Adjuvant Vaginal Cuff Brachytherapy and Patterns of Failure for Early Stage Uterine Carcinosarcoma

Abstract

Uterine carcinosarcoma (UCS) is a rare histologic type, for which the optimal adjuvant management has not been fully established. In this study, we evaluated treatment outcomes and patterns of failure in patients with FIGO stage I-II UCS. After IRB approval, our prospectively-maintained database for women with endometrial carcinoma was queried. Seventy-four women with early stage UCS who underwent surgical staging between August 1992 and January 2018 at our institution were identified. After surgery, patients were managed with observation, chemotherapy (CT) alone, radiotherapy (RT) alone, or both CT and RT (CRT). Wilcoxon Rank test and Chi-Squared Test of Independence were used to assess the effect of treatment modality on various clinical parameters. The Kaplan-Meier product-limit method was used to calculate the rates of relapse-free survival (RFS), disease-specific survival (DSS), and overall survival (OS). Univariate and multivariate analyses were employed to identify significant predictors for the survival rates. Median follow-up duration was 51 months (range, 5.1-291.6 months). The median age was 67.0 years. Distribution of FIGO stages was as follows: 48 women (65%) with FIGO stage IA, 17 (23%) IB, and 9 (12%) II. After surgery, 8 (11%) patients received RT alone, 19 (26%) had CT alone, 28 (38%) were treated with CRT, and 19 (26%) were clinically observed and did not receive any adjuvant therapy. All women who had RT received it in the form of high-dose rate (HDR) vaginal cuff brachytherapy alone (median surface dose of 30 Gy in 5 fractions) using a single channel vaginal cylinder. For patients who received chemotherapy, a median of 6 cycles (range, 3-6 cycles) of carboplatin and paclitaxel were administered. As of the most recent follow-up, 25 (34%) patients experienced recurrence of disease: 3 in the vagina only, 1 in the pelvic region only, 4 in both vagina and pelvic region, 1 in paraaortic region, and 16 in distant sites. Wilcoxon Rank test and Chi-Squared Test of Independence showed that neither the rate nor location of recurrence could be predicted by the modality of adjuvant therapy (p>0.05). Five-year OS was 66%, 38%, 81%, and 35%, respectively for CT alone, RT alone, CRT, and observation, respectively (p=0.008). Five-year RFS was 67%, 73%, 76%, and 44%, respectively for CT alone, RT alone, CRT, and observation, respectively (p=0.16). Five-year DSS was 76%, 71%, 81%, and 45%, respectively for CT alone, RT alone, CRT, and observation, respectively (p=0.07). Our results suggest that adjuvant vaginal cuff brachytherapy combined with chemotherapy for women with early stage uterine carcinosarcoma is likely to produce the best survival outcome, but a larger study is needed to further confirm the benefits.