Hyperfractionated radiotherapy of human tumors: overview of the randomized clinical trials.

PURPOSE Hyperfractionation (HF) is the altered fractionation schedule most frequently studied in clinical Phase III trials. In this overview, surviving fractions, rates of complete responses, and estimates of the long-term locoregional tumor control probabilities after HF and conventional fractionated irradiation (CF) available from the various reports were compared. METHODS AND MATERIALS A metaanalysis was performed of the randomized studies on hyperfractionation vs. conventional fractionation published since 1980 on different tumor types in various locations. RESULTS Compared with CF, HF significantly reduced the odds of death for patients with head and neck tumors (three studies, odds ratio 0.48 (0.40-0.58), p < 0.0001) and bladder cancer (two studies, odds ratio 0.53 (0.36-0.78), p = 0.001), while there was a trend in nonsmall cell lung cancer (three studies, odds ratio 0.69 (0.51-0.95), p = 0.02), and malignant gliomas (three studies, odds ratio 0.67 (0.48-0.93), p = 0.02). The probability of long-term loco-regional control of head and neck tumors was significantly enhanced after HF (four studies, odds ratio for loco-regional recurrence or related events 0.35 (0.28-0.45), p < 0.0001). In trials on head and neck tumors and bladder cancer, complete responses were seen more often after HF compared with CF (odds ratio for failure of complete response: 0.43 (0.32-0.57), p < 0.0001, and 0.43 (0.27-0.70), p = 0.0007). CONCLUSIONS This overview demonstrates that the effectiveness of radiotherapy is consistently higher for HF than for CF. The assumption that tumors have a small effective fractionation sensitivity (alpha/beta > 5 Gy) seems to be fulfilled especially for head and neck cancers.