Late rectal toxicity after image-based high-dose-rate interstitial brachytherapy for postoperative recurrent and/or residual cervical cancers: EQD2 predictors for Grade ≥II toxicity.

Abstract

To investigate the correlation of rectal dose volume metrics with late rectal toxicity after high-dose-rate pelvic interstitial brachytherapy. From October 2009 to November 2012, 50 patients with residual or recurrent cervical cancer were included. Patients received external radiation 50 Gy in 25 fractions over 5 weeks with weekly cisplatin. Rectum and rectal mucosal (RM) contours were delineated retrospectively. RM was defined as the outer surface of the flatus tube inserted at brachytherapy. The dose received by 0.1, 1, 2, 5 cc of rectum, RM, and sigmoid was recorded. Cumulative equivalent dose in 2 Gy (EQD2) for organs at risk was calculated assuming α/β of 3. Univariate analysis was performed to identify predictors of rectal toxicity. At a median follow-up of 34 months (12-51 months), Grade II and III late rectal toxicity was observed in 9 (18%) and 2 (4%) patients, respectively. On univariate analysis, rectal doses were not significant predictors; however, D 0.1-cc RM dose >72 Gy (p = 0.04), D 1-cc RM dose >65 Gy (p = 0.004), D 2-cc RM dose >62.3 Gy (p = 0.004), and D 5-cc RM dose >60 Gy (p = 0.007) correlated with Grade ≥II toxicity. On probit analysis, the estimated dose in EQD2 for a 10% and 20% risk of rectal toxicity was D 2-cc rectum of 55 and 66 Gy, and RM <55 and 63 Gy, respectively. Limiting 2-cc RM and rectal doses within the proposed thresholds can minimize Grade ≥II toxicity for gynecologic high-dose-rate interstitial brachytherapy.