent large-scale population analysis of the Massachusetts Male
Aging Study (MMAS) and the multi­
national Men’s Attitudes to Life Events and
Sexuality (MALES), revealed that 21% and
25% of men reported improvement, over
7 and 3 years of follow-up, respectively.2
Worsening of ED was reported by 51% and
28% of men in MMAS and MALES, respectively. As might be expected, the proportion
of men reporting progression increased
with age, and men younger than 50 years
were more likely to experience improvement than worsening. The age-specific rates
of progression were higher in MMAS than
in MALES, which might be explained by the
fact that the MMAS population is older or
that there was a longer period of follow-up
in MMAS than MALES.
Only a small proportion of men (14%
in MMAS and 28% in MALES) reported
use of phospho­diesterase type 5 (PDE5)
inhibitors, which is not uncommon in
population-­b ased studies. Furthermore,
most patients were recruited before the
commercialization of PDE5 inhibitors.
Although this study reported several pertinent findings, there are problems associated
with combining data from several longitudinal studies, including the difference in
follow-­u p duration between studies.
Moreover, self-reported outcomes might
not represent formal diagnosis of a condition, and the lack of access to PDE5 inhibitors or other erectogenic agents may further
restrict the generalization of this study.
A novel treatment for men with vasculogenic ED was reported in 2011: extracorporeal shock wave therapy (ESWT). In
contrast to high-intensity ESWT, which is
commonly used to treat urolithiasis and
orthopedic conditions, a low-intensity
protocol (LI-ESWT) has previously been
shown to improve the hemodynamics of
many organs via an improvement in vascular supply and endothelial function.
Neovascularization is mediated by the nonenzymatic production of physiologic levels
of nitric oxide, activa­tion of intracellular
signaling pathways, and increased expression of vascular endothelial growth factor
and its receptor.
This year, Gruenwald et al. 3 reported
that penile LI‑ESWT (two treatment sessions per week for 3 weeks in 29 patients)
can improve erectile function and penile
hemodynamics. The mean IIEF-ED
scores increased from 8.8 ± 1 at baseline to
12.3 ± 1 at 1‑month follow-up (P = 0.035).
At 2 months, while patients were on active
PDE5 inhibitor treatment, the IIEF-ED
further increased to 18.8 ± 1 (P <0.0001),
■■ An attempt to describe the natural
history of erectile dysfunction (ED) has
revealed that 25–51% of men experience
progression and 21–25% of men report
■■ A new predictive algorithm5 can
accurately predict erectile function after
definitive prostate cancer treatment
■■ PDE5 inhibitors can increase erectile
variables without sexual stimulation in a
laboratory setting7
■■ Shock wave therapy has been suggested
as a novel treatment option for severe
vasculogenic ED3
■■ Data from a population-based study9
suggests a lack of association
between sexual dysfunction and serum
testosterone level
and 72.4% of patients reached an erection
hardness score of ≥3 (P <0.0001), allowing
full sexual intercourse. These encouraging results suggest that penile LI‑ESWT
could be used to treat patients with severe
vasculogenic ED. However, many questions—such as the long-term impact of
ESWT on penile tissue—remain unanswered. In an animal model, Muller et al.4
have previously shown that ESWT results in
time-­dependent and treatment-dependent
reduction of intra­cavernosal pressure and
alteration in smooth muscle cells, both of
which are potentially deleterious to erectile
function. Large-scale long-term doubleblind multicenter comparative studies, with
specific patient selection criteria using standardized treatment protocols, are required
before this novel therapeutic modality can
be accepted as a mainstay treatment for ED.
The association between ED and prostate
cancer treatment is well established, and the
loss of sexual function in previously potent
men is closely related to outcome satisfaction. A longitudinal multicenter study by
Alemozaffar et al.5 has reported that 37% of
all patients (368 of 1,027) and 48% of men
who had functional erections before treatment (335 of 694), reported functional erections 2 years after definitive prostate cancer
therapy. Multivariate analysis revealed that
younger age, lower PSA levels, better pretreatment sexual functioning score, and
utilization of nerve-sparing surgery were
associated with increased probability of
functional erections after radical prostat­
ectomy (P <0.05 for each; area under curve
[AUC] 0.7 [95% CI 0.72–0.82]). Similarly
for men who received external beam radiotherapy, those who had a lower PSA level,
better pretreatment sexual functioning
score and did not receive neoadjuvant
hormone therapy, were more likely to report
erectile function after treatment (P <0.05
for each; AUC 0.83 [95% CI 0.78–0.88]).
Ethnicity and lower BMI, in addition to
the risks mentioned above, were found to
increase the log-odds of better erectile function after brachytherapy (P <0.05; AUC 0.89
[95% CI 0.85–0.94]).
Investigators generated models based
on the above data that could predict erectile function after primary prostate cancer
treatment, and validated their performance
using data from the Cancer of the Prostate
Strategic Urologic Research Endeavor database. The multivariable logistic regression
models could estimate probabilities of erectile function at 2 years after treatment, from
as low as 10% to as high as 70%, depending
on the patient’s pretreatment character­istics
and treatment details. This important study
demonstrates the use of predictive models
to analyze practical and clinically relevant
patient-reported outcomes for the first
time. Consistent with existing knowledge,
the benefit of nerve-sparing prostatectomy
and the detri­ment associated with adjuvant hormone therapy during radiation
were observed. Unfortunately, this study is
limited by its observational design, possible
selection bias, and absence of reported data
on penile rehabilitation. Furthermore, the
relatively short follow-up of patients did not
take into account that some men continued
to notice improvement or deterioration of
sexual function beyond 2 years. These predictive models can be added to the growing
list of nomograms and artificial neural
networks for prostate cancer management,
which will hopefully result in improved
outcomes for patients.
Penile rehabilitation after radical
prostat­ectomy using PDE5 inhibitors has
attracted considerable interest, and proofof-concept has been demonstrated. Nighttime use of PDE5 inhibitors is thought to
increase the occurrence of nocturnal erections, which aid preservation of corporal
smooth muscle.6 The absence of regular
sexually stimulated erections or nocturnal
erections may lead to insufficient oxygen
tension causing increased fibrous tissue
and decreased smooth muscle content,
which further aggravates ED. In 2011,
Gökçe et al.7 reported that penile erection
can occur in the absence of sexual stimula­
tion in a laboratory-­b ased double-blind
study. 80 men with lifelong premature
ejacula­tion but not ED were divided equally
into four groups to receive a single dose of
JANUARY 2012 | S81
either placebo, vardenafil (10 mg), sildenafil
(50 mg) or tadalafil (20 mg), after 3 days
of sexual abstinence. Despite the absence
of any sexual stimulation, 10%, 41%, 26%
and 55% of men in the placebo, sildenafil,
tadalafil and vardenafil groups, respectively, were able to achieve ≥60% base or
tip rigidities (P <0.05). The median time
to first measured base rigidity was 58.0,
21.5, 54.5 and 57.0 min and the median
total duration of recorded base rigidity was
4.0, 27.5, 10.0 and 11.5 min in patients who
took placebo, sildenafil, tadalafil and vardenafil, respectively. This is the first study
to show that substantial penile rigidity can
be obtained with PDE5 inhibitors, with no
sexual stimula­tion, in a laboratory setting.
Evidence that PDE5 inhibitors can affect
erectile variables supports the idea that
dysfunctional endo­thelium can be rehabilitated, and suggests that men who require
penile rehabilitation might benefit from
PDE5 inhibitors.
Finally, new research has assessed the
complex relationship between metabolic syndrome, sexual dysfunction, and
serum testosterone levels. Metabolic syndrome is often associated with male hypo­
gonadism and a recent meta-analysis 8
has confirmed that patients with metabolic syndrome have significantly lower
total testo­sterone levels (about 3 nmol/l)
than those without. Furthermore, after
adjusting for age and BMI, both type 2
dia­b etes and metabolic syndrome independently predicted lower testosterone level
(adjusted r = –0.752 [P <0.001] and –0.271
[P <0.05], respectively).
In another study, Marberger et al. 9
reported that age, International Prostate
Symptom Score (IPSS), BMI and diabetes
or glucose intolerance, but not serum testosterone or total prostate volume, were
signifi­cant independent predictors of sexual
dysfunction, based on baseline data in the
REDUCE study population. On multi­
variate analysis, only age and IPSS were
significant predictors of all four sexual
function criteria assessed (sexual inactivity,
impotence, decreased libido and Problem
Assessment Scale of the Sexual Function
Index score <9). Several criticisms can be
leveled at this population-based study,
including the varied timing of serum testosterone collection, the lack of reporting
of other testosterone-related variables (such
as free testosterone or sex hormone-binding
globulin measurements) and the confounding factor that higher BMI is likely to result
in metabolic syndrome, which contributes
S82 | JANUARY 2012
to the development of impaired glucose
tolerance and decreased androgen levels.
The lack of association between sexual dysfunction and serum testosterone reported
here questions the value of using modestly
reduced testosterone level as a criterion for
testosterone replacement in older men with
sexual dysfunction.
Studies published in 2011 have improved
and consolidated our understanding
of sexual dysfunction. The large-scale
population-­b ased studies of the natural
history of ED, the proposed prediction
models of ED after prostate cancer treatment and the various emerging and novel
ED therapies have provided new management approaches for this sexually debilitating condition. The association between
testosterone levels, metabolic syndrome and
sexual dysfunction remains complex, contentious and poorly defined. There is no
doubt that continued innovative research
over the next few years will further enhance
our understanding of the field.
Department of Urology, Princess Alexandra
Hospital, Ipswich Road, Brisbane, QLD 4102,
Australia (E. Chung). Division of Urology,
St Joseph Health Care, 268 Grosvenor Street,
London, ON N6A4V2, Canada (G. B. Brock).
Correspondence to: G. B. Brock
[email protected]
Competing interests
The authors declare no competing interests.
Prins, J., Blanker, M. H., Bohnen, A. M.,
Thomas, S. & Bosch, J. L. Prevalence of
erectile dysfunction: a systematic review of
population-based studies. Int. J. Impot. Res. 14,
422–432 (2002).
Travison, T. G. et al. The natural progression
and regression of erectile dysfunction:
follow-up results from the MMAS and MALES
studies. J. Sex. Med. 8, 1917–1924 (2011).
Gruenwald, I., Appel, B. & Yardi, Y. Low-intensity
extracorporeal shock wave therapy‑A novel
effective treatment for erectile dysfunction in
severe ED patients who respond poorly to
PDE5 inhibitor therapy. J. Sex. Med.
Muller, A. et al. The impact of shock wave
therapy at varied energy and dose levels on
functional and structural changes in erectile
tissue. Eur. Urol. 53, 635–642 (2008).
Alemozaffar, M. et al. Prediction of erectile
function following treatment for prostate
cancer. JAMA 306, 1205–1214 (2011).
Padma-Nathan, H. et al. Randomized, doubleblind, placebo-controlled study of postoperative
nightly sildenafil citrate for the prevention of
erectile dysfunction after bilateral nerve-sparing
radical prostatectomy. Int. J. Impot. Res. 20,
479–486 (2008).
Gökçe, A., Demirtas, A., Halis, F. &
Ekmekcioglu, O. The effects of
phosphodiesterase on penile rigidity variables
during a period of no sexual stimulation:
a laboratory setting double-blind study. BJU Int.
107, 264–267 (2011).
Corona, G. et al. Testosterone and metabolic
syndrome: a meta-analysis study. J. Sex. Med.
8, 272–283 (2011).
Marberger, M., Wilson, T. H. & Rittmaster, R. S.
Low serum testosterone levels are poor
predictors of sexual dysfunction. BJU Int. 108,
256–262 (2011).
Semen quality, sperm selection
and hematospermia
Amichai Kilchevsky and Stanton Honig
The past year has seen several new developments in the field of male
reproductive health, including a major revision of how urologists should
assess semen quality following the release of new WHO guidelines, novel
approaches to sperm selection for intracytoplasmic sperm injection,
and changes in how clinicians might evaluate and treat patients with
Kilchevsky, A. & Honig, S. Nat. Rev. Urol. 9, 68–70 (2012); published online 17 January 2012;
Updated reference values for human semen
analysis from the WHO were reported by
Cooper et al.1 in 2010, and much discussion followed in 2011. The new guidelines include significant changes from
prior versions, and it is crucial that urologists understand them. Moreover, the 5th
edition of this manual is the first to include
evidence-­based data as opposed to consensus-based recommendations. The authors
derived the updated reference ranges from
a collection of multicenter, multinational
studies encompassing 1,953 men for whom
the time to pregnancy was <12 months.
However, it is important to note that the
reported ranges are not ‘normal’ ranges but
rather information on the 95th percentile of
semen analysis data. As such, they inform
the reader that 95% of men who achieve
pregnancy within 12 months of trying will
have sperm concentrations of >15 million
cells/ml, >40% of observed sperm demonstrating good movement, and >4% of sperm
with normal morphology.
Unfortunately, these criteria do not
predict the likelihood of achieving a pregnancy in the following 12 months, which
is typically when urologists first see these
patients in consultation. Furthermore, the
reference range must be viewed as a continuum, given that many patients at the low
end of the range might still achieve pregnancy, and men at the high end might not.
The nondefinitive nature of these guidelines
confirms the need for further sophisticated
testing—such as DNA fragmentation analy­
sis, oxidative stress analysis and sperm
evaluation for genomic, proteomic and
metabolic factors—in certain cases of
male factor or unexplained infertility. For
example, a patient whose semen parameters
fall within the new reference ranges but
has not achieved pregnancy might benefit
from DNA fragmentation testing to identify
subtle sperm abnormalities. Esteves et al.2
have analyzed the new reference values
and highlight further issues that may face
the urologist, including whether the referral of male partners will decrease, whether
we were previously overtreating our male
patients and how to better interpret these
reference values by focusing on the 50 th
percentile of data.
Over the past decade, progress in the
field of assisted reproduction has led to a
change in the management of severe male
factor infertility not amenable to medical
or surgical correction. Currently, intracytoplasmic sperm injection (ICSI) is
the treatment of choice for patients who
suffer from either severe oligospermia
or non­obstructive azoospermia (NOA).
Historically, any motile sperm present in the
ejaculate would be preferentially utilized for
ICSI. Alternatively, if no sperm were found
in the ejaculate, sperm surgically extracted
from the testis were used instead. When
very low numbers of sperm were present
in the ejaculate and initial ICSI results with
motile ejaculated sperm were poor, then
testis sperm was considered.
Evidence against this ejaculate-first
approach was recently reported by Hauser
et al.,3 who found that fertilization rates in
patients with relative or virtual azoospermia
were higher when fresh or frozen-thawed
KEY ADVANCES IN MEDICINE testicular sperm cells were used than when
ejaculated sperm cells were used. This
finding is particularly interesting considering that although more motile sperm
cells were found in the ejaculated specimens than in the testicular samples, the
quality of embryos from testicular sperm
(fresh and frozen) was significantly higher
than of those from ejaculated sperm. This
observa­tion led the authors to conclude that
it is the source of sperm cells, and not their
motility, that plays a crucial role in fertility
outcome. This pilot study suggests a possible role for testicular sperm extraction
(TESE) coupled with ICSI in patients with
severe oligo­asthenospermia or relative or
virtual azoo­spermia.
If testicular sperm leads to better fertility outcomes, does it matter if fresh or
frozen-thawed testicular spermatozoa are
retrieved? According to Hauser et al.,3 the
answer is yes. While frozen-thawed spermatozoa may be more conveniently obtained,
the researchers found that for patients with
virtual or relative azoospermia, fresh testis
sperm yielded better implantation rates
than frozen testicular sperm.
Although these results support the use
of fresh testicular sperm for patients with
relative or virtual azoospermia, there is
still no consensus on the best approach for
retrieving testis sperm from men with pure
NOA. This year, the value of diagnostic testis
biopsy in the era of ICSI was addressed by
Kalsi et al., 4 who provide evidence that
microsurgical TESE (m-TESE)— introduced
by Schlegel and Li5 in 1998—is the optimum
sperm retrieval method in patients with
NOA, preferential to fine-needle aspira­tion
and traditional TESE.
Researchers were able to successfully
retrieve spermatozoa from 50 of 100 men
with NOA who underwent m‑TESE at their
center, which includes a success rate of 57%
in men with previously failed attempts at
sperm retrieval. The only significant positive predictor of a successful retrieval was
a previous histological diagnosis of hypospermatogenesis, and therefore the authors
recommend against the common practice
of performing isolated diagnostic testicular
biopsies on men with NOA, and suggest
instead that biopsy should always be combined with a TESE procedure. Therefore,
the take home message for any urologist
treating a patient with NOA is to either
proceed with a combined diagnostic testis
biopsy and send tissue to an andrology laboratory for processing and cryo­preservation,
or to refer the patient to a reproductive
Key advances
■■ The updated 5th edition of the WHO
semen guidelines includes significant
changes from prior versions and is the
first edition to include evidence-based
■■ In certain cases, testicular sperm may
be used for intracytoplasmic sperm
injection, in preference to ejaculated
sperm, for patients with relative or virtual
■■ Diagnostic testis biopsy alone (without
tissue processing) has limited value
in the management of nonobstructive
■■ The underlying cause of hematospermia
can be evaluated using transrectal
■■ Finasteride is a feasible treatment
option for men with recurrent idiopathic
urologist who has this capability. Our
current approach is to begin with a standard TESE and if no sperm are observed to
proceed with an immediate m‑TESE.
Abnormal findings in ejaculate are not
always related to male factor infertility. In
fact, one of the most frequently encountered
problems in general urology is hemato­
spermia, which can be a cause of great
concern and anxiety for affected men. Until
recently, hematospermia was assumed to
be idiopathic and patients were reassured
that their condition was benign. A recent
study by Zhao et al.,6 however, may alter the
urologist’s approach to hematospermia. In
their study, researchers performed trans­
rectal ultrasonography on 270 men with
hematospermia, and found abnormalities
in 95% of the cohort. These abnormalities
were universally benign in patients under
40 years of age, including prostatic calcifications, ejaculatory duct calculi, and benign
prostatic hyperplasia. Patients over the age
of 40 years, however, were significantly
more likely to have a malignant disease; 8
Courtesy of A. Kilchevsky, Yale–New Haven Hospital, USA
JANUARY 2012 | S83
of 126 (6.3%) men over 40 years old were
found to have prostate, seminal vesicle, or
bladder cancer. So, although reassurance
may be appropriate in a younger population, hematospermia in men over 40 years
of age must be evaluated more closely.
A small randomized study by Badawy
et al.,7 published this year, evaluated the
role of finasteride in the treatment of recurrent idiopathic hematospermia. This study
reports the response of 24 patients to treatment with finasteride at 5 mg daily for
3 months. Men who received finasteride
reported a statistically significant improvement in hematospermia compared to those
in the placebo group (66% versus 25%;
P = 0.05). Although this is a small study, it
gives the general urologist an off-label option
for the treatment of recurrent hemato­spermia
after organic causes (such as infection
and malignancy) have been ruled out.
In summary, there have been several new
developments in the fields of male factor
infertility and andrology over the last year
that are of great interest to the general urologist. The new WHO guidelines for semen
quality provide updated reference ranges
rooted in evidence-based data. New techniques for surgical sperm extraction have
both clarified current approaches and provided novel treatment options for NOA and
relative azoospermia. Finally, new information regarding the evaluation and treatment of hematospermia will help us better
identify causality, reassure our patients, and
provide treatment with better outcomes.
Department of Urology, Yale–New Haven
Hospital, Yale Physicians Building, 800 Howard
Avenue, 3rd Floor, New Haven, CT 06519, USA
(A. Kilchevsky). Division of Urology, University
of Connecticut Health Sciences Center, 263
Farmington Avenue, Farmington, CT 06030,
USA (S. Honig).
Correspondence to: S. Honig
[email protected]
Competing interests
The authors declare no competing interests.
Cooper, T. G. et al. World Health Organization
reference values for human semen
characteristics. Hum. Reprod. Update 16,
231–235 (2010).
Esteves, S. C. et al. Critical appraisal of World
Health Organization’s new reference values for
human semen characteristics and effect on
diagnosis and treatment of subfertile men.
Hauser, R. et al. Virtual azoospermia and
cryptozoospermia—fresh/frozen testicular or
ejaculate sperm for better IVF outcome?
J. Androl. 32, 484–490 (2011).
Kalsi, J., Thum, M. Y., Muneer, A., Abdullah, H.
& Minhas, S. In the era of micro-dissection
S84 | JANUARY 2012
sperm retrieval (m-TESE) is an isolated
testicular biopsy necessary in the management
of men with non-obstructive azoospermia? BJU
Schlegel, P. N. & Li, P. S. Microdissection TESE:
sperm retrieval in non-obstructive
azoospermia. Hum. Reprod. Update 4, 439
Zhao, H. et al. The value of transrectal
ultrasound in the diagnosis of hematospermia
in a large cohort of patients. J. Androl. http://
Badawy, A. A., Abdelhafez, A. A. &
Abuzeid, A. M. Finasteride for treatment of
refractory hemospermia: prospective placebocontrolled study. Int. Urol. Nephrol. http://
Objectifying risk for localized renal
Marc C. Smaldone and Robert G. Uzzo
Contemporary treatment guidelines for the localized renal mass are
largely driven by expert opinion and retrospective observational data.
Three articles published in 2011 add important information to the
existing body of literature, enabling improved objectification of risk and
individualization of clinical tradeoff decisions for patients presenting with
localized renal tumors.
Smaldone, M. C. & Uzzo, R. G. Nat. Rev. Urol. 9, 70–72 (2012); published online 13 December 2011;
Increased utilization of abdominal imaging
has led to a significant rise in the inci­
dental detection of clinically localized
renal tumors. Although the gold standard
treatment for clinically localized small renal
masses remains surgical excision, contemporary practice patterns have evolved in
an effort to match treatment strategy to
tumor characteristics, as well as minimize
the adverse effects associated with chronic
kidney disease (CKD).1 The armamentarium of management options has expanded
to include open and minimally-invasive
radical nephrectomy, nephron-sparing
surgical techniques, tumor ablation, and
active surveillance with curative intent. In
2009, the American Urological Association
Guidelines committee issued the following recommendation: “surgical excision
by partial nephrectomy is the reference
standard for the management of the clinical T1 renal mass, whether for imperative
or elective indications, given the importance of preservation of renal parenchyma
and avoidance of CKD.”2 However, when
evaluating these guidelines it is important
to bear in mind that the evidence quality
of the contemporary body of literature
is poor, and is limited primarily to retrospective observational studies. As a result,
treatment decisions are driven primarily by
expert opinion and surgeon experience, and
a number of important questions regarding
the safety and oncologic efficacy of current
management strategies remain unanswered.
Three studies published in 2011 provide
evidence that will improve the communication of risk and enable tradeoff decisions
when counseling patients diagnosed with
localized renal tumors.
Until recently, level 1 evidence comparing the oncologic efficacy of radical and
partial nephrectomy was lacking. In 2011,
Van Poppel et al.3 reported the results of a
European Organization for Research and
Treatment of Cancer (EORTC) prospective,
noninferiority, multicenter, phase III trial
assessing oncologic outcomes in patients
with a normal contralateral kidney randomized to either radical nephrectomy (n = 273)
or nephron-­sparing surgery (n = 268) for
solitary localized renal tumors ≤5cm in
size.3 Oncologic efficacy was found to be
equivalent between groups, with 10-year
progression rates of 4.1% (95% CI 1.7–6.5%)
and 3.3% (95% CI 1.2–5.4%) for partial and
radical neph­rectomy, respectively (P = 0.48).
Furthermore, only 12 renal-cancer-related
deaths were noted in the study, with a
median follow-up of 9.3 years, suggesting
that the risk of cancer-­s pecific death in
patients surgically treated for stage 1 tumors
is extremely low. However, in the intentionto-treat analysis, the authors reported a
significant 10-year overall survival benefit
for patients undergoing radical surgery
compared to partial nephrectomy (81.1%
versus 75.7%; HR 1.50 [95% CI 1.03–2.16];
P = 0.03). When the analysis was limited to
patients with histologically confirmed renal
cell carcinoma, these differences in overall
survival were less pronounced.
The findings of this study must be placed
in context of its limitations. Designed to
recruit 1,300 patients, the trial was closed
early due to poor accrual, and did not
achieve the target sample size chosen to
identify a 3% difference in 5-year survival. In addition, there were differences
in baseline comorbidities and considerable
crossover between treatment arms (5.6%
of patients randomized to radical neph­
rectomy underwent partial neph­rectomy,
and 14.6% of patients randomized to partial
nephrectomy underwent radical neph­
rectomy), which may have influenced the
reported survival outcomes. Despite these
limitations, the findings raise important
questions regarding the relative contributions of preserved renal function and competing risks from pre-existing comorbidities
to overall survival following surgical resection for localized renal tumors. Although
these results are thought-provoking, further
study is necessary to determine which
patients stand to benefit the most from
nephron-sparing surgery.
The benefit of renal function preserva­
tion in nephron-sparing surgery must
be weighed against the increased risk of
postoperative complications compared
with radical nephrectomy. 2 However, a
lack of standardized reporting methods
has resulted in a body of literature with
limited generalizability that is challenging to interpret. In 2011, investigators at
Fox Chase Cancer Center attempted to
address these issues by rigorously evaluating the complica­t ions associated with
partial neph­r ectomy in their institutional database4 using the Clavien-Dindo
classifica­t ion system (CCS). 5 Aiming to
provide a benchmark for comparative
studies between institutions, patients were
stratified using the nephrometry scoring
system 6 to investigate the relationship
between tumor complexity (low complexity score 4–6; intermediate complexity score
7–9; high complexity score 10–12) and risk
of postoperative complications. The study
cohort consisted of 390 patients with available imaging data for review, who underwent partial neph­rectomy between 2007
and 2010; 109 (28%), 217 (55.6%), and
64 (16.4%) patients underwent nephron-­
sparing surgery for low, intermediate, and
high complexity lesions, respectively. The
overall proportion of patients incurring
minor (CCS I–II) and major (CCS III–V)
complications were 26.7% and 11.5%,
respectively. No significant differences were
observed in minor complications between
complexity groups, but patients with highly
complex lesions were more likely to develop
a major complication requiring secondary
intervention than intermediate and low
complexity tumors (22% versus 11% versus
6%; P = 0.001). Controlling for demographic
and clinical characteristics, patients with
highly complex renal tumors were 5.4 times
(95% CI 1.2–24.2) more likely to sustain a
major complication compared to those with
low complexity lesions. Although this study
was limited by its retrospective methodology and lack of external validation, these
findings highlight two important points: the
rigorous reporting of complications using
standardized reporting methodology is
essential for comparing outcomes between
institutions, and the benefit of nephron
preserva­t ion in highly complex tumors
comes with the attendant risk (>20%) of a
major post­operative complication.
Recent evidence suggests that a substantial proportion of the rapidly rising
number of incidentally diagnosed small
renal masses represents indolent disease
Key advances
■■ In a recent phase III randomized trial,3
partial nephrectomy demonstrated
equivalent cancer-specific survival to
radical surgery for tumors ≤5 cm, but did
not demonstrate the expected overall
survival benefit
■■ The benefit of nephron preservation
in highly complex tumors comes with
the attendant risk (>20%) of a major
postoperative complication4
■■ A large proportion of small renal masses
under observation grow slowly, and
the short-term risks of progression or
metastasis under active surveillance are
that may not require treatment. 7 As a
result, much attention has been directed
towards describing the natural history of
untreated renal tumors in an effort to identify which lesions are safe to observe and
which require early definitive intervention.8
In 2011, Jewett et al.9 reported the results
of a multicenter prospective phase II trial
investigating the growth kinetics and progression rates of 209 incidentally diagnosed cT1a lesions under observation in
patients deemed unfit for surgery owing
to advanced age, comorbidity, or refusal
of other treatment.9 In this cohort of 178
patients (mean age 73 years, mean tumor
size 2.3 cm), the authors defined tumor progression as growth to ≥4 cm, tumor volume
doubling time ≤12 months, or progression
to metastatic disease. Notably, 99 patients
(56%) underwent renal biopsy and 127
patients (151 masses) were followed up for
>12 months (mean 28 months). Important
study findings include an average growth
rate of 0.13 cm per year, documented progression (as defined above) in 27 patients
(15%), and only two patients (1.1%) developing evidence of metastatic disease.
Furthermore, there was no difference in
growth rate between biopsy-proven malignant and benign disease (0.14 cm per year
versus 0.17 cm per year; P = 0.8), and 36% of
biopsy-proven renal cell carcinomas showed
either no evidence of growth or a decrease
in size. Although limited by lack of central
pathology review, elevated nondiagnostic
biopsy rate (33%), and short duration of
follow-up, this study should be commended
for its rigorous eligibility criteria, use of protocol renal mass biopsy, and strict definition
of tumor progression. These findings add
significantly to the growing body of literature documenting that a large proportion of
small renal masses under observation grow
slowly, and that short-term risks of progression or metastasis under active surveillance
are low.8,10
In the absence of level 1 evidence, physi­
cians are increasingly challenged to manage
risk on a patient by patient basis. Although
the list of unanswered questions remains
long, the studies described above each
provide evidence to improve risk communication and enable tradeoff decisions
when counseling patients diagnosed with
localized renal tumors. Until the ability to
match treatment to tumor biology has been
achieved, urologists must continue to vigilantly evaluate and challenge contemporary
management strategies, objectify tradeoff
risks, and individualize treatment strategies.
JANUARY 2012 | S85
Division of Urologic Oncology, Department of
Surgery, Fox Chase Cancer Center, 333
Cottman Avenue, Philadelphia, PA 19111, USA
(M. C. Smaldone, R. G. Uzzo).
Correspondence to: R. G. Uzzo
[email protected]
Competing interests
The authors declare no competing interests.
Go, A. S., Chertow, G. M., Fan, D.,
McCulloch, C. E. & Hsu, C. Y. Chronic kidney
disease and the risks of death, cardiovascular
events, and hospitalization. N. Engl. J. Med.
351, 1296–1305 (2004).
Campbell, S. C. et al. Guideline for
management of the clinical T1 renal mass.
J. Urol. 182, 1271–1279 (2009).
Van Poppel, H. et al. A prospective, randomised
EORTC intergroup phase 3 study comparing the
oncologic outcome of elective nephron-sparing
surgery and radical nephrectomy for low-stage
renal cell carcinoma. Eur. Urol. 59, 543–552
Simhan, J. et al. Objective measures of renal
mass anatomic complexity predict rates of
major complications following partial
nephrectomy. Eur. Urol. 60, 724–730 (2011).
Dindo, D., Demartines, N. & Clavien, P. A.
Classification of surgical complications: a new
proposal with evaluation in a cohort of 6336
patients and results of a survey. Ann. Surg.
240, 205–213 (2004).
Kutikov, A. & Uzzo, R. G. The R.E.N.A.L.
nephrometry score: a comprehensive
standardized system for quantitating renal
tumor size, location and depth. J. Urol. 182,
844–853 (2009).
7. Hollingsworth, J. M., Miller, D. C., Daignault, S. &
Hollenbeck, B. K. Rising incidence of small renal
masses: a need to reassess treatment effect.
J. Natl Cancer Inst. 98, 1331–1334 (2006).
8. Chawla, S. N. et al. The natural history of
observed enhancing renal masses: metaanalysis and review of the world literature.
J. Urol. 175, 425–431 (2006).
9. Jewett, M. A. et al. Active surveillance of small
renal masses: progression patterns of early
stage kidney cancer. Eur. Urol. 60, 39–44 (2011).
10. Smaldone, M. C. et al. Small renal masses
progressing to metastases under active
surveillance: A systematic review and pooled
analysis. Cancer
Don’t miss out!
Sign up to free e-ALERTS and be notified when new articles are published
in Nature Reviews Urology. Research Highlights, News & Views, Reviews,
and Perspectives—you won’t miss a thing!
S86 | JANUARY 2012