SAFETY LABELING TIDBIT - No. 2 (Mar 3, 2021): It is necessary to understand how the concept “reasonable possibility” of causation is interpreted in the universe of regulatory safety reporting as we will see it appear in one of the definitions of AR for labeling purposes. - - - In the early 1990s, ICHE2A operationalized the term AR as follows: “All noxious and unintended responses to a medicinal product … should be considered adverse drug reactions”, followed by “the phrase ‘responses to a medicinal product’ means that a causal relationship between a medicinal product and an adverse event is at least a reasonable possibility, i.e., the relationship cannot be ruled out.” This radical interpretation of reasonable possibility has not been adopted in US FDA’s and the EU’s more recent safety reporting regs. They define reasonable possibility more restrictively as “there is evidence to suggest a causal relationship” (FDA) and “there are facts (evidence) or arguments to suggest a causal relationship” (EU). MORE IN A LATER TIDBIT. References and additional information in my book.

SAFETY LABELING TIDBIT - No. 3 (Mar 4, 2021): There are 4 regulatory authorities that published definitions of the concept “adverse reaction”, specifically operationalized for labeling purposes: US FDA, Health Canada, the EU (and the UK’s MHRA) and Swissmedic. I will discuss them in several upcoming TIDBITS. - - - When analyzing and comparing such definitions we must look for two things. (1) Do they give us an actionable idea of the causality threshold (aka. “evidentiary standard” for accepting a type of event as an AR for labeling; in other words, how confident we should be about causation to accept the event as an AR for labeling)? (2) Does the agency mention that strength of evidence for causation is not the only factor to consider but that the potential clinical relevance of the AR candidate must also be considered in decision making? - - - The latter aspect (relevance) was first prominently mentioned by the CIOMS III/V recommendations for CCSI in the 1990s. They described various relevance aspects as causality-threshold-lowering factors. Obviously, relevance had long been an important consideration from a legal/litigation point of view. MORE IN A LATER TIDBIT. References and additional information in my book

SAFETY LABELING TIDBIT - No. 4 (Mar 5, 2021): FDA’s definition of adverse reaction (AR) for labeling - Part 1. In its labeling regulation, FDA defines AR as follows: “For purposes of prescription drug labeling, an adverse reaction is an undesirable effect, reasonably associated with use of a drug, ... This definition does … include … only those adverse events for which there is some basis to believe there is a causal relationship between the drug and the occurrence of the adverse event.” - - - Note the phrase “for the purposes of prescription drug labeling” (see TIDBIT No. 1 for FDA’s different definition for the purposes of safety reporting). - - - The causality threshold (aka evidentiary standard) is given as “reasonable [causal] association” and then explained as existence of “some basis to believe there is a causal relationship”. Not super helpful, to say the least. We will look at FDA’s 2006 AR section guidance and at an explanation FDA sent in 2018 to the US Supreme Court to better understand this threshold and how it differs from the threshold for elevating a risk to Warnings and Precautions. I must keep these TIDBITs within their character limits, so MORE IN A LATER TIDBIT. References and additional information in my book.

SAFETY LABELING TIDBIT - No. 5 (Mar 8, 2021): FDA’s definition of adverse reaction (AR) for labeling - Part 2. Before we move on to FDA’s Adverse Reaction (AR) section guidance for a better understanding of the causality threshold, a note on how the term “adverse event” is used in regs and guidelines (worldwide). “Event” is NOT the opposite of AR, as we may use it in casual conversations (e.g., “No, nausea is an event, not a reaction”). In labeling regs and guidelines, “event” is sort of a superordinate concept, which includes as sub-concepts both “ARs” and “events that are not ARs”. In the US safety-reporting universe, event includes the sub-concepts “suspected adverse reactions” (in turn including “AR for purposes of reporting” as a sub-concept; see TIDBIT No. 1) and “events that are not SARs”. - - - So don’t be confused by FDA’s clarification of the definition of AR for labeling purposes, which says: “This definition does not include all adverse events …, only those adverse events for which there is some basis to believe there is a causal relationship between the drug and the occurrence of the adverse event.” We not label “mere events” in the US PI AR section (rare exceptions apply). MORE IN A LATER TIDBIT. References and additional information in my book.

SAFETY LABELING TIDBIT - No. 6 (Mar 9, 2021): FDA’s definition of adverse reaction (AR) for labeling - Part 3. The evidentiary standard for accepting a type of event as an adverse reaction (AR) for labeling (“reasonable causal association”) represents a wide range of strengths of association, reaching from perceived proof down to one that is considered comparatively weak. Exactly how weak an association can be and still permit adding the type of event as AR to labeling is not described in FDA’s regs and guidance. However, it seems implied that something for which a causal association would be “un-reasonable” to assume should not qualify to be portrayed in labeling as an AR (we will come back to this point when we discuss the causality threshold in the EU). - - - That an association is, on occasion, permitted to be rather weak is understandable when we realize that decision making takes into account the clinical relevance of an AR candidate. This means: In principle, the more relevant an AR candidate, the sooner (evidence-wise) it is added to labeling. In the next TIDBID, we review how this (which is also one of the pillars of the CIOMS III/V recommendations) is visible in FDA’s AR section guidance. - References and additional information in my book.

SAFETY LABELING TIDBIT - No. 7 (Mar10, 2021): FDA’s definition of adverse reaction (AR) for labeling - Part 4. That the potential clinical relevance of an AR candidate is taken into account shows in two locations of FDA’s AR section guidance. - 1. (Section II) “FDA reviewers and applicants should assess SUCH FACTORS AS SERIOUSNESS, SEVERITY, FREQUENCY, and STRENGTH OF CAUSAL ASSOCIATION in determining which adverse reactions to include in the ADVERSE REACTIONS section …” (3 examples of relevance factors; “frequency” here does not mean “reporting frequency-based evidence”, which would belong into “strength of causal association”, but stands for how likely a user would be affected by the potential reaction, an aspect of relevance). - 2. (Section III B4) “Serious, low-frequency adverse events generally will be listed when THERE IS REASON TO SUSPECT that the drug may have caused the event.” “Non-serious, low-frequency adverse events should be listed only when there is STRONG EVIDENCE that the drug caused the event.” Here, seriousness is mentioned as a factor that lowers the required evidentiary standard in decision-making based on clinical trial adverse event data. MORE IN A LATER TIDBIT. References and additional information in my book.

SAFETY LABELING TIDBIT - No. 8 (Mar 11, 2021): Why is the potential relevance of an adverse reaction (AR) candidate so important in labeling decision making? A: Because of the limitations of available data we can often not be as confident in the existence of a causal association was we wanted to be. However, waiting for more/better data could sometimes delay or prevent notifying users about know-worthy potential risks. Fortunately, regulatory definitions of “AR for labeling” do not require that a causal association be well-established - although it is somewhat disconcerting that labeling usually calls both well-established and more doubtful ARs indiscriminately “adverse reaction”, and that we can usually not convey our doubt in causation. Anyway, being allowed (and invited) to add potential ARs to labeling the sooner (evidence-wise) the more relevant they are, provided a certain minimum level of confidence in causation is reached (see upcoming TIDBITS), helps to prevent “decision paralysis” and ensure we meet user expectations about the timeliness of safety alerts. MORE IN LATER TIDBITS. References and additional information in my book. See www.pharmiceutics.com for previous TIDBITS.

SAFETY LABELING TIDBIT - No. 9 (Mar 12, 2021): Canada’s Product Monograph guidance defines AR for labeling as follows: “’Adverse reaction’, for the purpose of this guidance, is an unintended event, reasonably associated with the use of a drug and conforms to the regulatory definition of ‘adverse drug reaction’.” Conceptually the same as US FDA’s definition. The reference to the regulatory definition is to clarify that, for labeling, the word “response” in that definition (“… reaction means a … response …”) means a reasonably associated effect (i.e., no certainty required). - - - Health Canada is less explicit than FDA in pointing out that an AR’s relevance is taken into account. This is only visible where it states that serious ARs “should be listed even if there are only one or two reported events, unless it is clear that a causal relationship can be excluded” - which is, by the way, a more rigorous rule than FDA’s, which does not extend to ANY serious AR. - - - References and additional information in my book. See www.pharmiceutics.com for previous TIDBITS.

SAFETY LABELING TIDBIT - No. 10 (Mar 15, 2021): Adverse reaction (AR) for labeling in the EU - Part 1. The SmPC guideline states: “4.8 Undesirable effects - This section should include all adverse reactions … for which … a causal relationship between the medicinal product and the adverse event is at least a reasonable possibility, … Adverse events, without at least a suspected causal relationship, should not be listed in the SmPC.” The “at least” means that, as in the US, for labeled ARs the level of confidence in causation may vary from perceived proof down to reasonable possibility. It does not ask that things be added as soon as a reasonable possibility is perceived. How does this compare with US FDA’s approach - theoretically and practically? Is this “reasonable possibility” the same concept we have seen in the EU definition of adverse reaction (aka suspected adverse reaction) for reporting? This and MORE IN LATER TIDBITS. References and additional information in my book. See www.pharmiceutics.com for previous TIDBITS.