Skip to main content

FDA Perspective on Safety Reporting of Serious Adverse Events

August 2017, Vol 7, No 8: Special Edition – Highlights from ASCO 2017

Chicago, IL—Although regulations state that sponsors are required to notify the US Food and Drug Administration (FDA) of potential serious risks from clinical trials, the large volume of adverse event report submissions has been identified as a burden to the effective conduct of clinical research, reported Meredith K. Chuk, MD, Medical Officer, FDA Office of Hematology and Oncology Products (OHOP), Silver Spring, MD, at the 2017 ASCO annual meeting practice management sessions.

Dr Chuk reviewed the expectations for investigational new drug (IND) safety reporting, as stated in the published regulations, and discussed efforts to streamline the process.

An audit by OHOP determined that from 2006 to 2014, an average of 17,686 safety reports were submitted to OHOP annually. Furthermore, only 14% of initial safety reports from commercial sponsors were informative, said Dr Chuk.1

“The large volume of uninformative reports has become a huge problem. The biggest issue is that valid safety signals are much more difficult to detect. All of the noise and all of the uninformative reporting limits resources for not only the FDA, but also clinical researchers and Institutional Review Boards,” she said.

According to guidance published by the FDA, sponsors are to report any “suspected adverse reaction that is both serious and unexpected” to the FDA and all participating investigators, said Dr Chuk. Adverse events that do not meet all 3 criteria should not be reported as IND safety reports.

Serious Adverse Events

“The criteria of being serious is probably the most straightforward. It’s defined as any event that results in death, life-threatening adverse event, inpatient hospitalization or prolongation of hospitalization, persistent or significant incapacity, congenital anomaly or birth defect, or any other medical events that may jeopardize the patient and require intervention to prevent one of the listed outcomes,” explained Dr Chuk.

For reporting purposes, the sponsor or the investigator can determine whether an adverse event is serious.

Unexpected Adverse Events

Unexpected adverse events are those that are not listed in the investigator brochure or that are not listed at the observed severity or specificity of the adverse event that occurred.

“In order to make this determination, the investigator brochure has to have a list of adverse events for which a causal relationship is suspected or confirmed. This list needs to be updated throughout the development of the drug as the safety profile evolves and as that drug is being used in combination with other products,” she said.

Suspected Adverse Events

The decision of whether an adverse event is suspected to be related to the drug is what gives individuals the most angst and is the most difficult to determine, said Dr Chuk. Nevertheless, this is the most important determination for safety reporting purposes, as well as for developing and understanding the safety profile of the drug, she explained.

Regulations state that there should be “reasonable possibility” that the drug caused the event (ie, evidence to suggest a causal relationship). Another key issue is that the determination of causality for the purposes of FDA reporting is made by the sponsor, not by the investigator.

“We feel that the sponsor, rather than the investigator, is in the best position to make the most comprehensive and consistent assessment of causality since they had information about the adverse event profile of the drug from across the development programs and across clinical trials,” said Dr Chuk.

Single adverse events are usually uninterpretable and should not be reported, although there are exceptions for adverse events that are strongly associated with drug exposure (eg, Stevens-Johnson syndrome, angioedema), she said.

According to OHOP’s safety reporting audit, of the 86% of initial safety reports that were uninformative, 54% of the adverse events were already expected (ie, listed in the drug’s safety information), and the sponsor did not include determination of causality for 50% of the adverse events.

“We determined that 24% of events met all 3 criteria, but a large portion of these were considered anticipated for the patient population under study and should not have been reported as single events,” reported Dr Chuk.

What Can Be Done?

For the past couple of years, the FDA has contacted sponsors individually with examples of uninformative reporting and has attempted to communicate its expectations.

“Sponsors who have invested in complying with the FDA’s 2010 final rule [regarding IND safety reporting] have really seen a drastic reduction in the number of safety reports that they’ve been submitting. The establishment of consistent thresholds for causality assessment and identification of clinically relevant follow-up information are also important,” said Dr Chuk, who emphasized that a companywide commitment is a key requirement for the successful implementation of the FDA’s 2010 final rule.

Finally, based on the success of a safety reporting pilot, the FDA is planning for the digital submission of IND safety reports. Having an electronic format that the FDA can process, review, and archive will lead to a more streamlined and consistent process and improve the ability to detect valid safety signals, while identifying uninformative reports, concluded Dr Chuk.


  1. Jarow JP, Casak S, Chuk M, et al. The majority of expedited investigational new drug safety reports are uninformative. Clin Cancer Res. 2016;22:2111-2113.

Related Items