A pilot study of concomitant boost accelerated superfractionated radiotherapy for stage III cancer of the uterine cervix

Purpose: Retrospective studies suggest that prolonged treatment time adversely affects control rates of squamous carcinomas managed by radiotherapy. From 1989 to 1994 a prospective clinical trial was conducted to assess the feasibility and efficacy of concomitant boost accelerated superfractionated (CBASF) radiotherapy for advanced uterine cervical carcinoma. Methods and Materials: Twenty newly diagnosed patients with FIGO stage III squamous cell carcinoma of the cervix were irradiated using a CBASF regimen. Patients received 45 Gy administered to the whole pelvis in 25 fractions in 5 weeks. On Monday, Wednesday, and Friday of the last 3 weeks, an additional 1.6 Gy boost was given 6 hours after the whole pelvis treatment. The 9 boost treatments, totaling 14.4 Gy, were given via lateral fields encompassing the parametria and primary tumor for a cumulative tumor dose of 59.4 Gy. A single low-dose rate brachytherapy procedure was performed within 1 week after the external beam radiotherapy to raise the point A dose to 85-90 Gy in 42 days. Primary endpoints of analysis were local control, complications, and patterns of failure. Results are compared with the outcomes of 21 patients treated with conventionally fractionated (CF) radiotherapy during the same years. Results: Median total treatment time was 46 days in the CBASF group (range 37-62). Median follow-up interval among surviving CBASF patients is 3.8 years. The four-year actuarial local control rates are 78% and 70% in the CBASF and CF groups, respectively (p = ns). Only 2 CBASF patients required a treatment break because of acute toxicity, but severe late complications occurred in 8/20 CBASF patients for a crude rate of 40%. Distant failure was more common than local failure in the CBASF group, and para-aortic node failure occurred in six of the eight CBASF patients with distant failure. Conclusions: In the management of stage III cervix cancer, the CBASF regimen produced a trend toward improved local control when compared with the CF regimen, shifting the patterns of failure toward a higher rate of isolated distant failures. The high frequency of para-aortic node failure warrants consideration of elective treatment to this region in stage III patients treated with curative intent. Although the high local control rate of the CBASF regimen supports further investigation of accelerated treatment regimens for locally advanced cervix cancer, the unacceptable risk of late complications necessitates refinement in technique and scheduling to improve the therapeutic ratio. (C) 1997 Elsevier Science Inc.