Phase III trial of gemcitabine plus docetaxel (GD) compared to capecitabine plus docetaxel (CD) with planned crossover to the alternate single agent in metastatic breast cancer (MBC)

1000 Background: GD and CD are efficacious in patients (pts) with MBC. This study compared safety and efficacy of GD and CD induction regimens, where the alternate, single-agent, crossover therapy (GD to C or CD to G) was predetermined. Primary endpoint was time to progressive disease (TTP). Secondary endpoints included toxicities, overall response (ORR), and overall survival (OS). Methods: This multicenter, open-label, phase III study enrolled MBC pts with possible prior anthracycline therapy, adjuvant or neoadjuvant taxane therapy, but no taxane therapy for MBC ≤6 months prior to entry. Enrollment of 442 pts (221 per arm) was planned with 385 progressions required to achieve 80% power for a 2-month observed difference in median TTP between arms. Pts were randomized to: GD: G 1,000mg/m2 Days 1, 8 plus D 75 mg/m2 Day 1, q21 days; or CD: C 1,000 mg/m2 BID, Days 1–14 plus D 75 mg/m2 Day 1, q 21 days. Upon disease progression, pts were given crossover C or G at doses and schedules identical to induction. ORR was assessed by RECIST. Results: Demographics of 472 enrolled pts were balanced between arms; 57% had prior anthracycline. GD caused greater myelosuppression than CD, but without greater febrile neutropenia. Gastrointestinal toxicities, mucositis, and hand-foot syndrome were greater with CD. More pts in the CD arm (n=61, 26.2%) versus the GD arm (n=41, 17.2%) discontinued due to toxicity (p=0.023). ORR, TTP, and OS were not significantly different comparing GD and CD. However, ORR and TTP were significantly greater for the GD to C crossover monotherapy compared to CD to G. Post-hoc analysis of crossover pts showed that the TTP sum from induction through crossover was 6.1 months greater for GD to C. Conclusions: GD and CD had similar efficacy with toxicity profiles consistent with prior clinical experience. Results suggest that the GD to C crossover sequence may provide a clinical benefit over CD to G. [Table: see text] [Table: see text]