Genentech has come out fighting on multiple fronts in its attempt to retain Avastin’s breast cancer indication, proposing a new confirmatory study, submitting new Phase IV data and attacking FDA’s interpretation of both the underlying scientific data and its regulations on withdrawal.

In a 98-page document that seeks a hearing on FDA’s proposal to withdraw Avastin’s accelerated approval in metastatic breast cancer, the company makes a sweeping attack on the agency’s regulatory treatment of bevacizumab while simultaneously offering concessions – in the form of a new study, a Risk Evaluation and Mitigation Strategy and continued suspension of marketing – that it believes justify keeping the indication available for the time being.

Genentech is offering to do a new study, add a REMS and continue to suspend marketing.

Genentech’s multi-pronged push for a public hearing also raises policy issues that go beyond Avastin. Among these are the company’s assertion that FDA has changed the approval standards for first-line treatments of metastatic breast cancer without adequate public discussion, and the resulting implications for cancer drug development.

FDA announced on Dec. 16 that it planned to pursue revocation of approval for first-line, combination therapy with paclitaxel in HER2-negative patients. The indication received accelerated approval in 2008 based on a statistically significant 5.5-month improvement in median progression-free survival in the E2100 study, in which bevacizumab was studied in combination with paclitaxel.

Genentech conducted two confirmatory trials: AVADO, which studied the addition of Avastin to docetaxel, and RIBBON1, which examined Avastin in combination with taxane/anthracyclinc-bascd chemotherapy or capecitabine. Although both showed a significant improvement in PFS, they failed to verify the magnitude of benefit seen in E2100, FDA said, concluding that the drug’s benefits in breast cancer no longer outweighed its risks (Also see "Genentech May Not Get A Chance To Appeal FDA's Decision To Withdraw Avastin's Breast Cancer Approval" - Pink Sheet, 20 December, 2010.).

Genentech has requested a hearing on the withdrawal. However, a hearing is not automatic, and the agency told Genentech it must present specific facts showing a genuine and substantial issue of material fact exists that requires a hearing (Also see "How The Avastin Appeal Will Proceed" - Pink Sheet, 20 December, 2010.).

“Fundamental Differences” Abound

In a data package supporting its hearing request, Genentech asserts there are fundamental differences with FDA on interpretation of the available data, Avastin’s benefit-risk profile (including consideration of data for other approved first-line therapies), the consistency of FDA’s proposed withdrawal with other approval and withdrawal actions, and the broader implications for development of cancer drugs. “In light of these fundamental differences, denial of a hearing would present serious due process concerns both for Genentech and for women with MBC.”

The briefing package purports to contain “substantial new material” that was not considered by FDA’s Oncologic Drugs Advisory Committee at its July 20 meeting on Avastin’s breast cancer indication, and either was not evaluated or only cursorily examined by the agency.

That includes an assessment, “considered in only preliminary form by FDA,” suggesting the choice of chemotherapeutic agent for the combination influences the magnitude of Avastin’s effect.

The cornerstone of Genentech’s request is a push for more time to prove its theory that Avastin’s magnitude of benefit varies depending upon the chemotherapy with which the VEGF inhibitor is paired. “The current lead hypothesis is that chemotherapies that provide prolonged combined exposure with Avastin may yield the strongest treatment effects,” Genentech says.

FDA has already taken a skeptical view of this theory. In his decision memorandum on the withdrawal, FDA Office of Oncology Drug Products Director Richard Pazdur said the assertion of a unique interaction between Avastin and paclitaxel as a means for explaining the magnitude of PFS change seen only in E2100 “has not been substantiated by either clinical or non-clinical evidence.”

Genentech believes that given the existing data, FDA overreached in dismissing this theory out of hand. “Genentech should not be required to have proven that paclitaxel is a preferred chemotherapy partner in order to maintain accelerated approval; rather, Genentech should have the opportunity to conduct a further study while accelerated approval is maintained.”

The company says its theory on the beneficial effect of paclitaxel combination treatment is bolstered by data from smaller, independent studies and a Phase IV trial conducted by parent company Roche. The open-label, 2,264-patient ATHENA study examined Avastin in combination with paclitaxel, docetaxel and other chemotherapies.

Genentech’s data package includes new Phase IV data from an ex-U.S., single-arm study conducted by Roche.

Genentech’s data package includes an efficacy analysis of time-to-progression in a subgroup of 325 patients administered Avastin and weekly paclitaxel. Median TTP was 10.6 months, “approaching the median PFS of 11.3 months observed in the Avastin arm of E2100 and further supporting that the E2100 results do not represent an outlier.” The company notes that it has not previously provided these data to FDA.

The ClinicalTrials.gov listing for ATHENA says the single-arm study is currently recruiting patients. The trial is being conducted entirely outside the U.S.

Another Bite At The Confirmatory Trial Apple

Following ODAC’s July meeting, where it voted 12-1 in favor of removing the breast cancer indication, Genentech submitted a proposal to FDA for a confirmatory, double-blind, randomized, multi-center Phase III study of Avastin with paclitaxel, the company says. The primary efficacy endpoint would be PFS, with overall survival, one-year survival and response rate as secondary endpoints.

The study would include a biomarker component to identify patients who may be more likely to benefit from Avastin – which is a route FDA suggested in announcing its decision. Though at the time the company said it did not have any such data, Genentech is now citing an analysis of AVADO plasma biomarker data that was presented as an abstract at the San Antonio Breast Cancer Symposium in December. The analysis suggested that patients with high levels of VEGF-A may be more likely to derive a substantial benefit than those with low levels.

“Genentech would apply its expertise and resources to conduct a confirmatory study with due diligence to determine the magnitude of benefit that patients can derive from Avastin in combination with paclitaxel and to evaluate further whether there is a valid predictive biomarker for patient benefit,” the company says.

ODAC members and FDA officials have encouraged Genentech to conduct studies of subpopulations and biomarkers to better identify evidence of Avastin’s benefit. In announcing the agency’s withdrawal decision, Pazdur and Center for Drug Evaluation and Research Director Janet Woodcock said FDA was willing to work with Genentech on proposals for such studies (Also see "Avastin Down, But Not Out Yet In Metastatic Breast Cancer Indication" - Pink Sheet, 20 December, 2010.).

The company says FDA has not yet fully assessed its proposal to maintain accelerated approval pending completion of a confirmatory study.

The extent of FDA’s review of Genentech’s proposed study remains an open question, as the proposal was not discussed by the agency in its withdrawal announcement. Given the strength of FDA’s decision memo in favor of withdrawal, it seems unlikely that a proposal for another, multi-year confirmatory study alone – in essence, giving Genentech another bite at the apple while holding onto the breast cancer indication – would be enough to convince FDA to both grant Genentech a hearing and reverse its earlier decision.

Genentech outlines a series of steps it would take while the confirmatory trial was under way to assure appropriate use of Avastin in breast cancer. These include modifying the label to reflect the current breast cancer data and clarifying that the drug is for use only in combination with paclitaxel. The company vows to maintain the cessation on “affirmative marketing” for use in breast cancer that has been in place since August, and it would add a REMS with a medication guide and communication plan.

Rebutting The “Random High” Theory

Genentech also submits what it describes as a new statistical evaluation rebutting FDA’s view that the results from E2100 are questionable because they could reflect a “random high.”

The trial’s PFS evaluation was based on an interim analysis, and the study was stopped early when 65% of the planned events had occurred. Pazdur’s decision memo suggests Avastin’s true effect size may have been overestimated in E2100 because an interim analysis is more variable than a full analysis at study completion.

Genentech counters that the interim results were based on a robust number of PFS events and a long follow-up period, comparable to that of AVADO and RIBBON1; when more PFS events accrued, the results did not change.

New Approval Standard?

Genentech further argues that a hearing is warranted because revoking the metastatic breast cancer indication would amount to setting a new approval standard for the disease that could deter drug development.

“The agency previously – and appropriately – followed a more flexible approach based on data showing a medicine’s effect on PFS or other disease progression endpoints, together with data on [overall survival],” Genentech says. “Now FDA has stated that improvement in PFS is not adequate unless it is of a large (but unspecified) magnitude, (such as the substantial 5.5-month improvement observed in E2100) and that otherwise there must be an improvement in OS to support approval.”

Such a standard will create hurdles for future breast cancer drug development, Genentech warns. No other drug in this first-line setting has shown the magnitude of effect on PFS seen in E2100, and there are substantial obstacles to demonstrating a significant survival benefit. “Existing drugs for first-line MBC currently on the market were not approved on the basis of such a standard, and few drugs for first-line MBC in the future will be able to make such a showing.”

FDA stands accused of failing to clearly specify what magnitude of progression-free survival benefit it would consider clinically meaningful.

FDA also has not clearly specified what magnitude of PFS benefit it would deem clinically meaningful or explained its rationale for that determination, the company says.

On top of a lack of guidance, Genentech also challenges FDA’s view of its obligation to hold a hearing.

The federal regulations governing FDA’s proposal to withdraw the breast cancer indication “do not permit FDA to deny a sponsor a hearing by finding a lack of a material factual dispute,” Genentech says, contradicting statements in the agency’s letter to the company. Even if FDA had the power to deny a hearing in this case, doing so would be inappropriate, the firm argues, because Genentech has demonstrated “ample factual questions suitable for a hearing, and policy concerns militate against denying a hearing.”

By Sue Sutter

Avastin Data, Policy Implications Warrant Hearing On Breast Cancer Use, Genentech Tells FDA

More from Archive

More from Pink Sheet