OHRP still grappling with how to report AEs to IRBs, despite guidance.

BOSTON The reporting of adverse events and unanticipated problems varies from FDA to OHRP despite efforts at harmonization, and while there is some effort to get as much consistency as possible into the reporting process - notably a federalwide electronic portal for adverse events reporting that will be available in the near future and will streamline submissions - it still remains problematic.

That was the message of a panel speaking at the annual HRPP conference in Boston here. Panelists were Sara Goldkind, M.D., senior bioethicist at FDA; Michael Carome, M.D., associate director of regulatory affairs at OHRP; and Amy Patterson, M.D., director of the Office of Biotechnology Activities at the National Institutes of Health (NIH).

FDA itself has not included a definition of "adverse events" in its regulations (although the term is used in those regulations); but in general, according to Goldkind, it refers to "serious and unexpected adverse drug experiences, new adverse effects, new observations," along with "unanticipated adverse device effects and unanticipated problems."

But it was agreed that OHRP diverges the most in terms of defining what is reported, and while some harmonization of definitions (and thereby procedures) seems to be taking place, the agency's emphasis on "unanticipated problems" as reportable and "adverse effects" as non-reportable would seem to be the major point of disagreement among federal agencies overseeing clinical trials and subject protection.

In March 2004, the Secretary's Advisory Committee on Human Research Protections (SACHRP) passed a resolution recommending that OHRP and FDA promptly issue clear and consistent joint guidance on IRB review of both internal and external adverse event reports, which led to the formation of the Federal Adverse Events Task Force, chaired by NIH and formed from among ex officio members of SACHRP.

A draft guidance was produced, comments solicited, and "after multiple revisions, the current document was published in January of this year, Carome said, referring to the guidance,"Reviewing And Reporting Unanticipated Problems Involving Risks To Subjects And Other And Adverse Events."

But he said it is "very difficult to harmonize regulatory language." In fact, while FDA has significant guidelines for adverse event reporting, there are no OHRP regulations pertaining to adverse events and no provisions for their reporting.

Carome asked the questionin the current OHRP guidance in title 45 of the Code of Federal Regulations, part 46 - public welfare provisions for the protection of human subjects - "what adverse event reporting requirements are stipulated by the HHS regulations for the protection of human subjects?" The answer: none. This is obviously very different from FDA's provisions.

"What has to be reported," Carome said, "are any unanticipated problems involving risks to subjects or others." In addition, according to 45 CFR 46.113, the IRB is given the authority to "suspend or terminate approval of research that has been associated with unexpected serious harms to subjects."

"I think this is important," Carome said. "The only way an IRB can invoke their authority is if they're informed about the problems." The definition of unanticipated problems put forth by the OHRP is extensive and detailed (and available here: www.hhs.gov/ohrp/policy/AdvEvnt Guid.pdf

However, he said, the crux of the matter is the understanding that adverse events are only reportable as unanticipated problems. Any adverse event not falling under the unanticipated problems definition is outside of the scope of reporting.

"In the past, we avoided issuing our own definition of adverse events," Carome noted. "We just used others', taken from other agencies, in the glossary. But we got a lot of negative reaction to that, so a definition was included in the final guidance. We tried to make it similar to others' definitions for purposes of harmonization."

Under the OHRP guidance, then, you must first determine which adverse events are not unanticipated problems (not reportable), which adverse events are also unanticipated problems (reportable), and which unanticipated problems are not adverse events (reportable). "The vast majority of adverse events," Carome said, "are not unanticipated problems." On the other hand, unanticipated problems can include a number of incidents, experiences, and outcomes that are not adverse events.

The guidance further establishes a timeline for the reporting of unanticipated problems: within one week of the investigator becoming aware of a serious unanticipatedproblem event, and within two weeks for non-serious unanticipatedproblem events. All unanticipated events should be reported to appropriate institution officials, the supporting agency head and OHRP within one month of the IRB's receipt of the report of the problem from the investigator.

In addition, Carome said there is a federal-wide portal for adverse events reporting that is currently being worked on. Once it is in place, an electronic submission system will streamline submission of reports from a number of different reporters and go to the NIH, FDA, and other government agencies. Duplication of reporting will be eliminated or greatly reduced. It is the sense of these panelists that this portal will assist in some of the harmonization efforts that have been thus far difficult to achieve.

The guidance for reporting AEs to OHRP can be found here: www.hhs.gov/ohrp/policy/procedures for reporting 052505.pdf

By Jeannette Cezanne New England Correspondent