Inferior Clinical Outcomes for Patients With Positive Post-Radiation Therapy Prostate Biopsy: Results From Prospective Randomized Trial RTOG 94-08

Anatomic recurrence patterns were classified as a local, lymphotropic, or osteotropic pattern if FFS and all subsequent RS were confined to the prostate/SVonly, LNs only, or bones only, respectively, for at least 2 years. The median follow-up (f/u) for all patients and those with BF was 7.0 and 10.1 years (yr), respectively. Median f/u after BF was 4.7 yr. Results: For the entire cohort, the 10 yr LF rate was 6%, 14%, and 17% for the LR, IR, and HR groups (overall p < 0.001). The 10 yr risk of isolated PLN relapse as a FFS was 0%, 1%, and 4%, respectively (p < 0.001). We next focused on outcomes following BF in the 602 patients with BF. At 5 yr, the rates of BM and PCSM were 42% and 19%. The most common FFS was local (42% of FFS), followed by bone (26%), PLN (17%), and retroperitoneal LN (7%). Patients displayed unique anatomic distributions of relapse, with local (31%), lymphotropic (7%), osteotropic (16%), and multiorgan patterns (22%). These anatomic patterns had markedly different prognoses, with 5 yr PCSM of 2%, 15%, 32%, and 48%, respectively (p < 0.001). Anatomic recurrence pattern was an independent predictor for PCSM (overall p < 0.001) and OS (overall p < 0.001) in MVA amongst patients with clinical failure. Gleason score 8-10, stage T34, and time to BF < 5 yr predicted for BM and PCSM following BF in MVA. Conclusions: In the dose escalation era, the most common first site of failure is within the prostate and SV, even for HR patients, suggesting improved local cancer eradication via addition of brachytherapy and/or radiosensitizers may be beneficial. On the other hand, patients treated without PLN EBRT have a relatively low risk of isolated PLN relapse. Failure patterns displayed a tropism for specific anatomic distributions, with divergent prognoses, suggesting underlying biological differences amongst different tumors. Author Disclosure: Z.S. Zumsteg: None. D.E. Spratt: None. P. Romesser: None. X. Pei: None. W. Polkinghorn: None. M. Kollmeier: None. Y. Yamada: D. Employment Other; Speakers Bureau, Institute for Medical Education. G. Consultant; Varian Medical Systems. M.J. Zelefsky: None.