Predictors for Rectal and Intestinal Acute Toxicities During Prostate Cancer High-Dose 3D-CRT : Results of a Prospective Multicenter Study

Purpose: To find predictors for rectal and intestinal acute toxicity in patients with prostate cancer treated with ≥70 Gy conformal radiotherapy. Methods and Materials: Between July 2002 and March 2004, 1,132 patients were entered into a cooperative study (AIROPROS01-02). Toxicity was scored using the Radiation Therapy Oncology Group/European Organization for Research and Treatment of Cancer scale and by considering the changes (before and after treatment) of the scores of a self-administered questionnaire on rectal/intestinal toxicity. The correlation with a number of parameters was assessed by univariate and multivariate analyses. Concerning the questionnaire, only moderate/severe complications were considered. Results: Of 1,132 patients, 1,123 were evaluable. Of these patients, 375, 265, and 28 had Grade 1, 2, and 3 Radiation Therapy Oncology Group/European Organization for Research and Treatment of Cancer toxicity, respectively. The mean rectal dose was the most predictive parameter (p = 0.0004; odds ratio, 1.035) for Grade 2 or worse toxicity, and the use of anticoagulants/antiaggregants (p = 0.02; odds ratio, 0.63) and hormonal therapy (p = 0.04, odds ratio, 0.65) were protective. The questionnaire-based scoring revealed that a greater mean rectal dose was associated with a greater risk of bleeding; larger irradiated volumes were associated with frequency, tenesmus, incontinence, and bleeding; hormonal therapy was protective against frequency and tenesmus; hemorrhoids were associated with a greater risk of tenesmus and bleeding; and diabetes associated highly with diarrhea. Conclusion: The mean rectal dose correlated with acute rectal/intestinal toxicity in three-dimensional conformal radiotherapy for prostate cancer, and hormonal therapy and the use of anticoagulants/antiaggregants were protective. According to the moderate/severe injury scores on the self-assessed questionnaire, several clinical and dose-volume parameters were independently predictive for particular symptoms.