SAFETY LABELING TIDBIT - No. 11 (Mar 16, 2021): Adverse reaction (AR) for labeling in the EU - Part 2. The SmPC guideline does not contain an explanation of the evidentiary standard (threshold) “[at least] reasonable possibility”, but it seems reasonable to use the explanation from the universe of regulatory safety reporting, which reads in the EU: “there are facts (evidence) or arguments to suggest a causal relationship” (see TIDBIT No. 2). The practical difference between its meaning for labeling and reporting lies in who decides if a reasonable possibility exists. For reporting, it is in the EU usually the investigators (looking at individual events with, at the time, limited information about the product). For labeling it is the company (or agency), having - by the time ARs for labeling are selected - hopefully much additional information available for a “best-evidence assessment of all observed adverse events and all facts relevant to the assessment of causality”, as the SmPC guideline calls it. MORE IN LATER TIDBITS. References and additional information in my book. See www.pharmiceutics.com for previous TIDBITS.
SAFETY LABELING TIDBIT - No. 12 (Mar 17, 2021): Before we start comparing the evidentiary standards (causality thresholds) for including a type of event in labeling and call it adverse reaction (AR) in the USA, Canada, EU and (not yet presented) Switzerland, I have to manage expectations: A better understanding of these threshold concepts does not, per se, make it much easier to decide when to label something as an AR. They are too vague and leave a lot of room for judgement and subjectivity. The keys to good practical decision-making lie elsewhere. However, understanding how these threshold concepts compare and how, in the USA, the threshold for the AR section is meant to compare with the evidentiary standard for elevating risk information to Warnings and Precautions may help structure our deliberations. MORE IN LATER TIDBITS. References and additional information in my book. See www.pharmiceutics.com for previous TIDBITS
SAFETY LABELING TIDBIT - No. 13 (Mar 18, 2021): Comparison of evidentiary standards (ES) for including adverse reactions (AR) in labeling - Part 1: Starting with the ES (causality threshold) for including an AR in the Warnings and Precautions (W&P) section of the US PI, because this gives us a relatively solid reference point for comparisons. - - - For being elevated to W&P of the US PI, an AR is supposed to meet both the applicable seriousness/significance criteria (see FDA guidance) and, ES-wise, a so-called “reasonable evidence of a causal association” threshold. This ES has been declared by FDA (in a 2018 amicus brief to the US supreme Court) to be higher than the one for including an AR in the AR section. This means something that crosses the threshold for the AR section does not necessarily also qualify for inclusion in W&P, even if otherwise serious enough. For being elevated, it should also cross the W&P section’s higher causality-threshold. Careful: This principle may not apply to labeling in other markets. MORE ON THIS IN THE NEXT TIDBITs. References and additional information in my book. See www.pharmiceutics.com for previous TIDBITS.
SAFETY LABELING TIDBIT - No. 14 (Mar 19, 2021): Comparison of evidentiary standards (ES) for including adverse reactions (AR) in labeling - Part 2 (read Part 1 first): FDA’s 2018 amicus brief to the US Supreme Court states: “The Warnings and Precautions section must identify “clinically significant adverse reactions” and certain other safety hazards where “reasonable evidence of a causal association” between the drug and such hazards exists. 21 C.F.R. 201.57(c)(6); see 73 Fed. Reg. 49,603, 49,604 (Aug. 22, 2008) (stating that a “‘preponderance’ of evidence” is not required).” The latter means that existence of a causal association does NOT have to be more likely than its non-existence or, in other words, less than 50% perceived likelihood of causal association is sufficient. Get educated about this by your legal counsel. - - - The amicus brief then states that “The causal threshold for including an adverse reaction in [the Adverse Reactions section] is lower than that for the Warnings and Precautions section …” MORE ON THIS IN THE NEXT TIDBITs. IMPORTANT additional information (about other FDA statements) in my book. See www.pharmiceutics.com for previous TIDBITS.
SAFETY LABELING TIDBIT - No. 15 (Mar 21, 2021): Comparison of evidentiary standards (ES) for including adverse reactions (AR) in labeling - Part 3 (read Parts 1 & 2 first): When FDA declares that, for elevating an AR to W&P, the level of confidence in a causal association does not have to be perceived “preponderance of evidence” for causation [i.e., casually expressed, “more speaking against causation than for it” or “being less than 50% sure” can be good enough] and that the threshold for including it in the AR section is even lower, then we can conclude that “reasonable [causal] association” (i.e., the ES for the US PI’s AR section) represents a range of confidence in causal association that, while ranging up to “100% sure”, starts significantly below “preponderance of evidence” (see TIDBIT No. 6 for background). One would, however, expect that a potential AR would be added to the AR section with such low confidence in causation only if it were of particular clinical relevance (see TIDBITS No. 7 & 8). MORE ON THIS IN THE NEXT TIDBITs. IMPORTANT additional information in my book. See www.pharmiceutics.com for previous TIDBITS.
SAFETY LABELING TIDBIT - No. 16 (Mar 22, 2021): Comparison of evidentiary standards (ES) for including adverse reactions (AR) in labeling - Part 4 (read Parts 1, 2 & 3 first): After we understood how low, for an event of particular clinical relevance (!), the causality threshold is for adding it to the US PI’s Adverse Reaction section as an “AR for labeling purposes”, do we know where, in comparison, the threshold is for the EU SmPC’s section 4.8? Is the EU’s “reasonable possibility of a causal relationship” ES for such an event the same, higher or lower? To my knowledge there is no reg, guideline or other agency publication that can provide an answer. However, experience appears to show that, for all practical purposes, the FDA’s and EU agencies’ thresholds for adding high-relevance ARs to labeling are about the same - with unavoidable exceptions that are likely attributable to differences in situation-specific judgement and to subjectivity. - - - By the way, even if we assumed that a systematic difference between agencies existed, would that be reason enough to propose a label change to any agency later than to the others? Additional info in my book. See www.pharmiceutics.com for previous TIDBITS.
SAFETY LABELING TIDBIT - No. 17 (Mar 23, 2021): Evidentiary standards (ES) for including adverse reactions (AR) in labeling - SWITZERLAND: In 2020, the Swiss labeling guideline changed the ES for adding ARs from an unrealistic “a causal connection cannot be clearly ruled out” to “causal relationship … considered to be plausible in the light of a systematic analysis based on all available data (e.g. …)”. Interpreting “plausibility” as “believability in the light of current knowledge, but not necessarily existence of proof”, I propose that the Swiss ES be understood as equivalent to FDA’s “reasonable [causal] association”. Missing is, however, the “second dimension” one can read into the word “reasonable” as used for the US and EU ES: that it does not only mean (1st) “plausible for an educated, rational thinker” but also (2nd) “adequate to the situation” (which would refer to adjusting the level of required confidence in causation to the potential relevance of the ARs). As the EU guidelines, the Swiss guideline does not address clinical relevance as a factor in decision making. Additional info in my book. See www.pharmiceutics.com for previous TIDBITS.
SAFETY LABELING TIDBIT - No. 18 (Mar 24, 2021): Criteria for “identified” and “potential” risks compared with the criteria for adding adverse reactions (AR) to labeling - Part 1: Applicable ICH guidance provides criteria for sorting risks into “identified” or “potential”. These two groups of risks have been coined to help, among other things, structure decision-making on adequate pharmacovigilance activities. They have NOT been coined to help decide what to include in labeling. Over time, the EU has further operationalized these two concepts, potentially adding to the confusion about their suitability for labeling decision-making purposes. The most consequential labeling errors (“under-labeling”) might result from the assumption that, to be label-worthy, a risk has to be an “identified risk”. In the following TIDBITS, we will review how the criteria for categorizing risks into “identified” and “potential” do (not) translate into the criteria for adding a risk to labeling. Additional info in my book. See www.pharmiceutics.com for previous TIDBITS.
SAFETY LABELING TIDBIT - No. 19 (Mar 25, 2021): Criteria for “identified” and “potential” risks compared with the criteria for adding adverse reactions (AR) to labeling - Part 2: ICH provides the following definitions. Identified risk: “An untoward occurrence for which there is adequate evidence of an association with the medicinal product of interest”. Potential risk: “An untoward occurrence for which there is some basis for suspicion of an association with the medicinal product of interest but where this association has not been confirmed”. Both definitions are accompanied by illustrative examples of such risks. - - - Based on the definitions themselves we already get a sense that identified risks will qualify for inclusion in labeling. But the definition of potential risk is simply too “fuzzy” for a direct comparison with the criteria for labeling and requires that we look at the examples for further information. - - - I will first focus on the ICH definitions proper. Then at an extension that was more recently added to the definition of identified risk in the EU (and which I have seen to cause confusion). Additional info in my book. See www.pharmiceutics.com for previous TIDBITS.
SAFETY LABELING TIDBIT - No. 20 (Mar 26, 2021): Criteria for “identified” and “potential” risks compared with the criteria for adding risks to labeling - Part 3 (read Parts 1 & 2 first): First example for identified risk (ICH wording): “An adverse reaction adequately demonstrated in non-clinical studies and confirmed by clinical data”. The corresponding example for potential risk (ICH wording): “Non-clinical safety concerns that have not been observed or resolved in clinical studies”. Think, for example, about potential teratogenic effects. Or about non-clinically well-documented effects that might plausibly also manifest in humans, and for which even corresponding adverse events were seen in clinical trials … but, not unexpectedly, not at a higher rate than with placebo; not unexpected in the light of, e.g., the size of the study population, the likelihood that an event of this type would be reported in a trial of this design, presumed background rate, etc. - - - Conclusion: Some of the non-clinical data-based potential risks will clearly cross the threshold into labeling (but usually not appear in the Adverse Reactions section). More info in my book. See www.pharmiceutics.com for previous TIDBITS.