The Oncologic Drugs Advisory Committee (ODAC) of the FDA has recommended that the Agency revoke approval of bevacizumab (Avastin) in first-line treatment of metastatic breast cancer. The FDA will decide whether or not to accept this recomendation later in the year. E2100, AVADO, and RIBBON-1 clearly showed that the addition of bevacizumab increases response rate and time to progression, although for reasons not yet clear, overall survival (OS) was not prolonged, nor was there an apparent nonsignificant trend for improved survival.

This makes the situation with bevacizumab somewhat different than what we have seen with a number of other drugs submitted for FDA approval. But what is also different is the interesting quirk we see in terms of adverse events and toxicity. They appear formidable on paper, yet the bulk of these events are hypertension and proteinuria, which are not only asymptomatic in most patients but are easily controlled. Testimony to this comes in the form of 800,000 patients exposed to this drug. Still, ODAC emphasized its potential for toxicity.

Unbiased Review?

From my perspective, the FDA reviewer's presentation was disturbingly unfair. While the FDA must fulfill its function of protecting the public, decisions should be based on an unbiased review of the data. Such was not the case here. The reviewer noted, for example, that a hazard ratio for survival of 0.87 was not significant for bevacizumab. However, she suggested that a hazard ratio of 1.003 "favored the placebo" arm. It was a subtle use of language that, in my opinion, introduced bias against the drug.

Similarly, while actual mortality was lower with bevacizumab, the reviewer concluded that drug-related mortality was higher-regardless of the potential for serious toxicities (cardiac complications, bowel perforations) to be attributable to the anthracycline or taxane components. I felt this conclusion was thoughtless, disingenuous, ignorant, or clearly biased, and it suggests the reviewer's agenda was to paint this drug in the most unfavorable light.

Additionally, there is some unfairness in the outcome in that Genentech was asked in 2008 to perform confirmatory trials that showed improvements in PFS without decrements in survival. Most reasonable observers would say Genentech satisfied this requirement. However, ODAC then questioned the clinical significance of findings, without providing a definition of such. A situation in which sponsors and investigators must guess at the FDA's wishes is not transparent or predictable and is therefore unfair.

There is an urgent need for the FDA and the cancer community to agree upon exactly what qualifies a drug for approval. Although there would certainly be differences of opinion from the various corners as to the requisite degree of benefit, by having this conversation we should arrive at what is commensurate with the expectations for advanced breast cancer and deliver something that patients would consider of value. ■

-Gabriel Hortobagyi, MD
M. D. Anderson Cancer Center, Houston

Dr. Hortobagyi attended the ODAC meeting as a consultant to Genentech and made a presentation in support of preserving and expanding the indications for bevacizumab.
Dr. Hortobagyi indicated that "For this meeting [ODAC, July 20, 2010], the company [Genentech] provided my travel and lodging expenses. I have otherwise no financial interest in the outcome of this meeting."