The misuse of tumour markers and other tests: Health Care System

The misuse of tumour markers and other
tests:
What it means to you, the patient and the
Health Care System
Elizabeth MacNamara
64th McGill Family Medicine Refresher Course
December 4th 2013
Disclosure of Commercial Support
 This Jewish General Hospital Laboratory has received financial
support from Roche Diagnostics of $10,000 per annum for five
years as an unrestricted educational grant.
 I have received no monies, or in kind support, from any
commercial company in any capacity apart form the above
Goals
 Understand that the excessive use of laboratory tests
results in a waste of resources, unnecessary follow up tests
and possible harm to the patient.
Workload
Controlling unnecessary tests
1. Urine microscopy
Savings – $280,000 per year
2. Folic acid
Savings $50,000 per year
Controlling unnecessary tests
3. Anticipation of exponential growth
Ex: Vitamin D
4. Accepting the limitations of a new test
Ex: BNP
3. Multidiciplinary approach for complex algorithms
Ex: hs Troponin
Réduction de 14% des tests hs Troponin
Family Physicians and the lab
 Many physicians do not follow guidelines for testing
 Physicians can easily overcome any restriction
 Most physicians do not see the negative impacts of over
testing
 How can we work together to decrease unnecessary
tests such as tumour markers?
Uses of tumour markers
 Screening
 Diagnosis
 Monitoring
 Prognosis
The ideal tumour marker
Characteristics
Highly specific
Detectable only in one tumor type
Highly Sensitive
Non-detectable in physiological or benign
disease states
Long lead time
Sufficient time for alteration of natural
course of disease
Levels correlate with
tumour
Prognostic and predictive utility of the
tumour marker
Short half-life
Frequent serial monitoring of the marker
levels after 5-6 half lives
Simple/inexpensive/
easily attainable
Applicability as screening test &
acceptability by target population
Tumour Markers for Screening
 Not recommended in healthy population
 No tumour marker is 100% specific
 Often elevated with non-malignant conditions
 Most not elevated in early stages of malignancy
Non-Malignant Conditions associated
with elevated tumour markers
 AFP: viral hepatitis, liver injury, IBD, pregnancy
 B-HCG: testicular failure, marijauna smokers, pregnancy
 CEA: smokers, IBD, hepatitis, Cirrhosis, Pancreatitis, Gastritis
 Ca 15-3: G-CSF use, cirrhosis, benign breast disease
 Ca-125: peritoneal irritation, endometriosis, PID, Hepatitis,
pregnancy
 Ca: 19-9: biliary disease, pancreatitis, jaundice, IBD, cirrhosis
 PSA: prostatitis, BPH, DRE, ejaculation, urinary retention
THE CASE OF PSA
 PSA is one of the few tumour markers routinely
recommended for screening, diagnosis, prognosis and
monitoring
Canadian Urology Association 2011
 Offer to men > 50 with at least 10 years life expectancy
 Offer to men > 40 if high risk (family hx, African
American)
 May benefit from offering baseline PSA in men 40-49
 If bx indicated (PSA >4 +/- abnormal DRE), 10-12 core
TRUS peripheral zone
 Consider treatment (surgery, radiation or androgen
deprivation) according to tumour volume, grade, DRE,
PSA
Controversy
 k
Controversy
“…final guideline says there’s little if any evidence
that PSA testing saves lives – while too many
men suffer impotence, incontinence, heart
attacks, occasionally even death from
treatment of tiny tumours that never would
have killed them.”
New York Times, July 2012
Reasoning for USPTF for not
recommending PSA screening
 Lifetime risk for diagnosis 15.9%
 Lifetime risk for dying of prostate cancer is 2.8%
 Most cases have good prognosis even without treatment
but some cases are aggressive
 Rare before age 50
 70% of deaths occur after age 75 (as with all causes of
death)
Harms of Screening
 The reduction in prostate cancer mortality after PSA
based screening is very small (0-1/1000)
 33% of those screened suffer significant harmful side
effects
 Pain, fever, bleeding, infection, and transient urinary
difficulties associated with prostate biopsy, psychological
harm of false-positive results
 90% of positive biopsies go on to active treatment
Harms of Treatment
 Harms of treatment
 Erectile dysfunction (29/1000)
 Urinary incontinence (18/1000)
 DVT or PE (1/1000)
 Cardiovascular event (2/1000)
 Death (<1/1000)
 Bowel dysfunction
 Psychological harm
Reaction in the US; which do you
think is correct?
 American Urological Association:
 “Outraged”
 Recommendations are “inappropriate and irresponsible”
 Sited significant methodological flaws, and inadequate
follow up to make a judgment
 Annals of Internal Medicine:
 The task force is “ideally suited to provide an objective,
unbiased assessment” because it’s members, unlike many of
it’s critics “have no emotional, ideological or financial
conflicts of interest”
So… how is PSA used at the JGH?
How is PSA used at the JGH
How is PSA used at Herzl
HERZL requisition changed to include PSA – increase from ~150 to 560 per yr
What was the greatest # of PSAs
requested for one patient in 2012?
What was the greatest # of PSAs
requested for one patient in 2012?
27 PSAs
 Dr. Melnychuk??
 Requisitions under his name but…
 He did not order all of them!!
 So who did?
 ? Phelbotomist
 ? Nurse
 ? Secretary
 ? Patient
Patient Pressure
 How many of you have been asked to order tests that
you were not convinced were indicated by patients?
American Urological Association
 No PSA screening if




<40
>70
< 15 year life expectancy
Average risk of prostatic cancer and >40 <54 yr.
 55 to 69yr weigh the benefits of preventing prostate
cancer mortality in (1 man for every 1,000 men screened
in 10 years) against the known potential harms
associated with screening and treatment.
 If screening every two or more years preferable to annual
Vitamin D: Is routine measurement
necessary?
Increase in Testing 2004 to 2010
 Ontario increase X 24 ($66,000,000/yr)
 Quebec increase X 5
 France increase X 3
 US increase X 12
 UK increase X 6
Exponential increase in requests for serum vitamin D at JGH
7000
6000
y = 8E-232e0.2688x
Number of assays performed
5000
4000
3000
2000
1000
0
2007
2008
2009
2010
Year
2011
2012
2013
Vitamin D: Is routine measurement necessary?
Increase in Vitamin D testing
 2004 to 2010
 Ontario increase X 24
 Quebec increase X 5
 France increase X 3
 US increase X 12
 UK increase X 6
2
4
6
(Months)
8
10
12
14
16
Modified from Am J Clin Nutr 96(5) 1152-3
Physiological (sun exposure)
 Vitamin D intakes of 2,000 IU/day are safe
 IOM 4,000 IU/day
 10,000 IU/day from UV
 Lifeguards in the United States and in Israel, farmers in the
Caribbean all >100 nmol/L
 Regular sun-tan parlor users >80 nmol/L
 The highest 25(OH)D concentrations up to 235 nmol/L from UV
 No hypercalcemia
JGH lab policy
 Toxicity in normal adults requires intake of more than
40,000 IU/day
 <50 nmol/L supplement and repeat after 6 months
 50 – 75 nmol/L
supplement and repeat after 1 year
 >75 nmol/L no repeat
 Decrease of 30% of assays performed
Summary
 High prevalence of insufficiency - 85%
 Target is likely > 75 nmol/L
 Likely need 1500 – 2000 IU/day
 6-8 months to reach a plateau
Patient harm from excessive testing
Summary
 Consider whether there is truly an indication for the blood
test
 Ask, will the results alter patient management and if so
how
 Ask if the result can harm the patient if wrong
 Wasting Health Care money damages our Health Care
System
Thank you
 Questions?
Canadian Urology Association 2011
 Watchful waiting: no therapy until the patient develops
symptoms secondary to advanced prostate ca
 PSA Q 6-12 months
 Active Surveillance: close follow up on patients dx with
early stage, low risk prostate ca
 PSA Q 3-6 months, bx 1x/year
 Therapy recommended when cure deemed possible
and disease progression occurs (rapid PSA doubling time,
progression on DRE, grade progression or tumour volume
progression on repeat Bx)
Cut-off: 50 or 75 nmol/L
Best Practice & Research Clinical Endocrinology & Metabolism 25 (2011) 681 - 691
Cut-off: 50 or 75 nmol/L
Best Practice & Research Clinical Endocrinology & Metabolism 25 (2011) 681 - 691
Cut-off: 50 or 75 nmol/L
Best Practice & Research Clinical Endocrinology & Metabolism 25 (2011) 681 - 691
Prevalence
 50% <50 nmol/L European countries
 25% <25 nmol/L in some groups
 JGH
 85%
<75 nmol/L
 55%
<50 nmol/L
 6%
<25 nmol/L
Dosing
 600 IU/day probably insufficient
 1000 IU/day will increase 25 Vitamin D by 20-25 nmol/L
 Increase to mean from 50 to 75
 Osteoporosis Canda
 800 - 2000 IU/day
`