Abstract BACKGROUND: Active surveillance (AS) is now the preferred treatment strategy for the >50% of prostate cancer (PCa) patients diagnosed with low-risk disease, where five-year survival rates are >99%. However, AS requires frequent and costly healthcare visits. In a case-only study of AS patients, we investigated whether polygenic risk scores (PRS) could identify patients who may benefit from intensive vs passive surveillance to reduce these burdens. METHODS: This study included 1,220 PCa patients on AS in the Canary Prostate Active Surveillance Study (PASS). A previously developed multi-ancestry PRS of 451 PCa risk variants (PRS451) was constructed in PASS, along with a PRS of 400 variants, excluding 51 PSA-associated variants (PRS400). Outcomes included upgrading and extreme upgrading at follow up and prostate volume, PSA, percent of biopsy cores with cancer, and Gleason grade at diagnosis. We investigated PRS associations with risk of upgrading with Cox proportional hazards regression adjusting for age at diagnosis and 10 principal components (PCs) of ancestry. Associations with risk of extreme upgrading were performed similarly but with a competing risk analysis, where the competing event was treatment. Discriminative ability was evaluated with Harrell’s C-index. To better account for population structure, we also fit mixed models with fixed effects for PRS, age at diagnosis, and 10 PCs, with a random effect for a genetic relationship matrix. RESULTS: Of the 1,220 AS patients, 38.5% (n=470) upgraded during the median 4.2 years of follow up and 85% (n=1041) self-reported as non-Hispanic White. In time-to-event analyses, each SD unit increase in PRS451 was associated with a 23% increased risk of upgrading (P=4.2 × 10−5), while PRS400 was associated with a 27% increased risk (P=1.6 × 10−6). PRS were similarly associated with 21% (95% CI=1.02-1.44, P=0.03) and 27% (95% CI=1.06-1.52, P=0.01) increased risk of extreme upgrading, respectively. Compared to a model with age at diagnosis, PCs, and clinical risk factors (C-index=0.617, 95% CI=0.590-0.644), including PRS400 slightly improved the ability to detect risk of upgrading (C-index=0.632, 95% CI=0.604-0.659). Mixed model results were highly consistent. Each SD unit increase in PRS400 was associated with 1.38 higher percentage points of biopsy cores with cancer (95% CI=0.76-2.00, P=1.1 × 10−5), 1.48 cc smaller prostate volume (95% CI=-2.82, -0.14, P=0.03), and 0.18 ng/mL higher PSA (95% CI=0.01-0.35, P=0.05) at diagnosis. Except for PSA, associations were similar or weaker for PRS451. CONCLUSION: In patients on AS, high PRS was associated with increased risk of upgrading, extreme upgrading, smaller prostate volume, and possibly tumor multifocality. As such, PRS could inform the intensity of surveillance for low-risk PCa patients, reducing the burden of AS for patients with low risk of disease progression. Citation Format: Louisa B. Goss, Menghan Liu, Yingye Zheng, Boya Guo, David V. Conti, Christopher A. Haiman, William J. Catalona, John S. Witte, Daniel W. Lin, Lisa F. Newcomb, Burcu F. Darst. Polygenic risk score associated with risk of upgrading and tumor features in a prospective cohort of prostate cancer patients on active surveillance [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6616.
Read full abstract