M. Alfred Haynes Research Training Institute for Social Equity Scholar 2013

M. Alfred Haynes Research Training Institute
for Social Equity
Ramona Bhatia, MD, MS
Scholar 2013
Among people living with
 75% are men
 44% are African-American (AA)
Due to antiretroviral therapy
(ART), 50% of PLWHA will be
aged 50+ by 2015
3-fold increase in non-AIDS
defining cancers (NADC) due to
aging of the HIV population
cdc.gov; Shiels et al., J Natl Cancer Inst, 2011; huffingtonpost.com
Prostate cancer (PCa) is the most commonly
diagnosed cancer in American men
AAs are at high risk for PCa
Studies on HIV-infected (HIV+) men indicate
PCa risk is that of general population
Predictors of PCa in HIV+ men are
Testosterone supplementation
seer.cancer.gov; Shiels et al., Cancer Epidemiol Biomarkers Prev, 2010
Compared to the general population, HIV+
patients are at higher risk for:
Lung and testis cancer development
 Advanced colorectal cancer
HIV infection and AA race are associated with
lung cancer under-treatment, contributing to
increased mortality
HIV+ and AA men with PCa are less likely to
receive radical prostatectomy
Albini et al., AIDS Res Hum Retroviruses, 2013; Shiels et al., JAIDS, 2010; Suneja et al., AIDS, 2013; Kumar et al., Med Oncol, 2011
Aim 1: To describe the incidence rate of PCa in
HIV+ men in a national cohort
Aim 2: To describe predictors of PCa in HIV+
men in a national cohort
The incidence rate of PCa in HIV+ men will
match that of the general population
Advanced age, AA race, and use of
testosterone will predict PCa development in
HIV+ men
27,000 HIV+ patients
aged 18+
 >60,000 patient-years
of follow-up
 80% male, 40% AA
 8 HIV primary care
sites across U.S.
 Electronic medical
record data (100s of
data points)
 PCa diagnosis
verified by pathology
Inclusion criteria:
 Men enrolled from
1995-2012 without
diagnosis of PCa on
cohort entry
65 incident PCa cases
Stratify incidence rates based on age, race, site,
and presence of prostate specific antigen (PSA)
screening (i.e., 2 tests at least 6 months apart
prior to PCa diagnosis)
Compare to national PCa incidence rates (i.e.,
CDC Wonder) using z-test
Compare covariates in PCa (N=65) and non-PCa
Randomly sample non-PCa group 1:4
Use computational analysis to identify clusters
Use intra-cluster correlations to identify key
Bivariate/multivariate analysis
CD4+ count
HIV viral load
A high burden of PCa in HIV+ men supports
further studies on:
HIV-related and racial disparities in PCa treatment
and outcome
Appropriateness of PCa screening guidelines in
HIV+ patients
Association of testosterone use and PCa
development calls for more research on
appropriateness of androgen supplementation
in HIV+ men
Multi-site study with diverse population
(external validity)
Validated PCa and clinical endpoints (less recall
PCa incidence accounting for screening
Longitudinal cohort reflecting natural history of
Etiologies of PCa treatment disparities in HIV+
Qualitative comparison of facilitators and
barriers to PCa and HIV screening in AA males
in Chicago
Guidelines on PCa screening and testosterone
use in HIV+ men
Funding plans
Career development award
Northwestern University
Dr. Adam Murphy (Dept. of Urology)
 Dr. Chad Achenbach (Div. of Infectious Diseases)
Dr. Agboto
 Dr. Langston
 Drs. Matthews-Juarez and Juarez
Supported by the Creative and Novel Ideas in HIV
Research (CNIHR) grant (NIH/IAS) and the
Northwestern University Specialized Program of
Research Excellence (SPORE) in Prostate Cancer
Medical conditions
Surgical conditions
Chronic pain (LaRue et al., 1997)
Depression (Bess et al., 2013)
Spine surgery for degenerative disease (King et al.,
Lung cancer (Suneja et al., 2013)
 Prostate cancer?
Suneja et al., AIDS, 2013
Retrospective study utilizing Enterprise Data
Warehouse (EDW)
Data from all encounters at Northwestern Memorial
Hospital, Chicago, IL and affiliates from 2001-2012
Inclusion criteria: all men with PCa from the ages of
HIV-infected cases age- and race-matched to HIVuninfected controls
2010 American Joint Committee Cancer Classification
System used for clinical staging
Treatment appropriateness defined by National
Comprehensive Cancer Network(NCCN) guidelines
for risk-appropriate therapy
 Majority of HIV
patients were on
antiretroviral therapy
(ART) (98%) and virally
suppressed (90%) at
PCa diagnosis
 Clinical stage and
NCCN classification
did not differ based on
HIV status
 Treatment
appropriateness was
similar between groups
 HIV infection, age, and
African-American race
predicted decreased
rates of RP
 All-cause mortality rate
in deaths/1000 personyears was 11.9 for HIVinfected vs. 7.4 for
controls (p= 0.47)
Recent data suggests that RP improves survival
compared to radiation for localized PCa
No studies on etiologies of surgical disparities
in HIV patients
In a general population, urologists offer
curative PCa treatment less often to Blacks,
contributing to treatment disparities
Potential surgical barriers include: risk to
surgeon, perception of poor healing, and
lifestyle judgment
To evaluate urologists’ beliefs, perceptions, and
attitudes on performing RP for clinically
localized PCa in HIV-infected men
We hypothesize that urologists overestimate
the risk of HIV transmission in the operating
room, contributing to decreased RP in HIV
Cross-sectional survey
Beliefs/Perceptions: risk of HIV transmission, rates
of wound healing compared to seronegative men
Attitudes: on HIV and alternative lifestyles
Outcome: estimated rates of RP performed on
eligible HIV-infected men compared to that of
Members of the American Urologic Association
Have performed at least ten RPs, with at least one in
the last year
Have seen at least one HIV patient in the last ten
Characteristics at cancer diagnosis:
2.8 (1.2, 5.0)
0.6 (0.2, 1.5)
1.4 (0.4, 3.7)
43 (38, 50)
46 (40, 54)
44 (39, 51)
White, n (%)
203 (59)
137 (45)
340 (52)
Male, n (%)
293 (85)
263 (86)
556 (86)
HBV/HCV infection, n (%)
60 (17)
74 (24)
134 (21)
IDU, n (%)
49 (14)
69 (23)
118 (18)
133 (39)
117 (38)
250 (38)
83 (24)
70 (23)
153 (24)
127 (37)
118 (39)
245 (38)
Nadir CD4 count, cells/µLb
62 (11, 174)
30 (4, 106)
45 (7, 137)
Pre-cART HIV RNA , log10 copies/mL b
5.3 (4.8, 5.7)
5.5 (4.9, 5.8)
5.4 (4.8, 5.8)
Total, n
Follow-up time, years
Age, years
Smoking, n (%): b
Prostate cancer (PCa) is the most commonly
diagnosed cancer and second leading cause of
cancer-related deaths in US men
Risk increases with age
Black men are 1.5 times more likely to develop PCa
Black men are less likely to receive treatment with
intent to cure
Black men are 2.4 times as likely to die from PCa
PCa is one of the most common non-AIDS
defining cancers, and age-adjusted risk of
developing PCa in PLWHA is at least that of
the general population
HIV patients with lung cancer are less likely to
receive chemotherapy and surgery, which may
increase mortality
2010 American Joint
Committee Cancer
Classification System used
for clinical staging
 Treatment
appropriateness defined
Recent data suggests that RP results in
by National
improved long term survival compared to
radiation therapy for clinicallyComprehensive
guidelines for riskappropriate therapy
 Calculated life expectancy
determined using
Charlson Comorbidity
Index (CCI)
Index ≥3 less than 10 year
life expectancy
National dataset analysis using Centers for
AIDS Research (CFAR) Network of Integrated
Clinical Systems (CNICS)
Validated cancer subset
 Comprehensive data on covariates and medications
(i.e., testosterone)
 PCa incidence
 PCa stage at presentation
 PCa treatment trends