AIM: The definition of very high-risk (VHR) prostate cancer is currently based on the study of radical prostatectomy. We aimed to identify a suitable definition for VHR group following external beam radiation therapy (EBRT) and long-term androgen-deprivation therapy (ADT). METHODS: This retrospective study included 356 high-risk patients treated with EBRT and long-term ADT. A median follow-up time was 68 months. At first, associations of previously described prognostic factors with biochemical disease-free survival (bDFS), clinical relapse-free survival (cRFS) and prostate cancer-specific survival (CSS) were examined. Second, the combination of significant adverse factors in the first analysis served as VHR test definitions. For each factor, a Cox proportional hazards model was used to calculate their hazard ratios for bDFS and cRFS. The logrank test was used to evaluate the association between each factor and CSS. RESULTS: Primary Gleason pattern 5, T4 and ≥ 5 or 4 cores with Gleason score 8-10 were risk factors associated with bDFS, cRFS and CSS. Eleven VHR test definitions composed of these adverse factors were associated significantly with bDFS, cRFS and CSS. The final definition was described by primary Gleason pattern 5 or T4 or ≥ 4 cores with Gleason score 8-10 because of the largest sample size of 38% among 11 test definitions. bDFS, cRFS and CSS of the VHR group were significantly lower compared with other high-risk patients (P < 0.001, P < 0.001 and P = 0.015, respectively). CONCLUSION: These VHR criteria were best fitted following EBRT with long-term ADT.
AIM: The definition of very high-risk (VHR) prostate cancer is currently based on the study of radical prostatectomy. We aimed to identify a suitable definition for VHR group following external beam radiation therapy (EBRT) and long-term androgen-deprivation therapy (ADT). METHODS: This retrospective study included 356 high-risk patients treated with EBRT and long-term ADT. A median follow-up time was 68 months. At first, associations of previously described prognostic factors with biochemical disease-free survival (bDFS), clinical relapse-free survival (cRFS) and prostate cancer-specific survival (CSS) were examined. Second, the combination of significant adverse factors in the first analysis served as VHR test definitions. For each factor, a Cox proportional hazards model was used to calculate their hazard ratios for bDFS and cRFS. The logrank test was used to evaluate the association between each factor and CSS. RESULTS: Primary Gleason pattern 5, T4 and ≥ 5 or 4 cores with Gleason score 8-10 were risk factors associated with bDFS, cRFS and CSS. Eleven VHR test definitions composed of these adverse factors were associated significantly with bDFS, cRFS and CSS. The final definition was described by primary Gleason pattern 5 or T4 or ≥ 4 cores with Gleason score 8-10 because of the largest sample size of 38% among 11 test definitions. bDFS, cRFS and CSS of the VHR group were significantly lower compared with other high-risk patients (P < 0.001, P < 0.001 and P = 0.015, respectively). CONCLUSION: These VHR criteria were best fitted following EBRT with long-term ADT.