Baseline biochemical and clinical predictors of castration resistance and survival in metastatic hormone-sensitive prostate cancer treated with first-line androgen receptor pathway inhibitors: A multicenter real-world study.

108 Background: Androgen receptor pathway inhibitors (ARPIs) plus androgen deprivation therapy (ADT), with or without docetaxel, are standard for metastatic hormone-sensitive prostate cancer (mHSPC). Predictors of progression and survival in real-world populations remain limited. We evaluated baseline biochemical and clinical factors associated with time to castration resistance (tCRPC) and overall survival (OS). Methods: We retrospectively analyzed 333 patients treated Jan 2020–Apr 2025 with abiraterone (ABI), apalutamide (APA), enzalutamide (ENZ), or darolutamide (DARO) plus ADT, with or without docetaxel. Baseline variables included age, ECOG PS, disease volume (CHAARTED), risk (LATITUDE), ISUP grade, visceral/liver metastases, LDH and ALP. Endpoints were tCRPC and OS; PSA response was assessed, with undetectable PSA ≤0.2 ng/mL. Kaplan–Meier and multivariable Cox regression analyses were performed. Results: Median age was 70; 142 received ABI, 157 APA, 28 DARO, 6 ENZ; 258 received doublet, 75 triplet therapy. Synchronous metastases were present in 229, high-volume disease in 186, high-risk features in 246. At 3 months, PSA50/PSA90 responses were 79%/64%, undetectable PSA in 30%. Median follow-up was 48 months. At 24 months, overall castration-resistance rate was 76%. High-volume rates: 68% doublets (67% APA, 70% ABI), 65% triplets (12m: 88% ABI, 86% DARO); low-volume: 88% doublets (86% APA, 92% ABI), 60% triplets (ABI). OS rates at 24 months: high-volume 61% doublets (56% ABI, 65% APA), 59% triplets (12m: 75% ABI, 84% DARO); low-volume 90% doublets (84% ABI, 94% APA), 72% triplets (80% ABI). After multivariable Cox analysis, high-volume predicted shorter tCRPC (HR 3.25, 95% CI 1.67–6.32, p<0.001) and worse OS (HR 2.58, 95% CI 1.44–4.62, p=0.001). LDH >220 U/L associated with shorter tCRPC (HR 2.79, 95% CI 1.35-5.80, p=0.06) and worse OS (HR 2.20, 95% CI 1.11-4.37, p=0.024). ALP >350 U/L, ISUP grade, and visceral metastases were not significant. In high-volume patients, LDH >220 U/L strongly predicted tCRPC in triplets (HR 20.0, p=0.006) but not in doublets. Conclusions: In this large real-world dataset, ARPI-based therapy achieved high PSA responses and durable disease control. High-volume disease and elevated LDH were the strongest predictors of earlier castration resistance and shorter OS, while ALP, ISUP grade, and visceral metastases were not prognostic. High LDH was a predictive factor for treatment outcomes in high-volume. This clinical parameter might be integrated with genomic markers to guide treatment intensification in mHSPC, specially in high-volume disease.