Background:
Prostate cancer is the leading cancer diagnosis among Veterans with HIV and will soon be the leading cancer among all US persons with HIV (PWH). Despite the substantial prostate cancer burden for PWH, there are little data on prostate cancer clinical characteristics and outcomes. Therefore, we studied prostate cancer characteristics at diagnosis and survival by HIV status in the Veterans Aging Cohort Study (VACS)-HIV, a national cohort of Veterans with HIV and demographically similar Veterans without HIV.
Methods:
We used data from VACS-HIV (2001-2018) to identify a cohort of male PWH prior to prostate cancer diagnosis (n=791), as well as male comparators without HIV (PWoH n=2,778). We compared patient demographics, prostate specific antigen (PSA) testing and prostate cancer clinical characteristics by HIV status. We then compared prostate cancer risk groupings (D’Amico) and prostate cancer-specific and overall survival by HIV status, stratified by risk group using age-adjusted Cox regression models.
Results:
VACS-HIV patients with prostate cancer had a median age of 62 years, which did not differ by HIV status. Race/ethnicity proportions were also similar, with non-Hispanic Blacks being the most common group diagnosed with prostate cancer. PWH with prostate cancer frequently had detectable HIV viremia at prostate cancer diagnosis (>60%). HIV infection was associated with higher PSA (median 6.8 vs. 6.3 ng/mL; p=0.005) but no difference in Gleason grade. There was less frequent PSA testing among PWH prior to prostate cancer diagnosis (1.25 fewer tests than PWoH, age adjusted; p<0.001). PWH were more likely to be diagnosed with D'Amico intermediate/high risk localized prostate cancer (68% vs. 63%; p= 0.02) and advanced prostate cancer (either nodal involvement or metastatic disease) than PWoH (4.0% vs. 2.7%; p=0.04). Both relationships persisted after adjustment for age. HIV was significantly associated with worse age-adjusted all-cause mortality for intermediate-, high-risk localized and advanced cancers. PWH did not have higher prostate-cancer specific mortality in any cancer risk group compared to PWoH.
Conclusions:
PWH were diagnosed with higher risk prostate cancers more frequently in VACS-HIV than those without HIV possibly reflecting lower rates of PSA testing in this group. Higher non-cancer mortality seen in those with HIV infection may impact the relative risks and benefits of prostate cancer management strategies-including observation-for appropriate patient groups.
