Single-Agent Trabectedin Versus Physician's Choice Chemotherapy in Patients With Recurrent Ovarian Cancer With BRCA-Mutated and/or BRCAness Phenotype: A Randomized Phase III Trial

PURPOSE Literature evidence suggests that trabectedin monotherapy is effective in patients with recurrent ovarian cancer (OC) presenting BRCA mutation and/or BRCAness phenotype. METHODS A prospective, open-label, randomized phase III MITO-23 trial evaluated the activity and safety of trabectedin 1.3 mg/m2 given once every 3 weeks (arm A) in BRCA 1/2 mutation carriers or patients with BRCAness phenotype (ie, patients who responded to >= two previous platinum-based treatments) with recurrent OC, primary peritoneal carcinoma, or fallopian tube cancer in comparison with physician's choice chemotherapy in the control arm (arm B; pegylated liposomal doxorubicin, topotecan, gemcitabine, once-weekly paclitaxel, or carboplatin). The primary end point was overall survival (OS) evaluated in the intention-to-treat population. RESULTS Overall, 244 patients from 21 MITO centers were randomly assigned (arm A = 122/arm B = 122). More than 70% of patients received >= three previous chemotherapy lines and 35.7% had received a poly (ADP-ribose) polymerase inhibitor (PARPi) before enrollment. Median OS was not significantly different between the arms: arm A: 15.8 versus arm B: 17.9 months (P = .304). Median progression-free survival was 4.9 months in arm A versus 4.4 months in arm B (P = .897). Among 208 patients evaluable for efficacy, the objective response rate was 17.1% in arm A and 21.4% in arm B, with comparable median duration of response (5.62 v 5.66 months, respectively). No superior effect was observed for trabectedin in the prespecified subgroup analyses according to BRCA mutational status, chemotherapy type, and pretreatment with a PARPi and/or platinum-free interval. Trabectedin showed a higher frequency of grade >= 3 adverse events (AEs), serious AEs, and serious adverse drug reactions compared with control chemotherapy. CONCLUSION Trabectedin did not improve median OS and showed a worse safety profile in comparison with physician's choice control chemotherapy.