2005N-0038 Reporting of Adverse Events to Institutional Review Boards; Public Hearing
FDA Comment Number : EC12
Submitter : Mrs. Margaret DeMeo Date & Time: 03/25/2005 05:03:09
Organization : UConn Health Center
Category : Health Professional
Issue Areas/Comments
(I)Role of IRB - (1)Determination of relatedness is pivotal in determining which AEs are reported. This is without consistent criteria for attribution determination. (1a) Whose opinion counts - the sponsor, external PI, local PI, or the IRB? Is it the role of the IRB to affirm or confirm the attribution determination? When there is conflict of opinion, is the IRB the final arbiter of the scientific debate? (1b) Temporal relationship is under-utilized in deciding relatedness. For example, an SAE report of a 59 year old subject death within 30 minutes of consent and 10 minutes of start of drug infusion. This is determined to be unrelated. An attribution category - temporal relationship only, would be helpful. (2)Unblinding: SAE review is especially difficult when the blind is not broken. Regulatory conditions for breaking a blind in relation to unexpected SAEs are needed. Backbone consistent with ICH Guideline for Industry Clinical Safety Data Management Definition and Standards for Expedited Reporting - Part III D, Managing Blinded Therapy Case), would ease this burden. (2a) Element of Consent: condition for unblinding in relation to an SAE should be a required element. (3) Protocol Adherence, review of unexpected SAEs requires review of the protocol and consent form. Protocol adherence issues may be identified, especially eligibility violations. How should these be reported for external SAEs. (4) Safety Report Content: content of SAE reports vary. Some external SAE reports are consistently lacking important information. At the least, this careful parsing of information is a disservice to subjects and at most, it compromises their safety. (5) Old Safety Reports: earlier trial SAEs, initially considered unrelated, but later determined possibly-related through a pharmacovigilance, are sent to IRBs 4 and 5 years after they occurred. It would be helpful if safety data was reviewed before the next phase of the trial is underway. (6)DSMB - The concept of DSMB is appealing, but their effectiveness and reliability requires ongoing evaluation. (6a) Short term trials, included double-blind, for short duration routinely do not have DSMBs. When multi-center and local site recruit 50 or less subjects, assembling a local DSMB appears impractical. (6b) This IRB suggests that independent DSMBs be required for multi-center trials with greater than low risk. (6c) Sponsor /internal DSMBs, there is potential for conflict of interest. (6d) DSMB reports are rarely substantive, relying on them for annual review may not meet 45 CFR 46.111. Does the IRB need detailed DSMB reports or should they rely that the safety plan is operational as described?
(II) Reportable adverse events - Substantive information of an internal / external SAE should include:(a)Accurate keyword description ? this should enhance not impair communication. Keywords often poorly represent (or misrepresent) the SAE. This stymies IRB review, and the miscommunication is perpetuated later, when the event is documented by keyword in the Investigator Brochure, reducing the utility of the IB. (b) meaningful description of the even; (c) assurance that conduct was consistent with protocol, elig, safety tests, dose reduction, stop rules; (d)DSMB information. (III) Practices of reporting AEs to IRBs - Consistent local policy for AE reporting, and an infrastructure that supports compliance is necessary. This institution has developed: (1)an internal, online AE reporting, review and tracking system. (2) The online system directs AEs (meeting local criteria for expedited review) to each of 3 clinical reviewers. Any reviewer may direct review to the IRB or to an advisory committee that reviews SAEs weekly.

'The rights, safety and well being of trial subjects are the most important considerations, and should prevail over the interests of science and society(ICH - GCP).' This requires an effective AE Reporting System.