Hypofractionated Radiotherapy as Primary Therapy for Prostate Cancer: Randomised Trial Comparing Toxicity Between 2 Different Hypofractionated Schedules

External beam radiotherapy (RT) is one of the standard curative treatment options for patients with prostate cancer (PC). Several randomised trials have shown excellent long-term biochemical outcome with higher radiation doses. Nowadays, RT for PC commonly consists of delivering 74-80 Gy in 2 Gy fractions, resulting in an overall treatment time of 7-8 weeks. The sensitivity of different tissues to fractionation changes can be quantified through the α/β ratio in the linear-quadratic model. Dose-response analysis of PC patients treated with both external beam RT and brachytherapy has led to the hypothesis that the α/β ratio of PC is lower than for most other tumors and approaches a value characteristic of late responding tissues. Values between 1.2 and 3.9 Gy have been calculated. If the α/β ratio of PC is indeed low, then hypofractionating RT treatments can theoretically maintain high bioequivalent tumor doses, shorten overall treatment time and decrease late toxicities.The advantages in terms of patient convenience and treatment cost are obvious. There is level I evidence that shows that hypofractionated radiotherapy schedules have at least equivalent biochemical outcome with only a small increase in acute but not late toxicity when compared to conventional fractionation RT schedules.

Results on different hypofractionation schedules have been reported, however the optimal hypofractionation is not clear so far. In this randomised trial we would like to compare 2 different radiotherapyschedules: 16 fractions à rato of 4 fractions a week versus 25 fractions à rato of 5 fractions a week. The incidence on acute toxicity and early late toxicity (i.e. within 2 year post radiotherapy) and the impact on quality of life will be registrated and compared. The study will be performed in 2 stages. For stage 1, sample size was calculated to rule out an upper limit of 40% of patients with RTOG grade 2 or worse bowel (GI) complications with an expected rate of 25%, based on a one-stage Fleming-A'Hern design. A power of 83.0% (alpha level 0.038 one-sided) was obtained when including 72 patients per group (144 patients in total). If 22 or more patients out of 72 had grade 2 or worse GI complications, then the study arm was to be rejected. To allow for a dropout of 10%, 160 patients were included in stage 1. Sample size for stage 2 was calculated analogously allowing ruling out an upper limit of 35% of patients with RTOG grade 2 or worse GI complications with an expected rate of 25%. When including 155 patients per group (310 in total) a power of 85.7% (alpha level 0.049 one-sided) was obtained. If 45 or more patients out of 155 had grade 2 or worse GI complications, then the study arm was to be rejected. The sample size for stage 1 and stage 2 combined was set at 346 (173 per group), with a 10% allowance for dropout.