By Kurt R. Karst –
Almost a year after having notified Genentech of a proposal to withdraw the approval of Avastin® (bevacizumab) for use with paclitaxel for the treatment of patients who have not received chemotherapy for metastatic HER2-negative breast cancer, and almost five months after holding a hearing (Docket No. FDA-2010-N-06211) on the topic, FDA, on November 18th, announced FDA Commissioner Margaret Hamburg’s decision to withdraw the approval. FDA approved Avastin® under BLA No. 125085 on February 22, 2008 for the breast cancer use under the Agency’s “accelerated approval” regulations. Avastin® was initially approved on February 26, 2004 as a first-line treatment in combination with intravenous 5-fluorocil-based chemotherapy in patients with metastatic carcinoma of the colon and rectum, and has since been approved for other uses. None of those uses are affected by FDA’s November 18th decision.
In December 1992, FDA promulgated final regulations under which the Agency will accelerate the approval of certain new drugs and biologics for serious or life-threatening illnesses, and when such products provide a meaningful therapeutic benefit to patients over existing treatments. These accelerated approval regulations are located in Subpart H (21 C.F.R. Part 314) of FDA’s drug regulations, and in Subpart E (21 C.F.R. Part 601) of the Agency’s biologics regulations. If a product meets these criteria, then FDA may grant marketing approval based on a demonstrated effect on a surrogate endpoint reasonably likely to predict clinical benefit and a sponsor’s commitment to complete with due diligence the required postmarketing studies to confirm the product’s clinical benefits. A surrogate endpoint is an alternative measurement of the symptoms of a disease or condition that is substituted for measurements of observable clinical symptoms.
Importantly, FDA may withdraw the approval of an application approved under the accelerated approval regulations if, for example, “[a] postmarketing clinical study fails to verify clinical benefit,” “[o]ther evidence demonstrates that [a] product is not shown to be safe or effective under its conditions of use,” or if a sponsor “fails to perform the required postmarketing study with due diligence.” FDC Act § 506 – Fast Track Products – was enacted in 1997 as part of the FDA Modernization Act and is generally considered to have codified FDA’s accelerated approval regulations. Like FDA’s accelerated approval regulations, FDC Act § 506 provides that FDA “may withdraw approval of a fast track product using expedited procedures,” if, for example, “a post-approval study of the fast track product fails to verify clinical benefit of the product,” or “other evidence demonstrates that the fast track product is not safe or effective under the conditions of use.”
FDA’s February 2008 approval of Avastin® was based on a single study – the E2100 study – in which metastatic breast cancer patients showed an improvement of 5.5 months of median Progression Free Survival (“PFS”), the surrogate endpoint, but that did not demonstrate an overall survival benefit or improvement in quality of life. The required postmarketing studies to confirm the benefit of Avastin® for the breast cancer indication – named AVADO and RIBBON1 – had already begun at the time of approval. In both studies, PFS was the primary endpoint. The results of the AVADO and RIBBON1 studies were submitted to FDA in November 2009. FDA determined that the results of the studies did not confirm that the increase in PFS was as substantial as the E2100 study supporting the accelerated approval of Avastin® had suggested – i.e., the studies did not, according to FDA, verify clinical benefit – and the Agency proposed to withdraw the breast cancer indication.
Genentech promptly requested a hearing on the proposed withdrawal. FDA and Genentech disputed the effectiveness information for the Avastin® breast cancer indication and the appropriate risk-benefit analysis to be applied in this case. At the conclusion of the hearing, the advisory committee unanimously voted that the AVADO and RIBBON1 studies failed to verify clinical benefit, that the available evidence on Avastin® demonstrates that the drug has not been shown to be effective for the breast cancer indication, that the available evidence on Avastin® do not demonstrate that it is safe for the breast cancer indication and that Avastin® has not been shown to present a clinical benefit that justifies the risks associated with the drug for the breast cancer indication, and that the FDA Commissioner should not continue the approval of Avastin® for the breast cancer indication while Genentech designs and conducts additional studies intended to verify the product’s clinical benefit if the Commissioner agrees with certain grounds for withdrawal.
FDA Commissioner Hamburg states in her 69-page decision that after having carefully reviewed the record, the conditions for withdrawal have been met, and that “the currently available data do not support continued accelerated approval of this drug for this indication.” The results of the AVADO and RIBBON1 studies, says Commissioner Hamburg, “substantially changed the profile of [Avastin®],” and “have not verified the clinical benefit shown in E2100.
The Avastin® withdrawal is not the first instance in which a product granted accelerated approval has been withdrawn from the market – either because a postmarketing study failed to verify clinical benefit, or because of a sponsor’s failure to complete a required postmarketing study with due diligence – although it is the first instance in which the hearing procedures have been completed and FDA has made a withdrawal decision. As we previously reported, Pfizer voluntarily withdrew (at least insofar as a voluntary withdrawal is truly voluntary when FDA requests it) MYLOTARG (gemtuzumab ozogamicin for Injection) from the market after a required postmarketing study failed to demonstrate clinical benefit. In addition, in August 2010, FDA issued a hearing notice and proposal withdraw approval of all marketing applications for the accelerated approval drug Midodrine HCl, which FDA first approved in September 1996 under the brand name PROAMATINE to treat orthostatic hypotension (see our previous post here). Since that time, a hearing has been requested, but FDA appears to have delayed scheduling a hearing. Instead, FDA opened a public docket (FDA-2010-N-0637) “to provide a forum to facilitate communication regarding the conduct of clinical trials needed to verify and describe the clinical benefit of midodrine hydrochloride.”