Defining a dose-response relationship for prostate external beam radiotherapy.

INTRODUCTION We aimed to quantify a relationship between radiotherapy dose and freedom from biochemical failure (FFBF) in low- and intermediate-risk prostate cancer. To reduce confounding we used data with a standardised end-point, mature follow-up, low competing risk of metastatic failure, conventional fractionation and separate reporting for outcomes with hormonal therapy (HT). METHODS A systematic review of the literature was carried out. Studies that reported the use of radiotherapy alone in 1.8-2 Gy fractions in low- and intermediate-risk prostate cancer were included. The primary end-point was Phoenix definition 5-year FFBF. A logistic regression was used to quantify the dose-response relationship. RESULTS Data from eight studies with 3037 patients met the inclusion criteria. The data from 810 low-risk patients and 2245 intermediate-risk patients were analysed. A strong association between radiotherapy dose and FFBF was found in low- and intermediate-risk patients managed with radiotherapy alone. In low-risk patients not treated with HT the dose required to achieve 50% biochemical tumour control (TCD50 ) is 52.0 Gy and the slope of the dose-response curve at TCD50 (γ50 ) is 2.1%/Gy. At 78 Gy this represented a FFBF of 90.3%. In intermediate-risk patients not treated with HT the TCD50 is 64.7 Gy and γ50 is 3.2%/Gy. At 78 Gy this translated into a FFBF of 84.3%. HT had a small effect for low-risk patients and an inconsistent effect for intermediate-risk men. CONCLUSION A strong association was found between radiation dose and biochemical outcome in both low- and intermediate-risk patients. Standardised reporting of results from future studies will make future analyses more robust.