To determine overall survival (OS) and factors associated with OS after pelvic exenteration for cervical cancer.Methods
Women with cervical cancer who underwent exenteration (n = 517) were identified from the 1998 to 2011 National Cancer Database. Kaplan-Meier and multivariate Cox proportional-hazards survival analyses were performed to test for associations of potential explanatory variables with OS. Analyzed confounders included age, insurance status, income, distance from home to treatment center, stage, exenteration type, surgical margin status, and treatment with adjuvant radiation and/or chemotherapy.Results
Among the entire cohort with clinical follow-up (n = 313), median OS was 24 months. Stage (P = 2.5 × 10−12), lymph node status (P = 1.3 × 10−7), insurance status (P = 1.5 × 10−5), and histologic type (P = 0.04) were significantly associated with OS by the log-rank test. Unadjusted median OS was 24.2 and 61.8 months for women with squamous and adenocarcinoma histologies, respectively. By multivariate Cox regression, age, insurance status, stage, margin status, and adjuvant radiation were associated with OS. Histology was not independently associated with OS on multivariate regression. Among women with node-negative disease, median OS was 73.2 months.Conclusions
Exenteration may be curative for more than half of women with node-negative cervical cancer. Stage, insurance status, lymph node status, and surgical margin are independently associated with differential OS after exenteration.