Real-world treatment patterns and outcomes among patients with high-risk and very high-risk non-metastatic prostate cancer.

Abstract

e17120 Background: Patients with high-risk (HiR) and very high-risk (VHiR) non-metastatic prostate cancer (nmPC) are at increased risk of biochemical recurrence (BCR), distant metastasis, and death. Real-world (rw) data on treatment patterns, including androgen receptor pathway inhibitors (ARPI) use, and long-term outcomes of patients with nmPC remain limited. We assessed treatment patterns and outcomes among patients with HiR and VHiR nmPC. Methods: This retrospective cohort study included patients from the EHR-derived, deidentified Flatiron Health Research Database, which included 384,351 patients with PC. Patients diagnosed with nmPC between January 1, 2011, and November 30, 2025, were classified as HiR if they met ≥1 criterion: Gleason score (GS) ≥8, grade group (GG) 4-5, prostate-specific antigen (PSA) ≥20 ng/mL, or clinical (c)T3-T4 or VHiR if they met ≥2 criteria: GS ≥8, GG 4-5, PSA ≥40 ng/mL, or cT3-T4. Patients with cN1 or distant metastasis within 90 days of diagnosis were excluded. rw time to BCR (rwTTBCR), rw metastasis-free survival (rwMFS), and rw overall survival (rwOS) were estimated using Kaplan-Meier and Cox proportional hazards models adjusted for age and race. Results: A total of 63,045 patients met criteria (median [IQR] age, 70 [64-75] years). 89.5% were HiR and 10.5% were VHiR. Baseline demographic characteristics and ADT use were similar across risk groups. However, radiation therapy (RT) use was higher in the VHiR group (78.4% v 67.6%), while surgery was more common in HiR (41.5% v 27.6%). Within 6 months of radical prostatectomy (RP) or RT initiation, 2.6% of patients received an ARPI (VHiR, 6.2%; HiR, 2.2%). Overall, 23.6% of patients developed BCR, with similar rates across risk groups (VHiR, 23.4%; HiR, 23.6%). Among patients with BCR, 51.8% received an ARPI after BCR and prior to distant metastasis (VHiR, 64.2%; HiR, 50.4%). Outcomes for the overall cohort and by risk group are presented below (Table). In adjusted Cox models, VHiR disease was associated with shorter rwTTBCR (HR, 1.28; 95% CI, 1.22-1.34; P < .001), shorter rwMFS (HR, 1.32; 95% CI, 1.27-1.37; P < .001), and worse survival (HR, 1.35; 95% CI, 1.29-1.42; P < .001). Conclusions: This large rw cohort observed worse clinical outcomes in patients with VHiR compared with HiR. ARPI use in the localized setting remains low, with increasing use post-BCR, particularly in patients with VHiR disease. As ARPIs continue to be approved in the nmPC setting, future research should examine rw outcomes among patients receiving ARPI to improve treatment strategies for patients with nmPC at highest risk of progression. rw Outcomes overall and by risk group Overall VHiR HiR rwTTBCR, mo (95% CI) 79.0 (77.7-80.4) 64.8 (61.3-67.5) 80.7 (79.4-82.5) rwMFS, mo (95% CI) 74.8 (73.9-75.8) 60.9 (59.1-62.8) 76.5 (75.6-77.5) rwOS, mo (95% CI) 124.0 (122.5-125.6) 105.4 (101.7-109.9) 126.1 (124.1-127.9)