By a 5-4 vote, the Oncology Drugs Advisory Committee failed to recommend that FDA approve Avastin (bevacizumab, Genentech) in combination with paclitaxel as a first-line treatment for locally recurrent or metastatic, HER2-negative breast cancer.
GAITHERSBURG, Maryland-By a 5-4 vote, the Oncology Drugs Advisory Committee failed to recommend that FDA approve Avastin (bevacizumab, Genentech) in combination with paclitaxel as a first-line treatment for locally recurrent or metastatic, HER2-negative breast cancer.
The narrow margin represented a major policy disagreement among ODAC members about progression-free survival (PFS) as an endpoint, as well as differing opinions about the quality of the data presented to support the new indication for Avastin.
Genentech based its efficacy argument on a single international phase III, open-label trial, conducted at 258 centers and designated E2100; more than 90% of the 722 patients were treated in the United States. Patients were randomized to Avastin plus paclitaxel or paclitaxel alone.
The study achieved its primary endpoint of PFS, but failed to show a statistically significant increase in overall survival, a secondary endpoint. The sponsor also presented safety data from a second study that compared Avastin plus capecitabine (Xeloda) to capecitabine alone in 462 patients with progressive metastatic breast cancer. That study found no difference in PFS or survival.
Avastin is currently approved for treating first- and second-line metastatic colorectal cancer in combination with 5-FU-based chemotherapy, and as a first-line therapy for unresectable or metastatic nonsquamous non-small-cell lung cancer in combination with carboplatin and paclitaxel. Both indications received approval based on overall survival.
Endpoint advice sought
In June 1999, ODAC recommended that FDA refuse to accept time to progression (TTP) as a primary endpoint for clinical trials in first-line metastatic breast cancer. The latest ODAC meeting became a reprise of that debate focused on a similar endpoint-PFS.
In his opening remarks, Richard Pazdur, MD, director of FDA's Office of Oncologic Drug Products, stressed that the agency sought the committee's advice on the use of PFS as an acceptable endpoint for regular approval, as well as its approval recommendation.
He acknowledged that FDA has accepted TTP and PFS as endpoints in special circumstances. However, he said the use of PFS in a clinical trial requires careful planning, and, unlike overall survival, bias can be easily introduced into the analysis of PFS.
Study data
Data from the E2100 study presented for Genentech by Kathy Miller, MD, of Indiana University School of Medicine, showed an investigator-determined PFS of 11.3 months for Avastin vs 5.8 months for controls in the primary analysis, and comparable PFS times of 9.2 months vs 5.8 months when cases not confirmed by independent radiologists were excluded. Both findings were significant.
E2100 researchers ended the trial after the primary endpoint was reached. This may have accounted for the study's failure to show a significant improvement in survival. Median overall survival was 26.5 months for the Avastin-treated patients and 24.8 months for the control group. Moreover, grade 3-5 adverse events were 20% higher in the Avastin arm.
The endpoint debate
FDA asked ODAC for a vote only on whether it should approve the new indication for Avastin. But it sought and received a long discussion of whether PFS without a proven survival advantage should be considered a measure of direct clinical benefit in the initial treatment of metastatic breast cancer.
Few committee members disputed that PFS offers some value. "There is no question that PFS is clinically meaningful," said Gary H. Lyman, MD, director of Duke's Health Services and Outcomes Research Program-Oncology.
He noted that the most recent analysis of all trials in the adjuvant setting conducted by the Oxford Early Breast Cancer Trials Collaborative Group showed that differences in disease-free survival at 5 years seem to translate into significant overall survival differences at 15 years. "This is a different setting, and I recognize we may not be able to extrapolate these data to other settings," Dr. Lyman said.
Several committee members questioned whether many patients, when first treated for locally recurrent or metastatic breast cancer, should be regarded as untreated for advanced disease in the traditional sense.
"Most of these women have been treated several times, and so very few truly are receiving first-line therapy, making it harder to assess overall survival advantage," said Joanne E. Mortimer, MD, of the Moore Cancer Center at the University of California, San Diego.
ODAC chair Maha Hussain, MD, wondered if FDA could devise an approval route somewhere between overall survival and PFS. "I respect my colleagues' opinions about progression-free survival," said Dr. Hussain, a prostate cancer specialist at the University of Michigan Medical School. "I also come from a field with one or two examples where early positive indications did not translate into a survival advantage." Indeed, treatment with the drugs resulted in worse survival, she noted. "Therefore, if you ignore survival and just go by progression-free survival, you could actually put harmful drugs on the market."
A dissenting outside voice
"Based on the E2100 trial, I believe that Avastin should be in the armamentarium for breast cancer, and that PFS should be considered an endpoint in the first-line setting as well as the second- and third-line settings," Eric Winer, MD, told ONI in an interview. Dr. Winer is director, Breast Oncology Center, Dana-Farber Cancer Institute, and associate professor of medicine, Harvard Medical School.
Dr. Winer was emphatic that PFS is "in and of itself a meaningful endpoint" as long as it is "substantial," is supported by other endpoints, and is balanced with toxicity.
"If your treatment is extraordinarily toxic on a daily basis, then living longer without progression may not be worthwhile," he said. "On the other hand, if your therapy is reasonably well tolerated, then for many people delaying time to progression can be quite beneficial. Of course, there is no question that treatment should be individualized."
Toxicity-one of the FDA's concerns-should always be strongly considered, he acknowledged. Avastin has been associated with some serious adverse events, such as bowel perforation and stroke, though these are very rare. The day-to-day toxicity is modest and manageable, he noted.
"For example, the most common toxicity, hypertension, is easy to manage in the metastatic setting where it remains asymptomatic, although in the adjuvant setting we have to think about it differently. In patients treated in the adjuvant setting, we worry about the long-term implications of hypertension," he said.
Dr. Winer noted that the 5-to-4 ODAC vote was hardly a resounding defeat and a decision that does not necessarily doom Avastin for approval for this indication.
"We will have to see what the FDA is going to do. This was not a clear recommendation," he said. But the committee vote raises issues surrounding clinical trials in general and the overall drug approval process, he pointed out.
"I believe we need to use PFS as an endpoint, but, more importantly, we need new and better drugs. If there is an agent that has a clear benefit, even if it is for PFS [not overall survival], this is a drug I would like to see become available," he said. "I think that setting the approval bar too high sends a very negative message to the pharmaceutical industry. I worry about this message at a time when we want the pharmaceutical industry to come forward with new agents for breast cancer. We should have high standards, but we also have to be reasonable."