Mathematical model for evaluating incidence of acute rectal toxicity during conventional or hypofractionated radiotherapy courses for prostate cancer.

Abstract

To describe the radiation-induced acute rectal toxicity (ART) using a modified Lyman-Kutcher-Burman normal tissue complication probability model and parameters set, taking into account the overall treatment time. A total of 160 patients underwent three-dimensional conformal radiotherapy to the prostate and seminal vesicles and were randomized to receive 80 Gy in 40 fractions within 8 weeks (Group A) or 62 Gy in 20 fractions within 5 weeks, 4 d/wk (Group B). An additional 52 patients (Group C) underwent intensity-modulated radiotherapy with a hypofractionation schedule consisting of 56 Gy, delivered in 16 fractions (4/wk) of 3.5 Gy. Patients were followed for ART weekly during treatment. The overall treatment time, rectal dose-volume histograms, and ART status, defined as Radiation Therapy Oncology Group Grade 2 or greater gastrointestinal toxicity, were used to determine the modified Lyman-Kutcher-Burman model parameters. The m and n values were obtained from the cohort, and the tolerance doses for 50% complication probability for uniform irradiation [TD(50)(1)(k)] were obtained for each fractionation schedule indicated with k. Of 212 patients treated with localized prostate radiotherapy, 65 developed Grade for > or = 1 week during treatment. The m and n value was 0.17 and 0.08, respectively. The TD(50)(1)(k) parameter was 79, 62.5, and 53 Gy, respectively for Group A, B, and C. The optimized modified Lyman-Kutcher-Burman normal tissue complication probability model allowed us to describe the ART data from conventional and hypofractionated regimens, using the dose-volume histograms and overall treatment time. This model could prove useful in designing hypofractionation schedules to reduce the incidence of ART.