Bevacizumab Fails to Improve Survival in Resected High-Risk Melanoma

Treatment with single-agent bevacizumab resulted in an improved disease-free interval in patients with resected melanoma, but no increase in overall survival compared with observation, according to results from the AVAST-M trial (abstract 9501) presented earlier this month at the 2017 American Society of Clinical Oncology (ASCO) Annual Meeting in Chicago.

“AVAST-M is the largest melanoma trial evaluating an anti-angiogenesis strategy for melanoma,” said Philippa G. Corrie, director of chemotherapy services in the cancer division at Addenbrooke’s Hospital in the United Kingdom. “Adjuvant bevacizumab improves long-term disease-free interval, but this did not translate into a significant improvement in distant metastasis-free or overall survival.”

The use of bevacizumab has shown modest activity in the treatment of advanced melanoma. Here, Corrie and colleagues hypothesized that it might be more effective in preventing metastases and tested this hypothesis in patients at high risk for recurrence.

In AVAST-M, 1,343 patients with resected AJCC stage IIb, IIc, or III melanoma were randomly assigned to observation or bevacizumab after resection of their disease. The primary endpoint was overall survival.

Results from a previous interim analysis showed that the disease-free interval was improved in patients assigned to bevacizumab (HR, 0.85; 95% CI, 0.74–0.99; P = .04) compared with observation.

“This has held true over time,” Corrie said.

The 1-year disease-free interval rates were 76% for bevacizumab and 70% for observation. At 5 years, the disease-free interval rate was 51% for bevacizumab and 45% for observation. Distant metastasis-free interval favored bevacizumab but did not reach statistical significance when compared with observation, Corrie noted.

For overall survival, assignment to bevacizumab or observation had identical outcomes (HR, 0.99; 95% CI, 0.84–1.18; P = .94). There were 251 deaths on the bevacizumab arm and 254 deaths on the observation arm. The 5-year overall survival was 64% for bevacizumab and 63% for observation.

The researchers also looked at the data to see if treatment for recurrence had any effect on overall survival findings. At the data cutoff, 49% of patients assigned bevacizumab and 55% of patients assigned to observation had a recurrence. Immune checkpoint inhibitors or other targeted therapies were used to treat recurrences in 16% of the bevacizumab patients and 14% of the observation patients, suggesting that “treatment for recurrence did not influence the overall survival endpoint.”

A retrospective analysis looking at genotype showed that patients with BRAF-mutant melanoma treated with bevacizumab tended to have better overall survival, however the improvement was not statistically significant.

Finally, the researchers assessed circulating tumor DNA (ctDNA) in the plasma of 161 patients whose tumor were either BRAF or NRAS positive. Plasma was collected within 12 weeks of surgical resection. Detection of ctDNA significantly predicted for both worse disease-free interval (HR, 3.12; P < .001) and overall survival (HR, 2.63; P = .003).

According to Corrie, this suggested that ctDNA may be a useful method to stratify patients for future adjuvant therapy trials.