Time Factor For Radiotherapy Of Prostate Cancer: Retrospective Analysis Of Biochemical Failure In 4839 Men Treated Between 1987 And 1995

Assess the importance of overall time (OT) and dose for biochemical failure (BF) after external-beam radiotherapy of prostate cancer in a retrospective analysis of a nine-institution database with 4839 patients treated at the institutions of our collaborators EM Horwitz (Fox Chase Cancer Center), P Kupelian (UCLA), A Martinez (William Beaumont Hospital), J Michalski (Mallinckrodt Institute of Radiology), T Pisansky (Mayo Medical School), H Sandler (University of Michigan), WU Shipley and A Zietman (Massachusetts General Hospital), and M Zelefsky (Memorial Sloan Kettering Cancer Center). Relevant baseline factors (T stage, Gleason score, initial PSA) were available for 4338 men. Cox models were used to estimate the effects of dose and OT corrected for baseline factors, treatment year, institution and interactions, and differences in post-treatment PSA measurement intervals. After exclusion of very short and long intervals, patient numbers were 1445 events / 3426 at risk (endpoint all BFs), and 1177 events / 3354 at risk (endpoint exclusion of BFs that were likely distant failures). Separate analyses were carried out by risk group for men who received < 70 Gy and ≥ 70 Gy. Neither dose nor OT was significant when the analysis was restricted to doses < 70 Gy, while for patients treated to 70 Gy or higher there were significant influences of both dose and OT on outcome in low- and intermediate-risk patients. These findings are consistent with reports of no time factor for patients treated in the pre-PSA era to doses predominantly less than 70 Gy (e.g., Lai et al. IJROBP 19:561,1990), vs. reports of a time factor for patients treated to doses higher than 70 Gy (d'Ambrosio et al. IJROBP 72:1402,2008). The time and dose effects were quantified as a relative increase after 5 years follow-up of 6% in BFs for a 1-week increase in OT, a relative decrease of 15% in BFs for a 6- Gy increase in dose, and a dose equivalent of proliferation of 0.24 Gy/day. As the dose per fraction was nearly constant, the data contain no information on the α/β ratio. Keeping in mind the usual caveats that attend a retrospective study such as the present one, the results show that OT and dose are significant determinants of outcome of radiotherapy in low- and intermediate-risk patients treated to 70 Gy or higher, and suggest that meaningful improvements in outcome may be targeted by modest increases in total dose and decreases in OT.