A radiobiological analysis of multicenter data for postoperative keloid radiotherapy.

PURPOSE: To identify factors significantly affecting recurrence rates after postoperative external beam radiotherapy (XRT) of keloids, and to delineate any radiation dose response and effects of radiation dose per fraction. METHODS AND MATERIALS: A comprehensive literature review was performed to compile a database of 2,515 resected keloids (36.9% earlobe). Postoperative XRT was 45- to 100-kV X-rays in 27.0% or 120- to 250-kV X-rays in 11.1%, Co-60 in 1.9%, Sr-90 in 4.7%, 1.5- to 9-MeV electrons in 26.5%, and no XRT in 28.8%. In the 1,791 irradiated patients, the median radiation parameters were as follows: total dose, 15 Gy (range, 6-30 Gy); dose per fraction, 5.0 Gy (range, 2-15 Gy); fractions, 3 (range, 1-10); and time, 7 days (range, 0-33 days). RESULTS: Multivariate stepwise logistic regression correlated decreased keloid recurrence with earlobe location (p = 1.98E-10; odds ratio, 0.34), biologically effective dose (p = 1.01E-27), and treatment with electron beam or Co-60 vs. other techniques (p = 0.0014; odds ratio, 0.72). Different radiobiological models calculated values of α/ß = 1.12 to 2.86 (mean, 2.08) and time (repopulation) correction factors for biologically effective dose from 0.98 to 2.13 Gy per day (mean, 1.34) starting 10 days after surgery. Different models (with α/ß = 2.08) predicted that doses needed for 90% and 95% control with 3 fractions of postoperative electron beam were 16.0 to 16.2 Gy and 18.3 to 19.2 Gy, respectively, in less than 10 days for earlobe keloids and 21.5 to 22.2 Gy and 23.4 to 24.8 Gy, respectively, in less than 10 days for other sites. CONCLUSIONS: Postoperative keloid radiotherapy requires moderately high doses and optimal technique to be effective. The relatively low α/ß ratio indicates that radiotherapy with a limited number of fractions and high doses per fraction is the best strategy.