Purpose: To identify dosimetric variables predictive of acute gastrointestinal (GI) and genitourinary (GU) toxicity and to determine whether hormonal therapy (HT) is independently associated with acute GI and GU toxicity in prostate cancer patients treated with conformal radiotherapy (RT).
Methods and materials: This analysis was performed on 336 patients participating in a multicenter (four hospitals) randomized trial comparing 68 Gy and 78 Gy. The clinical target volume consisted of the prostate with or without the seminal vesicles, depending on the risk of seminal vesicle involvement. The margin from the clinical target volume to the planning target volume was 1 cm. For these patients, the treatment plan for a total dose of 68 Gy was used, because nearly all toxicity appeared before the onset of the 10-Gy boost. Acute toxicity (<120 days) was scored according to the Radiation Therapy Oncology Group criteria. The dosimetric parameters were obtained from the relative and absolute dose-volume/surface histograms derived from the rectal wall (rectal wall volume receiving > or =5-65 Gy) and the bladder surface (bladder surface receiving > or =5-65 Gy). Additionally, relative and absolute dose-length histograms of the rectum were created, and the lengths of rectum receiving more than a certain dose over the whole circumference (rectal length receiving > or =5-65 Gy) were computed. The clinical variables taken into account for GI toxicity were neoadjuvant HT, hospital, and dose-volume group; for GU toxicity, the variables pretreatment GU symptoms, neoadjuvant HT, and transurethral resection of the prostate were analyzed. The variable neoadjuvant HT was divided into three categories: no HT, short-term neoadjuvant HT (started < or =3 months before RT), and long-term neoadjuvant HT (started >3 months before RT).
Results: Acute GI toxicity Grade 2 or worse was seen in 46% of the patients. Patients with long-term neoadjuvant HT experienced less Grade 2 or worse toxicity (27%) compared with those receiving short-term neoadjuvant HT (50%) and no HT (50%). The volumes of the prostate and seminal vesicles were significantly smaller in both groups receiving neoadjuvant HT compared with those receiving no HT. In multivariate logistic regression analysis, including the two statistically significant clinical variables neoadjuvant HT and hospital, a volume effect was found for the relative, as well as absolute, rectal wall volumes exposed to intermediate and high doses. Of all the length parameters, the relative rectal length irradiated to doses of > or =5 Gy and > or =30 Gy and absolute lengths receiving > or =5-15 and 30 Gy were significant. Acute GU toxicity Grade 2 or worse was reported in 56% of cases. For patients with pretreatment GU symptoms, the rate was 93%. The use of short-term and long-term neoadjuvant HT resulted in more GU toxicity (73% and 71%) compared with no HT (50%). In multivariate analysis, containing the variables pretreatment symptoms and neoadjuvant HT, only the absolute dose-surface histogram parameters (absolute surface irradiated to > or =40, 45, and 65 Gy) were significantly associated with acute GU toxicity.
Conclusion: A volume effect was found for acute GI toxicity for relative, as well as absolute, volumes. With regard to acute GU toxicity, an area effect was found, but only for absolute dose-surface histogram parameters. Neoadjuvant HT appeared to be an independent prognostic factor for acute toxicity, resulting in less acute GI toxicity, but more acute GU toxicity. The presence of pretreatment GU symptoms was the most important prognostic factor for GU symptoms during RT.