FDA's public hearing on Avastin's (bevacizumab) metastatic breast cancer claim opened June 28 with an attack by Center for Drug Evaluation and Research officials on Genentech's plan for a new confirmatory study.

Genentech's proposed study of Avastin in combination with paclitaxel in first-line MBC does not justify continued approval of the indication because the trial is unlikely to substantiate the magnitude of progression-free survival benefit seen in the E2100 trial and there are multiple factors that could delay the new study's completion, Division of Biologic Oncology Products Director Patricia Keegan said.

Although Genentech may conduct further studies to establish that Avastin provides a clinical benefit for a subset of patients, it is not appropriate for the VEGF inhibitor to continue to bear a label suggesting it is safe and effective for first-line MBC when the totality of currently available data suggests otherwise, she said.

Keegan was one of five CDER representatives who gave affirmative presentations during the first day of an unprecedented, two-day hearing on the drugs center's proposal to withdraw Avastin's accelerated approval in MBC.

Genentech will make its affirmative presentation to hearing officer Karen Midthun and members of the Oncologic Drugs Advisory Committee on the morning of June 29. The ODAC members will discuss and vote on four questions later in the day (Also see "Cancer Approval Standards, FDA Reputation At Stake With Avastin Hearing, NEJM Authors Say" - Pink Sheet, 27 June, 2011.).

Looking To Replicate E2100's Benefit

CDER's 2008 approval of the MBC claim was based on results from the E2100 trial, in which Avastin in combination with paclitaxel demonstrated a median PFS benefit of 5.5 months compared with paclitaxel alone.

Two confirmatory trials in the first-line setting, AVADO and RIBBON1, studied bevacizumab in combination with other chemotherapeutic agents and failed to show the same magnitude of PFS impact as that seen in E2100, instead demonstrating PFS benefit ranging from less than a month to about three months. No studies have demonstrated an improvement in overall survival or health-related quality of life, CDER said.

In December, CDER announced plans to withdraw the MBC indication, asserting that the biologic's toxicities outweighed the minimal PFS benefit seen in all trials except E2100.

Genentech has suggested that Avastin's effect may be predicated upon the chemotherapy partner with which it is paired as well as the duration of combination therapy. The company is fighting to retain Avastin's MBC indication while it designs and conducts a new confirmatory trial of bevacizumab in combination with paclitaxel.

The proposed study would include a prospective biomarker analysis premised upon data gathered in AVADO, as well as in other settings, that suggest plasma VEGF-A levels may be predictive of bevacizumab's efficacy (Also see "Genentech Banking On Biomarker Data To Save Avastin In Breast Cancer" - Pink Sheet, 6 June, 2011.).

CDER Turns New Phase II Data In Its Favor

However, Keegan maintained that the E2100 trial's PFS results are an outlier and this conclusion is supported by new data from a Phase II study.

Study CIRG-TORI-010 was a three-arm, randomized, placebo- and active-controlled trial in 282 patients with HER-2 negative metastatic breast cancer. The three study arms were: paclitaxel and placebo; paclitaxel in combination with Takeda/Amgen's investigational VEGF inhibitor motesanib; and paclitaxel plus Avastin. The latter arm employed the same dose and dosing schedule as that used in E2100.

Genentech originally cited the study's results, in abstract form, in its January request for a hearing on the MBC indication. Avastin in combination with paclitaxel demonstrated a median PFS of 11.5 months, which was consistent with the median PFS of 11.3 months observed in E2100, the company said.

However, CDER takes a decidedly different view of the data. Based upon its review of results published in The Lancet in April 2011, but not the underlying raw data, the regulator said the trial failed to replicate the PFS benefit seen in E2100.

The Avastin/paclitaxel arm was associated with a 2.5-month improvement in median PFS compared with paclitaxel alone. These results are in line with the more modest benefits seen in AVADO and RIBBON1, Keegan said.

"The results of Study 10 suggest that an additional study using the same dose and schedule of Avastin and paclitaxel in E2100 is unlikely to substantiate the magnitude of the PFS treatment effect seen in the E2100 study," Keegan said.

Genentech's proposed new study would be double-blinded, in comparison with the open label nature of E2100, and would stratify patients by high versus low VEGF-A levels. Despite these differences, however, the proposed treatment arms and patient population are similar to those in the E2100 trial, Keegan said.

Delaying Factors

She also cited several factors that may delay completion of the new confirmatory trial and argue against keeping the breast cancer indication on-label in the interim.

The trial's protocol remains under development, and the study would require a validated biomarker assay for measuring serum VEGF-A levels, which also is under development.

Finally, based upon recently conducted studies, submission of the results would not be expected until at least three years after enrollment begins, she said.

Genentech is expected to defend its study proposal and offer more details about it during the hearing's second day.

-Sue Sutter (s.sutter@elsevier.com)

CDER Attacks Genentech's Plans For New Confirmatory Trial At Avastin Hearing

More from Archive

More from Pink Sheet