Randomized phase III study of capecitabine, oxaliplatin, and bevacizumab with or without cetuximab in advanced colorectal cancer (ACC), the CAIRO2 study of the Dutch Colorectal Cancer Group (DCCG).

C. J. Punt
08 june 2008--Background: Cetuximab, a chimaeric monoclonal antibody against the epidermal growth factor receptor (EGFR) has shown efficacy in patients (pts) with ACC. Preclinical and early clinical studies suggest a benefit for combining VEGF and EGFR targeting agents in ACC. The CAIRO2 study was designed to investigate the effect of adding cetuximab to capecitabine and oxaliplatin (CapOx) and bevacizumab in ACC. Methods: Previously untreated pts were randomized between capecitabine 1,000 mg/m² orally b.i.d. day 1-14, oxaliplatin 130 mg/m² i.v. day 1, and bevacizumab 7.5 mg/kg i.v. day 1 (arm A) or the same schedule plus cetuximab 400 mg/m² i.v. in week 1, cycle 1, and 250 mg/m² i.v. weekly thereafter (arm B). All cycles were given q 3 weeks. The primary endpoint was progression-free survival (PFS). Response was assessed q 3 cycles (RECIST). The study was designed to detect an increase in the median PFS of 3 months from 11 to 14 months (21% reduction in the hazard ratio of progression, HR=0.79). Results: A total of 755 patients, WHO PS 0-1 and median age 62 yrs (range 27-83), were randomized between June 2005 and December 2006. Of 736 eligible pts, 730 patients started protocol treatment with a median number of cycles of 10 (range 1-39) in arm A and 9 (range 1-40) in arm B. At the time of this analysis, median follow-up is 14.7 months (range 4-28 months) with 140 patients (19%) still receiving treatment. Reasons for end of treatment did not differ between treatment arms, and 28% of pts in arm A and 30% of pts in arm B discontinued treatment for reasons of toxicity. Grade 3-4 toxicity occurred in 71.8% (arm A) and 81.9% (arm B, p=0.0012). Excluding cetuximab-related skin toxicity resulted in similar rates: 71.8% (arm A) and 74.5% (arm B, p=0.40). Most frequent grade 3-4 toxicities in arm A vs B were diarrhea (19.2 vs 25.8%), fatigue (12.6% vs 15.3%), hypertension (13.7% vs 8.8%), nausea (8.5 vs 6.3%), sensory neuropathy (9.6% vs 7.9%), acneiform rash (0.5% vs 25.2%) and hand-foot skin reaction (18.4% vs 18.1%). A total of 554 progressions or deaths were observed: 264 in arm A and 290 in arm B. Median PFS was 10.7 months (CI: 9.7;12.5) in arm A and 9.8 months (CI: 8.6;10.7) in arm B (p=0.019, HR 1.22 (1.03;1.44). Response rates (PR + CR) were 40.6% (CI: 34.5;46.9) vs 43.9% (CI: 37.8;50.1, p=0.44 ), median OS 20.4 (CI: 18.1;26.1) months vs 20.3 months (CI: 17.9;21.6, p=0.21), and 60-day mortality 1.9% vs 2.4%. Conclusions: The combination of both antibodies, cetuximab and bevacizumab, to CapOx results in a significant decrease in PFS compared to bevacizumab and CapOx. Besides skin toxicity, cetuximab did not add additional toxicity to CapOx and bevacizumab. Updated results will be presented at the meeting, including data on KRAS in relation to outcome of treatment.