Compliance to the prescribed dose and overall treatment time in five randomized clinical trials of altered fractionation in radiotherapy for head-and-neck carcinomas

Abstract

Purpose: To investigate compliance to the prescribed dose-fractionation schedule in five randomized controlled trials of altered fractionation in radiotherapy for head-and-neck carcinoma. Methods and Materials: Individual patient data from 2566 patients participating in the European Organization for Research and Treatment of Cancer (EORTC) 22791, EORTC 22811, EORTC 22851, Princess Margaret Hospital (PMH), and continuous hyperfractionated accelerated radiotherapy (CHART) head-and-neck trials were merged in the fractionation IMPACT (Intergroup Merger of Patient data from Altered or Conventional Treatment schedules) study database. The ideal treatment time was defined as the minimum time required to deliver a prescribed schedule. Compliance to the prescribed overall treatment time was quantified as the difference between the actual and the ideal overall time. An overall measure of compliance in an individual patient, the total dose lost (TDL), was calculated as the dose lost due to prolongation of therapy (assuming a D prolif of 0.64 Gy/day) plus the difference between the prescribed and the actual dose given. Results: The time in excess of the ideal ranged up to 97 days (average 3.9 days), and 25% of the patients had delays of 6 days or more. World Health Organization (WHO) performance status and nodal stage had a significant effect on TDL. TDL was significantly higher in the conventional than in the altered arm of the EORTC 22851 and CHART trials. In the PMH trial, TDL was significantly higher in the hyperfractionation than in the conventional arm. Centers participating in the three EORTC trials varied significantly in their compliance. There was a significant improvement in compliance in patients treated more recently. Conclusions: Even in randomized controlled trials, compliance to the prescribed radiation therapy schedule may be relatively poor, especially after conventional fractionation. This affects the interpretation of the outcome of these trials.