Systematic review of the effect of radiation dose on tumor control and morbidity in the treatment of prostate cancer by 3D-CRT

Purpose: A higher radiation dose is believed to result in a larger probability of tumor control and a higher risk of side effects. To make an evidence-based choice of dose, the relation between dose and outcome needs to be known. This study focuses on the dose-response relation for prostate cancer. Methods and Materials: A systematic review was carried out on the literature from 1990 to 2003. From the selected studies, the radiation dose, the associated 5-year survival, 5-year bNED (biochemical no evidence of disease), acute and late gastrointestinal (GI) and genitourinary (GU) morbidity Grade 2 or more, and sexual dysfunction were extracted. With logistic regression models, the relation between dose and outcome was described. Results: Thirty-eight studies met our criteria, describing 87 subgroups and involving up to 3000 patients per outcome measure. Between the (equivalent) dose of 70 and 80 Gy, various models estimated an increase in 5-year survival (ranging from 10% to 11%), 5-year bNED for low-risk patients (5-7%), late GI complications (12-16%), late GU complications (8-10%), and erectile dysfunction (19-24%). Only for the overall 5-year bNED, results were inconclusive (range, 0-18%). Conclusions: The data suggest a relationship between dose and outcome measures, including survival. However, the strength of these conclusions is limited by the sometimes small number of studies, the incompleteness of the data, and above all, the correlational nature of the data. Unambiguous proof for the dose-response relationships can, therefore, only be obtained by conducting randomized trials. (C) 2006 Elsevier Inc.