How low do you need to go? Association between various prostate-specific antigen (PSA) response measures and clinical outcomes in metastatic castration-sensitive prostate cancer (mCSPC) in the Veteran Health Administration (VHA) data.

Abstract

5092 Background: Clinical trials show that combined treatment (tx) with androgen deprivation therapy (ADT) and androgen receptor pathway inhibitors (ARPIs) improves PSA response and overall survival (OS) in patients (pts) with mCSPC. Our real-world study assessed the impact of PSA response on OS and progression in pts receiving ADT ± other tx. Methods: VHA data (2017–2024) were analyzed for adults with mCSPC who initiated index tx, ie, ADT ± other tx (ARPI, nonsteroidal antiandrogen [NSAA], docetaxel), and had PSA values at baseline (365 days prior to ADT initiation [index date]) and during index tx. PSA response was examined as: 1) ≥90% decline from baseline PSA; 2) PSA <0.2 ng/mL during index tx. Progression was defined as PSA progression (≥25% rise and ≥2 ng/mL increase from nadir PSA during PSA follow-up [index date to 8/31/2024]), initiation of a new tx, hormone resistance or death. Landmark analyses with Cox proportional hazards regressions assessed the association of PSA response (by 9 months [mo] post index date) with OS and time to progression after 9 mo post index date. Results: Overall, 4890 pts started first line mCSPC tx: ADT alone, 47%; ADT + ARPI, 40%; ADT + NSAA, 7%; ADT + docetaxel (± ARPI ± NSAA), 6%. Median follow-up was 24.7 mo, with a median PSA follow-up of 14.6 mo. During PSA follow-up, 44% of pts reached PSA <0.2 ng/mL and 74% had ≥90% PSA decline from baseline. PSA decline of ≥90% was related to a 22% reduced risk of death but was unrelated to progression (Table). In contrast, PSA <0.2 ng/mL was associated with greater reduction in the risk of death (54%) and progression (55%). Conclusions: In a real-world mCSPC setting, pts who achieved PSA <0.2 ng/mL within 9 mo of initiating ADT ± other tx had lower risk of death and progression than pts who did not. Reaching ≥90% PSA decline was modestly associated with improved OS but not progression, suggesting a PSA nadir of <0.2 ng/mL is needed for optimal outcomes. Disclosure: A genAI tool (01/09/25; Pfizer; GPT-4o) developed the 1 st draft; authors assume content responsibility. ≥90% PSA decline PSA <0.2 ng/mL Achieved Not achieved † Achieved Not achieved † OS ‡§ Pts at risk, n 2609 683 1512 1780 Events, n (%) 672 (26) 224 (33) 236 (16) 660 (37) Median (95% CI), mo NR (53, NE) 50 (39, NE) NR (NE, NE) 37 (33, 42) HR (95% CI); P value 0.78 (0.66, 0.91); <0.001 – 0.46 (0.40, 0.54); <0.001 – Time to progression § ║ Pts at risk, n 2395 569 1444 1520 Events, n (%) 1060 (44) 271 (48) 440 (31) 891 (59) Median (95% CI), mo 27 (24, 29) 26 (20, 32) 51 (44, 57) 15 (13, 16) HR (95% CI); P value 0.93 (0.81, 1.06); 0.28 – 0.45 (0.40, 0.50); <0.001 – † Comparator group. ‡ Time from post-index day 270 to death. § Adjusted for age, race, region, index year, log of time from metastasis to index date, site of metastasis and comorbidities. ║ Time from post-index day 270 to first evidence of disease progression.