A multicenter retrospective study to assess feasibility, safety and efficacy of first-line carboplatin-paclitaxel versus carboplatin monotherapy in a frail, elderly epithelial ovarian cancer population

Objective Underrepresentation of elderly ovarian cancer patients in clinical trials has led to lack of clarity regarding optimal first-line chemotherapy in this cohort. The Elderly Women with Ovarian Cancer (EWOC)-1 trial demonstrated that 3-weekly carboplatin (3wC) resulted in worse survival and feasibility compared with standard 3-weekly carboplatin-paclitaxel (3wCP) in frail, elderly ovarian cancer patients. Our retrospective study compares feasibility, safety, and efficacy of first-line 3wCP and 3wC in a frail ovarian cancer cohort. Methods Clinical data were retrospectively analyzed for newly-diagnosed, International Federation of Gynecology and Obstetrics (FIGO) stage III/IV, >= 70-year-old epithelial ovarian cancer patients, treated by clinician choice with 3wC or 3wCP at two London cancer centers over a 2 year period. Charlson Comorbidity Index (CCI) and Eastern Cooperative Oncology Group (ECOG) performance status provided surrogate markers of frailty. Common Terminology Criteria for Adverse Events v5.0 graded toxicity. Results A total of 107 patients were treated with 3wC (n=30) and 3wCP (n=77). Age, performance status, and CCI were significantly different between cohorts, with 3wC patients older (84 vs 75 years, p<0.001), with more comorbidities (median CCI 4 vs 3, p<0.001) and worse performance status (47% vs 17% PS >= 2, p=0.015). Surgical outcomes differed significantly between cohorts, with 20 (67%) 3wC patients not undergoing surgery, compared with 22 (29%) 3wCP patients (p<0.001). Median follow-up was 45.8 months (IQR 38.7-56.3 months). While we observed improved progression-free (HR 0.55, 95% CI 0.33 to 0.90, p=0.017) and overall survival (HR 0.44, 95% CI 0.27 to 0.73, p=0.001, log-rank test) in a univariate cox proportional hazards comparison between 3wCP and 3wC, this was not significant on multivariate analysis. Completion of six planned chemotherapy cycles was achieved by the majority, with similar discontinuation rates between groups (13% 3wC vs 8% 3wCP, p>0.05). Overall grade >= 3 hematological toxicity rates were similar between regimens (33% 3wC vs 44% 3wCP, p=0.37) with grade >= 3 neutropenia (p=0.019) and grade >= 3 thrombocytopenia (p=0.006) more common with 3wCP and 3wC, respectively. No treatment-related deaths occurred. Conclusion Our data demonstrates that standard 3wCP is a well-tolerated, feasible first-line treatment for frail, elderly ovarian cancer patients. Improved survival with 3wCP was not significant when corrected for established clinical prognostic factors.