Radiotherapy for T2 and T3 carcinoma of the bladder: the influence of overall treatment time.

The influence of overall treatment time on local control rate was studied on a group of 147 patients with muscle invasive T2 or T3 transitional cell carcinoma of the urinary bladder. All patients received external radiotherapy at the Catharina Hospital, Eindhoven, The Netherlands between January 1974 and December 1984. Patients treated with overall treatment times shorter than 75 days (n = 92) were irradiated during a continuous course; all but one patient, with overall treatment times of 75 days or more (n = 55), received split-course radiotherapy. Actuarial local relapse-free probability at 3 years (LRFP3) was computed from the onset of radiotherapy. LRFP3 proved to be dependent on overall treatment time. For T2 stage, LRFP3 was 80 +/- 18% (n = 5) and 54 +/- 13% (n = 13) for overall times between 15-44 and 45-74 days, respectively, 36 +/- 14% (n = 11) for overall times between 75 and 104 days and 64 +/- 15% (n = 11) for overall times longer than 105 days. For T3 stage, LRFP3 was 33 +/- 19% (n = 6) and 48 +/- 10% (n = 25) for overall times between 15-44 and 45-74 days, respectively, 25 +/- 14% (n = 12) for overall times between 75 and 104 days and 22 +/- 14% (n = 9) for overall times longer than 105 days. The figures between brackets are numbers of patients relapsing within 3 years or at risk of relapse during at least 3 years. Patients who died without local relapse before 3 years were censored. We have reasons to believe that patient selection bias leads to overestimation of LRFP3 for the split-course radiotherapy in retrospective studies where the 'intention to treat' cannot be recalled. This retrospective study suggests that prolonging overall time of radiotherapy has an effect on local control in T2 and T3 transitional cell carcinoma of the urinary bladder. Local control was the worst for patients treated by split-course radiotherapy with a gap of approximately one month. Local control was not further decreased (and seemed even improved) by longer gaps, but this observation is possibly biased as explained in the discussion section. For patients treated by continuous course radiotherapy we could not find a difference in local control rates between patients treated with overall times of 44 days or less and those treated with overall times of 45-74 days.