No Benefit With Bevacizumab Maintenance Therapy for Metastatic CRC During Chemo-Free Intervals

The addition of bevacizumab maintenance therapy failed to improve outcomes compared with no treatment when administered during chemotherapy-free intervals (CFI) to patients with metastatic colorectal cancer after induction chemotherapy, according to a randomized trial.

“The duration of first-line combination chemotherapy is a matter of controversy because of the toxicity that occurs with prolonged treatment,” wrote study authors led by Thomas Apericio, MD, PhD, of Saint Louis Hospital in Paris, France. CFIs were introduced to address this, and studies evaluating maintenance therapy during the CFIs have yielded conflicting results.

The new PRODIGE 9 trial was a phase III, open-label, randomized study comparing bevacizumab maintenance during CFI with no treatment (observation). All patients received induction chemotherapy with 12 cycles of fluorouracil, leucovorin, and irinotecan plus bevacizumab. It included a total of 491 patients, divided evenly between maintenance therapy and observation. The results were published online ahead of print on January 18, 2018 in the Journal of Clinical Oncology.

Disease progression or death occurred in 15.9% of the maintenance group and in 18.9% of the observation group. The median total number of chemotherapy cycles was 13 for the patients on maintenance therapy and 16 for the observation group.

There was no difference in the primary outcome of tumor control duration, with a median of 15 months in both arms (hazard ratio [HR], 1.07; 95% CI, 0.85-1.34; P = .57). The median progression-free survival (PFS) was 9.2 months with bevacizumab maintenance and 8.9 months without it (HR, 0.91; 95% CI, 0.76-1.09; P = .316). At 12 months, the PFS rate was 30.2% with maintenance therapy and 21.01% with observation.

The median time to treatment failure was also no different: 11.1 months with bevacizumab and 12.1 months with observation (HR, 1.17; 95% CI, 0.97-1.40; P = .092). Median overall survival (OS) was 21.7 months with maintenance therapy and 22 months without it (HR, 1.07; 95% CI, 0.88-1.29; P = .500). A multivariate analysis found that female gender, performance status ≥ 2, unresected primary tumor, and age older than 65 years were associated with a shorter OS.

Treatment-related adverse events of grade 3 or higher were seen in 80% of bevacizumab patients and in 79% of observation patients. There were more toxicities during the CFI in the bevacizumab group. No difference was seen with regard to quality of life measures.

“The PRODIGE 9 trial demonstrated that bevacizumab monotherapy has no effect when used as maintenance treatment after induction chemotherapy,” the authors concluded. “Additional research is needed to better define subgroups of patients who should receive maintenance chemotherapy after induction treatment or could undergo a true chemotherapy-free interval.”