Читать книгу Cobert's Manual Of Drug Safety And Pharmacovigilance (Third Edition) - William Gregory - Страница 25

In practice, these definitions are rather murky and confusing. They also change periodically. For a good summary of post-marketing pharmacovigilance definitions used in the EU (and often elsewhere), see the GVP Annex I Definitions, referenced above. This is a comprehensive document with definitions from the EMA. They are very similar to the FDA definitions.

AEs are unintended “bad things” that occur when taking a drug (or biologic or vaccine, etc.). They may or may not be due to the drug itself (the “active moiety”, or “active pharmacological ingredient” (API)), the formulation, excipients in the product (e.g., the inactive ingredients, fillers), the packaging (e.g., leaching of products from a container into the liquid drug product), a contaminant, manufacturing problems, the underlying disease, or some other unknown cause or causes. Thus, an AE does not imply that the drug (i.e., the active component) necessarily caused the bad thing to occur.

An AR is an AE in which there is “reasonable possibility” of a causal relationship between the drug and the AE. Some interpret this to mean that the relationship cannot be ruled out. This is probably too extreme as it implies that unless causality can be absolutely, positively ruled out, it is “possibly related” or that there is a “reasonable possibility” of causality. FDA in its guidance on IND Safety Reporting of December 2012 (https://www.fda.gov/downloads/drugs/guidancecomplianceregulatoryinformation/guidances/ucm227351.pdf) discussed this at length and indicated that they do not want to see cases reported as expedited IND reports if there is “not enough evidence to suggest that there was a reasonable possibility that the drug caused the AE”. This is done to increase the likelihood that the information sent to FDA will be “interpretable and will meaningfully contribute to the developing safety profile of the investigational drug and improve the overall quality of safety reporting”. The notion of causality is discussed in much greater detail in Chapter 22. Thus, an AE possibly or probably due to the drug is an ADR or AR.

These terms are being replaced in practice by SAR, which emphasizes the suspicion that the drug is a possible cause of the bad thing or is the possible cause of the bad thing. Following logically from this, we now have the phrase suspected, unexpected, serious adverse reaction. The addition of the words “serious” and “unexpected” to the SAR term represents the criteria for submission as expedited reports to government health agencies in many countries of serious reactions from clinical trials. This phrase is very similar in meaning to the acronym SUSAR that is used for expedited clinical trial reports in the EU.

Suspected, expected, serious adverse reactions usually do not have to be submitted as expedited reports to governmental agencies. They are usually submitted periodically (e.g., yearly) or at the end of the study in the final study report. However, some regulatory jurisdictions require submission of all “serious” SARs, regardless of expectedness, if there is a reasonable possibility of a causal relationship with the product.

Expectedness represents an often highly subjective area. An event or reaction is expected if it is found in the product reference document (IB for clinical trials or the post-marketing labeling for approved drugs). More specific or more severe events or reactions, however, are considered to be unexpected. Thus, if “pneumonia” is in the brochure or product labeling and the patient has “streptococcal pneumonia”, this is considered unexpected because the “streptococcal” designation is more specific. Similarly, “fatal pneumonia” is considered unexpected if only pneumonia is labeled (see Chapter 22).

The bottom line here is that there are multiple definitions and variants floating around. They all more or less add up to the same cases being “expeditable” in the United States, European Union, and elsewhere in many, but not all situations. There are nuanced differences in the definitions of related/unrelated, but fundamentally what they come down to is that cases that are serious (death, life-threatening, hospitalization, disability/incapacity, birth defect) and related (“reasonable possibility” that the AE is due to the drug) and unexpected (not in the IB or only included in the class labeling section) are expeditable in the clinical trial setting.

Another nuance is the responsibility for the causality determination for clinical trial reports: FDA has assigned this responsibility to the sponsor (only), whereas elsewhere, either the sponsor or the investigator judgment applies. In general, one should be conservative in applying the definitions, and if one has to discuss or debate whether something is serious and/or related and/or unexpected, then it is. That is, if there is any doubt about any of these three definitions, choose the more conservative approach (serious, related, unexpected).