By Lisa Seachrist

Washington Editor

BETHESDA, Md. — The FDA took another step toward defining what evidence is necessary to add additional indications to approved cancer therapies by presenting a draft guidance to the Oncologic Drugs Advisory Committee. The specifics of the agency's plan, however, kicked up some controversy among panel members.

The guidance document, which was made available in March, sets forth a series of requirements that would not only allow drug sponsors to claim the drug treats additional cancers, but also bring the labels of a number of chemotherapeutic agents up to date with current clinical practice.

"This document is not a dramatic departure from current practices but more of an attempt to be as transparent as possible in describing what we expect from drug sponsors," said Robert Delap, director of the division of oncology drug products at the FDA. "The document we are presenting today discusses the quality and quantity of data needed to add a new cancer treatment indication to an already marketed product."

More than any other set of drugs, cancer therapies are used to battle a variety of cancers other than those explicitly stated on the labels. Oncologists develop these treatment regimens based upon literature and scientific reports. This off-label use, while legal, concerns the FDA because the product label is meant to provide full prescribing information and should include all clinical indications for which adequate data are available.

In order to encourage companies to submit supplemental applications, the agency developed a guidance document to explain what levels of proof the agency would accept to allow new indications on the label. In the document, the agency describes a number of scenarios where a single trial would be sufficient. They include situations where the drug will be used to treat biologically similar cancers, when the drug will be used to treat cancer at an earlier stage than is indicated, and when the drug will be used as a single agent after receiving marketing approval for use as part of a combined regimen.

In addition, the agency is proposing a single study -- establishing safety and pharmacokinetics that include efficacy observations rather than proof of efficacy -- that would be sufficient to get a drug already approved in adults labeled for pediatric use.

The agency's plan to allow single trials as evidence -- largely an attempt to get around congressional language in the Food and Drug Act that implies the agency must look at two clinical trials in order to approve a drug for a specific indication -- prompted some panel members to question whether the FDA was lowering its standards for already marketed drugs.

"I really think you have to stand by what the agency considers as defining efficacious," said Arlene Forastiere, an associate professor of oncology at Johns Hopkins University. "You can take the safety data, but I don't think activity in another cancer can support an application where a single trial doesn't provide truly compelling evidence."

Forastiere went on to point out that a drug that works on one tumor type doesn't automatically allow one to extrapolate that it will be effective in another tumor type. As such, Forastiere noted that a single trial may not be compelling enough to warrant adding a new indication to the manufacturers label.

Delap pointed out that in comparison to labeling for these drugs in the past, the agency is approving drugs for "finer and finer slices of indications."

"More recently, we have been very precise about what cancers and what stage a particular drug is indicated," he said.

Richard Gelber, professor of biostatistics at the Dana-Farber Cancer Institute, in Boston, said, "It seems to me that if you've got one tumor type where the drug is effective you can accept one study; it affords some latitude to use one well-controlled study."

The panel also discussed the merits of using a single trial to support a supplemental application of a drug used in late-stage cancer for use in earlier stages of that same cancer. Forastiere noted that in this instance a single study "makes a lot of sense."

However, Richard Schilsky, director of the University of Chicago Cancer Research Center, noted that the agency may also have to look at the toxicity of the drug because it's effects are likely to be felt for a much longer time in earlier-stage cancers. The panel had the same toxicity concerns for drugs that are to be labeled for use in children.

A large problem facing the agency in updating cancer therapy labels is that many of the treatments no longer have patent protections so there is little incentive to update the label. The agency intends to conduct surveys of the literature and of practitioners, and to attend scientific meetings were new indications are presented as a means to keep labels up to date.

The agency officially ended its comment period for this draft on May 30; however, it will still consider comments submitted after that date. When the guidance is final, it will be published in the Federal Register. *