Purpose: To identify the clinical and dosimetric factors predictive of a greater risk of Grade 2 or worse late rectal bleeding in patients with localized prostate cancer treated with three-dimensional conformal radiotherapy in a prospective dose-escalation study.
Methods and materials: We performed a retrospective analysis of the clinical records and dose-volume histograms of 107 patients with Stage T1c-T3 prostate cancer treated at our institution with three-dimensional conformal radiotherapy who had a minimal follow-up of 1 year. Of the 107 patients, 21 were treated at dose level 1 (70.0 Gy), 57 at dose level 2 (72.0 Gy), and 29 at dose level 3 (75.6 Gy). The mean International Commission on Radiation Units and Measurements reference dose was 76.5 Gy (range, 69.8-82.6 Gy).
Results: The 4-year actuarial incidence of Grade 2 or worse late rectal bleeding was 7.7% +/- 2.5%. The results of the multivariate analysis indicated that the mean rectal dose (rectal D(mean); p = 0.003) and the percentage of rectum receiving >60 Gy (Vr(60); p = 0.002) correlated with Grade 2 or worse rectal bleeding. The receiver operating characteristic curve analysis showed that this logistic regression model using both Vr(60) and rectal D(mean) had good reliability to predict the risk of late rectal bleeding. The area under the curve for Vr(60) and rectal D(mean) was 0.889 and 0.892, respectively.
Conclusion: The results of the present study provide clear evidence of a dose-volume effect and the importance of intermediate doses (60.0 Gy) on the risk of rectal bleeding at this prescription dose level. On the basis of these results, new constraints have been implemented in our institution to keep the risk of developing Grade 2 rectal bleeding reasonably low (rectal D(mean) 50.0 Gy and Vr(60) 42%).