NEW YORK-"Bevacizumab (Avastin) appears to add to the efficacy of cetuximab(Drug information on cetuximab) (Erbitux) and cetuximab/irinotecan (Camptosar) in irinotecan(Drug information on irinotecan)-refractory bevacizumab(Drug information on bevacizumab)naive colorectal cancer patients," when compared with historic controls, according to Leonard B. Saltz, MD, of Memorial Sloan-Kettering Cancer Center in New York. He reported on a phase II trial of the concurrent administration of two monoclonal antibodies- cetuximab and bevacizumab- with or without irinotecan (abstract 3508). "This is the first trial in which these two antibodies were given to patients at the same time," Dr. Saltz noted. Cetuximab targets the epidermal growth factor receptor (EGFR), whereas bevacizumab targets the vascular endothelial growth factor (VEGF). Naive to Both Monoclonals All patients in the phase II trial had irinotecan-refractory disease and were naive to both monoclonal antibodies studied. Patients were randomized to receive one of the two following regimens:

• Cetuximab (400 mg/m² loading dose, then weekly at 250 mg/m²) plus bevacizumab (5 mg/kg every other week), and irinotecan (same dose and schedule as last received prior to the study)
• The same cetuximab and bevacizumab dosing schedules but without irinotecan.

Cetuximab was started on day 1, but for the first week only, bevacizumab was started on day 2 "in order to permit assessment of these agents separately," Dr. Saltz said. The trial was initially planned to accrue 75 patients in each arm, but statistics were recalculated for a lower number because of the widespread commercial availability of bevacizumab and "the scarcity of bevacizumab- naive patients," Dr. Saltz said. Data were reported for 81 patients randomized to the two arms: 41 in the arm with irinotecan and 40 in the arm without irinotecan. "There was a slight imbalance in terms of older patients in the irinotecan arm," Dr. Saltz said. "Otherwise, there were no significant differences between the arms." The median age in the irinotecan group was 64 years (range, 43-86 years), and in the arm without irinotecan it was 56 years (range, 24-80 years). Predictable Toxicities "Toxicities were as would be ex- pected from the individual agents, without clear evidence of synergistic effect," Dr. Saltz said. The primary toxicity related to the monoclonal antibodies was skin rash (see Table 1). "We don't see a significant interaction between the arms as would be expected," Dr. Saltz said. "We see a degree of toxicity that is not outside the parameters we might expect from cetuximab alone," he said. "In terms of irinotecan-related toxicities, as would be expected, we see substantial increases in neutropenia, diarrhea, both grades 2 and 3/4, as well as fatigue, with over 40% of patients having grade 2 or 3 fatigue when given irinotecan, as compared with only 5% having grade 2 fatigue with the antibodies alone" (see Table 2). "One of the concerns was whether the adverse-event profiles would amplify each other," Dr. Saltz noted. "We did see some evidence of those events, which could potentially be considered relatable to either bevacizumab or cetuximab. These are somewhat vague in nature and certainly could also be attributed to the advanced disease of these patients." The adverse effects included upper and lower gastrointestinal bleeding, rectal fistula, and one myocardial infarction (which was ultimately fatal). Usefulness of Regimen Remains Unknown Partial responses occurred in 15 patients (37%; 95% confidence interval [CI]: 22%-53%) in the irinotecan arm and 8 patients (20%; 95% CI: 9%- 36%) in the cetuximab/bevacizumabonly arm. Median time to progression was 7.9 months for patients receiving irinotecan and 5.6 months in those not receiving irinotecan. "The usefulness of bevacizumab with cetuximab in patients with prior bevacizumab exposure remains unknown. Studies to address this question are in development," Dr. Saltz said. "But I would emphasize that at this point, the applicability of these data to practice today is highly questionable, since patients who are bevacizumab- naive getting to this point in their therapy are quite rare," Dr. Saltz added. "Perhaps most importantly, the study served as the safety pilot for the current CALGB/SWOG study, which will be opening momentarily," Dr. Saltz said. The Cancer and Leukemia Group B/Southwestern Oncology Group study gives investigators the choice of using either FOLFOX (fluorouracil, leucovorin, oxaliplatin(Drug information on oxaliplatin) [Eloxatin]) or FOLFIRI (fluorouracil, leucovorin, irinotecan) and then randomizing patients also to receive cetuximab, bevacizumab, or the combination of the two monoclonal antibodies. Inhibitory Strategies The need for additional studies was supported by Neal J. Meropol, MD, of Fox Chase Cancer Center in Philadelphia, who served as a discussant for this ASCO session. "While the patient numbers are small and this should not lead to a change in practice, this pilot study provides sufficient clinical validation to warrant phase III testing of this concept," he said. "The study by Saltz and colleagues lends credence to the notion that approaching a cancer as a system of interrelated and interdependent pathways can yield clinical fruit. It is also clear that colorectal cancers are a heterogeneous group of diseases characterized by various pathway addictions and redundancies," Dr. Meropol continued. "Ultimately, it is hoped that describing the networks of individual tumors will lead to individualized selection of inhibitory strategies." As potential examples of such strategies, Dr. Meropol mentioned "combinations of receptor inhibitors or EGFR inhibitors in combination with inhibitors of downstream pathways."