PSA nadir predicts biochemical and distant failures after external beam radiotherapy for prostate cancer: A multi-institutional analysis

Purpose: To determine the significance of prostate-specific antigen (PSA) nadir (nPSA) and the time to nPSA (T_nPSA) in predicting biochemical or clinical disease-free survival (PSA-DFS) and distant metastasis-free survival (DMFS) in patients treated with definitive external beam radiotherapy (RT) for clinical Stage T1b-T2 prostate cancer. Methods and Materials: Nine participating institutions submitted data on 4839 patients treated between 1986 and 1995 for Stage T1b-T2cN0-NxM0 prostate cancer. All patients were treated definitively with RT alone to doses ≥60 Gy, without neoadjuvant or planned adjuvant androgen suppression. A total of 4833 patients with a median follow-up of 6.3 years met the criteria for analysis. Two endpoints were considered: (1) PSA-DFS, defined as freedom from PSA failure (American Society for Therapeutic Radiology and Oncology definition), initiation of androgen suppression after completion of RT, or documented local or distant failure; and (2) DMFS, defined as freedom from clinically apparent distant failure. In patients with failure, nPSA was defined as the lowest PSA measurement before any failure. In patients without failure, nPSA was the lowest PSA measurement during the entire follow-up period. T_nPSA was calculated from the completion of RT to the nPSA date. Results: A greater nPSA level and shorter T_nPSA were associated with decreased PSA-DFS and DMFS in all patients and in all risk categories (low [Stage T1b, T1c, or T2a, Gleason score ≤6, and PSA level ≤10 ng/mL], intermediate [Stage T1b, T1c, or T2a, Gleason score ≤6, and PSA level >10 but ≤20 ng/mL, or Stage T2b or T2c, Gleason score ≤6, and PSA level ≤20 ng/mL, or Gleason score 7 and PSA level ≤20 ng/mL], and high [Gleason score 8-10 or PSA level >20 ng/mL]), regardless of RT dose. The 8-year PSA-DFS and DMFS rate for patients with nPSA <0.5 ng/mL was 75% and 97%; nPSA ≥0.5 but <1.0 ng/mL, 52% and 96%; nPSA ≥1.0 but <2.0 ng/mL, 40% and 91%; and nPSA ≥2.0 ng/mL, 17% and 73%, respectively. The 8-year PSA-DFS and DMFS rate for patients with T_nPSA <6 months was 27% and 66%; T _nPSA ≥6 but <12 months, 31% and 85%; T_nPSA ≥12 but <24 months, 42% and 94%; and T_nPSA ≥24 months, 75% and 99%, respectively. A shorter T_nPSA was associated with decreased PSA-DFS and DMFS, regardless of the nPSA. Both nPSA and T_nPSA were significant predictors of PSA-DFS and DMFS in multivariate models incorporating clinical stage, Gleason score, initial PSA level, and RT dose. The significance of nPSA and T_nPSA was supported by landmark analysis, as well as by analysis of nPSA and T_nPSA as time-dependent covariates. A dose ≥70 Gy was associated with a lower nPSA level and longer T_nPSA in all risk categories, and a greater dose was significantly associated with greater PSA-DFS and DMFS in multivariate analysis. Regression analysis confirmed that higher clinical stage, Gleason score, and initial PSA were associated with a greater nPSA level. Conclusion: The results of this large, multi-institutional analysis of 4833 patients have provided important evidence that nPSA and T_nPSA after definitive external beam RT are not only predictive of a predominantly PSA endpoint (PSA-DFS), but are also predictive of distant metastasis in all clinical risk categories. Greater RT doses were associated with lower nPSA, longer T_nPSA, and improved PSA-DFS and DMFS.