Dose-volume correlation of cumulative vaginal doses and late toxicity after adjuvant external radiation and brachytherapy for cervical cancer.

PURPOSE: To investigate dose-response relationship between vaginal doses and long-term morbidity. METHODS AND MATERIALS: Patients receiving adjuvant pelvic (chemo) radiation and brachytherapy for cervical cancer from January 2011 to December 2014 were included. Baseline vaginal length was determined clinically and from imaging at BT planning. Dose points were defined along mucosa and at 5 mm depth at 12, 3, 6, and 9 'o' clock positions at every 2 cm from apex to introitus. Cumulative equivalent doses in 2 Gy were calculated. Vaginal stenosis was reported in reference to baseline length according to CTCAE version 3.0. Receiver operator characteristics curve was used to identify dose thresholds for univariate and multivariate analysis. RESULTS: Overall, 78 women with median age of 49 (32-71) years were included. The median dose at vaginal apex mucosa and 5 mm depth was 118 Gy3 (78-198) and 81 Gy3 (70-149) respectively. At median follow-up of 36 (18-60) months, vaginal stenosis ≥25%, and grade ≥ II telangiectasia was observed in 33.3% and 45.7%, respectively. On receiver operator characteristics analysis, apical mucosal dose >142 Gy3 and recto-vaginal point dose >86 Gy3 predicted for stenosis on univariate (p = 0.02, p = 0.06) and multivariate analysis (p = 0.04). The probability of stenosis increased from 32% at 70 Gy3, 38% at 80 Gy3, and 45% at 90 Gy3 rectovaginal point dose. No correlation was observed between vaginal doses and telangiectasia and vaginal stenosis and sexual quality of life. CONCLUSION: Vaginal apex mucosal dose >142 Gy3 independently predicts for vaginal stenosis.