Phase III study of standard triweekly versus dose-dense biweekly capecitabine (C) + oxaliplatin (O) + bevacizumab (B) as first-line treatment for metastatic colorectal cancer (mCRC): XELOX-A-DVS (dense versus standard): Interim analysis

Abstract

4078 Background: Every 3 week (Q3W) COB has been shown to be highly active and non-inferior to FOLFOX+B in first-line mCRC. Phase II data suggest that dose-dense every 2 week (Q2W) COB may be significantly more active and better tolerated than Q3W COB. Methods: XELOXA-DVS was a phase III, open-label study of 435 patients with chemonaive mCRC who met standard eligibility criteria. Patients were randomized to Q3W: C 850 mg/m2 BID d1–14 + O 130 mg/m2 d1 + B 7.5 mg/kg d1 or Q2W: C 1500 mg/m2 BID d1–7 + O 85 mg/m2 d1 + B 5 mg/kg d1 for up to 72 weeks. Complete surgical resection was allowed using pre-defined criteria. The primary endpoint, progression-free survival (PFS), was estimated using the Kaplan- Meier method, while the hazard ratio and 95% CI were estimated using Cox regression analysis, based on the intent-to-treat population. No formal statistical testing was conducted. Results: The median PFS was 8.4 months (Q2W) and 9.7 months (Q3W) (hazard ratio [HR]=0.84; 95% CI=0.62–1.13). The median PFS (on-treatment) was 9.1 months and 10.2 months, respectively (HR=0.81). Of the 72 and 73 patients with disease progression (DP), the median time to DP was 9.4 and 10.8 months, respectively (HR=0.86). The objective response rates were 21.7% vs 29.4%, respectively (HR=1.05). Patients in the Q2W vs Q3W group experienced higher rates of grade 3/4 diarrhea (29% vs 24%), hand-foot syndrome (12% vs 8%), and treatment discontinuation rates (40% vs 20%), respectively. Other grade 3/4 toxicities (>5%, Q2W vs Q3W) included fatigue (13% vs 13%), dehydration (12% vs 10%), nausea (8% vs 9%), peripheral neuropathy (5% vs 9%), anorexia (5% vs 7%), and abdominal pain (5% vs 7%). Conclusions: At the dose and schedule used, dose-dense Q2W COB was not superior to standard Q3W COB. These data further confirm the activity and tolerability of Q3W COB. The activity and tolerability of a lower C dose Q2W, with more aggressive dose reduction, combined with B and O or irinotecan is currently being evaluated (X-BIO). [Table: see text]