NIH Issues New Rules on Oversight and Reporting of Safety Problems in Gene-Therapy Trials

The National Institutes of Health issued new rules on Monday aimed at improving oversight and public disclosure of safety problems in gene-therapy trials. The changes follow criticism that inadequacies in safety oversight contributed to the 1999 death of Jesse Gelsinger, a research volunteer in an experiment at the University of Pennsylvania.

In a key change, the agency will establish a national committee to review the safety of gene-therapy experiments. These studies involve inserting DNA into patients to repair damaged genes, or replace missing ones, that result in illnesses. The new committee will review reports of serious medical complications and deaths from gene-therapy experiments around the nation to determine if they indicate a common pattern of safety problems.

In its announcement in Monday's Federal Register, the agency also requested further public comment on another proposed change that would relax somewhat the NIH's existing guidelines requiring researchers to promptly report safety problems, known as "adverse events." This proposal would make the NIH's reporting requirements similar to those of the Food and Drug Administration -- a change sought by researchers who said it would promote simplicity. The agency has invited public comment until December 19 on that idea.

The new safety board is a response to complaints about the oversight of gene-therapy trials. The criticism erupted after Mr. Gelsinger died at the University of Pennsylvania in 1999, the first death ascribed directly to a gene-therapy treatment itself and not to an underlying disease. The regulations have been winding their way through the NIH and federal bureaucracy since then.

In the months after Mr. Gelsinger's death, the NIH discovered that gene-therapy researchers had widely underreported adverse events. Scientists said that the reporting requirements were vague, and that the agency did not seem to analyze or use the data systematically. Meanwhile, patient advocates argued that the public had no easy way to find out about the adverse-event reports or what they meant.

For the past decade, the NIH's Recombinant DNA Advisory Committee has evaluated proposed gene-therapy studies -- including the one involving Mr. Gelsinger, who suffered from a liver ailment caused by a missing gene -- before they begin. Researchers meanwhile also had to report the same safety information to the FDA. But the FDA's regulations require it to keep the information secret, to protect the business interests of the research sponsors. (In December, the FDA published a proposal to make public such information, but the agency has not yet followed up.)

The NIH's new safety panel -- to be called the NIH Gene Transfer Safety Assessment Board -- will be a subcommittee of the Recombinant DNA Advisory Committee, or RAC. It will analyze safety trends in the approximately 360 gene-therapy studies now under way nationwide, and will provide summaries of its findings to the full committee for discussion. The RAC has no power to stop trials, but, unlike the FDA, these meetings will be held in public. The summaries about safety trends will be posted on an NIH online database that is scheduled to enter pilot testing next summer. The agency will remove identifying information about individual patients.

The changes reflect the NIH's belief that "whenever a patient gets enrolled in a trial, their experience counts," said Amy P. Patterson, director of the NIH's Office of Biotechnology Activities, which monitors gene-therapy studies.

In Monday's announcement, the NIH clarified existing regulations to say that the agency expects research sponsors -- which are usually pharmaceutical companies -- to submit adverse-event reports even if they contain information that a sponsor considers to be confidential, commercial information. Some companies had asked the NIH to keep the reports secret, arguing that publicizing them could allow competitors to replicate their research. Dr. Patterson said the agency would consider requests for exceptions on a case-by-case basis.

The NIH is holding off on another change that would loosen requirements on the timing of adverse-event reports. Now, researchers are required to make the reports "immediately," although there is no specific deadline. Under the proposal, researchers would have to report within 15 days all unexpected, serious adverse events -- such as deaths, life-threatening complications, and other major side effects -- that appear to be related to the experimental gene-therapy treatment. Scientists would have to report all other adverse events within one year. In addition, they would have to report any gene-therapy studies of animals that suggested the therapy posed significant risks for humans.

The agency had proposed similar wording in December; Monday's announcement included some fine-tuning, which triggered another round of public comment on this provision only. Some critics have worried that the proposed change in reporting requirements might lead the NIH's safety-assessment board to overlook problems that, viewed individually, did not seem serious, but that collectively suggested danger.

Dr. Patterson said the change in reporting requirements would allow NIH officials and the safety panel to focus their attention on the most serious potential problems. The NIH has received 3,700 adverse-event reports since 1989, most of them since Mr. Gelsinger's death in 1999, and many scientists have wondered whether the NIH and the Recombinant DNA Advisory Committee have the ability to make sense of all of that information.

Under the proposed change, the NIH estimated it would receive only 120 reports of serious adverse events yearly. "People felt there needed to be a triaging of effort at the local and federal level," Dr. Patterson said. She added that the safety-assessment board will also systematically review all adverse-event reports to look for safety patterns, and for ways to improve the efficacy of experimental treatments.

Still, Alan Milstein, a Philadelphia lawyer who represented Mr. Gelsinger's father in a lawsuit against the University of Pennsylvania, said that the NIH may be relying too much on the willingness of a research team to categorize and report health problems as adverse events. Researchers face conflicts of interest, and reporting problems "should be a decision by someone other than the researcher," he said.