Prediction of time to hormonal treatment failure in metastatic castration-sensitive prostate cancer with 18F-FDG PET/CT

Abstract

The aim of this prospective investigation was to assess the association of 18F-FDG PET/CT with time to hormonal treatment failure (THTF) in men with metastatic castration-sensitive prostate cancer. Methods: 76 men with metastatic castration-sensitive prostate cancer recruited from 2005 to 2011 underwent 18F-FDG PET/CT and were followed prospectively for THTF, defined as treatment change to chemotherapy or death. Patients who had not switched to chemotherapy were censored at the last follow-up date (median of 36 mo; range, 12–108 mo). Cox regression analyses were performed to examine the association between PET/CT measurements: sum of SUVmax, maximum SUVmax, and average SUVmax for up to 10 of the most active lesions and THTF. Survival probabilities were based on the Kaplan–Meier method. Results: 43 patients had hormonal treatment failure, and 8 died without documented treatment failure. Median THTF was 26.5 mo (95% confidence interval [CI], 15.5–46.6 mo). The THTF-free probability at 5 y was 35% ± 6%. On univariate analysis, all PET parameters, including number of lesions, were statistically significant for THTF. In a reduced multivariate model accounting for clinical variables, only sum of SUVmax (hazard ratio, 1.01; 95% CI, 1.002–1.03; P 5 0.024) and number of lesions (hazard ratio, 1.18; 95% CI, 1.08–1.29; P, 0.001) were independently associated with THTF. When sum of SUVmax was grouped into quartile ranges, there was a significantly worse survival probability for patients in the fourth-quartile range than in the first, with a univariate hazard ratio of 6.2 (95% CI, 2.8–13.6; P, 0.001). Conclusion: Sum of SUVmax and number of lesions derived from 18F-FDG PET/CT provide independent prognostic information on THTF in men with metastatic castration-sensitive prostate cancer.