Accelerated Approval: Good Intentions, Difficult Implementation

October 2010, Vol 1, No 5

A US Food and Drug Administration (FDA) decision originally set for September 17, 2010, on using bevacizumab (Avastin) in breast cancer has been postponed until December 17, 2010. Although bevacizumab is the first antiangiogenic drug for treating cancer and is the bestselling cancer drug in the world, the FDA Oncologic Drugs Advisory Committee (ODAC) recommended that approval for this indication be revoked in an almost unanimous vote (12-1) at a July 20, 2010, meeting.

Accelerated Approval
This particular indication was granted in 2008 under the FDA’s accelerated approval (AA) program. Unlike traditional approval, AA is based on a surrogate end point, an indirect or substitute marker representing a clinically meaningful outcome.

Progression-free survival is often chosen as the primary end point for AA trials, because it measures a direct effect of treatment. Overall survival (OS), ordinarily the “gold standard” of clinical trials, is increasingly difficult to measure for first-line treatment, because treatments administered after a clinical trial has ended cannot be controlled.

The FDA decides whether a particular surrogate end point is acceptable for a clinical trial and generally bases its decision on whether the end point is reasonably likely to predict a real clinical end point.

AA is predicated on the condition that the sponsor will conduct postmarketing trials. If the sponsor fails to conduct these trials, or if they do not confirm clinical benefit, the FDA can revoke the AA and remove the drug from the market.

GAO Finds Fault
AA has not been without its problems and detractors, one of which is the Government Accountability Office (GAO). Last year, it took the FDA to task with a report that concentrated on surrogate end points. The report noted that the FDA granted 90 AAs (79 for cancer) based on surrogate end points from 1992 through the end of 2008 and that 69 of the 204 drugs approved via the traditional process were new molecular entities, many for cancer. Of the 175 postmarketing studies requested, half have been closed as of February 2009.

The GAO report concluded that “weaknesses in FDA’s monitoring and enforcement process hamper its ability to effectively oversee postmarketing studies.” Moreover, the agency does not consider this oversight a priority.

Even more troublesome, the FDA has the authority to remove a drug from the market if a sponsor does not comply with the law, but it has not specified the conditions under which it would do so. “It has never exercised its authority, even when study requirements have gone unfulfilled for nearly 13 years,” the report stated.

The GAO recommended that the FDA commissioner clarify the conditions under which the agency would exercise its authority.

Two Examples of AA Gone Awry
Gemtuzumab (Mylotarg) was granted accelerated approval in May 2000 for the treatment of acute myeloid leukemia. However, postmarketing studies cast doubt on the agent’s effectiveness and a shadow over its safety.

Initial approval was granted on the basis of surrogate end points in 3 clinical trials with a total of 142 patients. A confirmatory trial with 627 patients was begun in 2004 to determine if adding gemtuzumab to standard chemotherapy would improve survival. The trial was stopped early because in addition to not being proved more efficacious, more patients in the gemtuzumab study arm died than in the control arm. Moreover, gemtuzumab was associated with an increasing incidence of venoocclusive liver disease.

A second trial of 1100 patients confirmed the dismal results. As a result, gemtuzumab will be withdrawn as of October 15, 2010.

In the case of bevacizumab, at their July 20 2010 meeting, members of ODAC reviewed 2 postmarketing studies comprising almost 2000 patients and concluded that bevacizumab does not favor benefit over risk. Specifically, there was no additional OS or time to disease progression, and the agent caused significant serious side effects.

Bevacizumab is a highly lucrative source of income for Roche: $6 billion in annual sales, $1 billion of which derives from breast cancer. Even if the FDA withdraws approval, it could still be used off label for that indication. However, Medicare and most private insurers would refuse to reimburse for it, and at a cost of upwards of $50,000 per year, few patients could afford it.

Picture Not All Bleak
Kip Piper, MA, FACHE, President, Health Results Group, Washington, DC, is positive about the AA process. “No regulatory process is perfect, but thousands of lives have been saved because treatments for cancer and other life-threatening conditions received expedited approval. From a pop ulation health perspective, the benefits of AA far outweigh the risks.”

Unfortunately, he added, we live in a risk-aversive, fear-centric country. “Exceptions get amplified, small risks are exaggerated, scientific nuances are obliterated, clinical context is lost in the 24-hour news cycle, and costs are confused with benefits.”

For life-threatening conditions, priority must be on ensuring that patients have maximum access to treatments, but premarket studies will never match the scope of postmarket clinical experience.

Patients may be negatively impacted by an on-again, off-again AA process. “But when lives are at stake, we need to maintain a regulatory system that ensures that perfect is not the enemy of good, or analysis is not the enemy of action,” he said.

Related Articles