Landmark-Based PSA90 and PSA Nadir Stratification in Androgen Receptor Pathway Inhibitor–Treated Metastatic Hormone-Sensitive Prostate Cancer

Background Prostate-specific antigen (PSA) decline is commonly used to monitor treatment response in metastatic hormone-sensitive prostate cancer (mHSPC). However, PSA90 at an early time point may not fully capture the heterogeneity of outcomes in patients receiving androgen deprivation therapy plus androgen receptor pathway inhibitors. We evaluated a landmark-based PSA response framework integrating 3-month PSA90 status and PSA nadir depth by 12 months. Methods This multicenter retrospective study included patients with mHSPC treated with androgen deprivation therapy plus abiraterone, enzalutamide, or apalutamide between January 2015 and December 2023. The primary analysis used a 12-month landmark approach. Patients alive and free of radiographic progression at 12 months were classified into three groups: Group 1, PSA90 at 3 months and PSA nadir <0.02 ng/mL by 12 months; Group 2, PSA90 at 3 months without nadir <0.02 ng/mL; and Group 3, no PSA90 at 3 months. The primary endpoint was radiographic progression-free survival (rPFS); overall survival (OS) was secondary. Results Among 206 initially identified patients, 160 were eligible for the 12-month landmark cohort. Group 1 included 53 patients, Group 2 included 76, and Group 3 included 31. Median rPFS was not reached in Group 1, compared with 18.9 months in Group 2 and 20.2 months in Group 3. In multivariable analysis, Group 2 and Group 3 had shorter rPFS than Group 1, with adjusted hazard ratios of 9.18 and 9.81, respectively. For OS, the corresponding adjusted hazard ratios were 3.76 and 4.76. In exploratory model-performance analyses, inclusion of the three-level PSA response group was associated with improved apparent model fit compared with clinical variables alone and PSA90 status alone. Conclusion In ARPI-treated mHSPC, PSA90 at 3 months was prognostically informative but insufficient alone. Combining PSA90 with PSA nadir depth by 12 months identified a favorable prognostic subgroup, but independent external validation is required before routine clinical implementation.