How unfavorable is AUA unfavorable intermediate-risk prostate cancer? Results from a large, institutional cohort.

Abstract

342 Background: Major clinical guidelines including the American Urology Association (AUA) defines unfavorable intermediate-risk prostate cancer (UIR-PCa) according to diagnostic PSA, clinical stage, biopsy positive cores, and Gleason grade group (GG). The current AUA guidelines separate patients into 3 primary groups: 1) GG1 with PSA 10-20 and stage T2b-c, 2) GG2 with PSA 10-20 and/or stage T2b-c and/or ≥50% biopsy cores positive, and 3) GG3 with PSA <20. We sought to utilize a large institutional radical prostatectomy (RP) cohort with long-term outcomes data to determine how UIR-PCa has evolved over time, and whether the current groupings are equally unfavorable as defined by the current guidelines. Methods: Participants were diagnosed with UIR-PCa at University of California, San Francisco in 2000-2024 and underwent RP. Disease risk was defined by clinical Cancer of the Prostate Risk Assessment (CAPRA) at diagnosis and post-surgical CAPRA (CAPRA-S) at RP. Lifetable estimates, Kaplan-Meier curves, and Cox proportional hazards regression were performed to evaluate outcomes of recurrence, defined as either biochemical failure (two PSA ≥0.2 ng/ml) or second treatment and metastasis after RP. The model was adjusted for age, PSA density, and high genomic score (OncotypeDx GPS>40, Decipher >0.6 or Prolaris>0.8). Results: 1,394 patients were included in the study. Median (IQR) follow-up was 44 (15-91) months. Clinical CAPRA at diagnosis was 6% low, 73% intermediate, and 21% high risk. From 2000 to 2023, the proportion of UIR-PCa patients defined by biopsy GG3 increased 34% to 63%. During the same period, the number of high-risk patients by CAPRA (≥6) increased from 10% to 31%. At RP, the proportion of patients with pathologic GG3 and GG4-5 increased from 22% to 41% and from 7 to 16%, respectively. Seven years following radical prostatectomy (RP), recurrence-free survival was 57% and metastasis-free survival was 92%. Kaplan-Meier analysis demonstrated that 7-year recurrence-free survival after radical prostatectomy was significantly lower for GG3 patients (48%) compared to GG1/GG2 patients (67%), log-rank p<0.01. Meanwhile, 7-year metastasis-free survival after radical prostatectomy was 90% for GG3 patients relative to 94% for GG1/GG2 patients, log-rank p<0.01. Multivariable Cox regression analysis showed a significant increase in the risk of recurrence after RP for GG3 patients compared to GG1/GG2 (HR 1.64, 95% CI 1.32 – 2.05, p<0.01, Table 1), but not for risk of metastasis. Conclusions: The study shows a significant increase in GG3 among UIR-PCa patients over time, and that UIR-PCa defined by GG3 was associated with adverse outcomes after RP. These findings suggest that the current UIR-PCa groupings as defined by the AUA are not created equal and may warrant revision to address an evolving risk profile of UIR-PCa over time. Cox proportional hazards regression for the outcome of recurrence after RP. Effect Parameter Global p Param p HR 95%CI Lower 95%CI Upper Year of diagnosis Per 10 years 0.28 1.01 0.99 1.04 Age at diagnosis Per 10 years 0.44 1.06 0.91 1.24 PSA density Logarithm <.01 1.53 1.29 1.82 Preop genomic Low vs High <.01 <.01 0.64 0.45 0.90 No test vs High <.01 0.49 0.37 0.66 Unfav subgroup GG3 vs GG1/GG2 <.01 1.64 1.32 2.05