During BRG project meetings, we engage in decision-making as well as philosophical discussions that cross a number of epidemiological topics. These often bloom from the most recent regulatory authority interactions. We thought we would share some of the more cohesive thoughts, thus, Epidemiology Topics will be presented a few times in the next six months. The first topic, Population Dechallenge, was based on observations made during drug adverse event projects.
When investigating adverse eventcausation of an identified exposure in an individual, possibly the most obvious concept is that of challenge-dechallenge-rechallenge. In fact, this determination is used on the United States Food and Drug Administration (FDA) Form 3500 that accompanies reports to the Agency of pharmaceutical-related adverse reactions. The idea is as follows:
- Challenge: original exposure to agent A.
- Positive (+) challenge: adverse event B occurs.
- Negative (-) challenge: adverse event B does not occur.
- Dechallenge: If challenge was positive, remove (or reduce the dose) of agent A.
- Positive (+) dechallenge: adverse event B abates.
- Negative (-) dechallenge: adverse event B does not abate.
- Rechallenge: If dechallenge was positive, reapply exposure (or original dose) of agent A.
- Positive (+) rechallenge: adverse event B reoccurs or intensifies.
- Negative (-) rechallenge: adverse event B does not reoccur or does not intensify.
The challenge-dechallenge-rechallenge assessment is a common theme in pharmacovigilance (and in certain cases, device surveillance) and may aid in determination of causation at the individual patient level. Causation is almost always considered more likely when a positive (+) re challenge exists. Review of this information falls under experimental evidence within Bradford Hill’s criteria.1
Asimilar concept can be applied when considering disease association/causation with a drug (or device product)at the population level. However, there is an important conceptual difference between examining exposures at the individual versus population levels. We have just briefly discussed the individual level challenge-dechallenge-rechallenge situation. Exposure and adverse event associations at the population level are focused on greater amounts of information on a wider scale. In fact, many epidemiological studies and report observations are performed in a retrospective manner and may assess hundreds of thousands or millions of people over many years.
The information for these types of studies and assessments may come from a variety of sources (e.g. large databases, spontaneous reports). Another difference in assessing associations and causality using the concept of challenge-dechallenge-rechallenge at the individual and population levels is the applicability of the results. At the individual level, a positive (+) dechallenge is rarely considered proof of causation. It is considered an association. In the instance where a positive (+) rechallenge exists, there is stronger evidence for causation. However, these types of individual assessments are often difficult to generalize to a larger population. Thus, the concepts of challenge, dechallenge, and rechallenge must be defined with respect to a population, not an individual. It is important to note that rechallenge may not ‘exist’ in a population evaluation.
The population challenge-dechallenge-rechallenge concept is as follows:
- Population Challenge: original overall population exposure to agent A.
- Positive (+) population challenge: adverse event B numbers/rates increase.
- Negative (-) population challenge: adverse event B numbers/rates do not increase.
- Population Dechallenge: if population challenge was positive, observe that exposure to agent A was subsequently removed or exposures were reduced in the overall population.
- Positive (+) population dechallenge: adverse event B numbers/rates decrease.
- Negative (-) population dechallenge: adverse event B numbers/rates do not decrease.
- Population Rechallenge: if population dechallenge was positive, observe if there was the reintroduction of exposure to agent A in the overall population.
- Positive (+) population rechallenge: adverse event B numbers/rates increase.
- Negative (-) population rechallenge: adverse event B numbers/rates do not increase.
With the concept of population challenge-dechallenge-rechallenge, exposure is assessed in a population. Therefore, dechallenge is not necessarily removing (or reducing) the exposure from a person who was being exposed. Instead, it is the removal or reduction of availability of the exposure in the population. A population dechallenge does not consider whether an individual was previously exposed or individually experienced a dechallenge. A finding of positive (+) population dechallenge has the potential to be a strong case for causation. As for population rechallenge, examples are rare.
However, there are examples of true positive (+) population dechallenge that are definitive for exposure-adverse event causation.One example of positive (+) population dechallenge occurred during the thalidomide birth defect disaster. Soon after thalidomide was introduced to pregnant women in the 1950s, thousands of birth defect cases, including rarely occurring phocomelia, were observed in babies whose mothers were exposed.2 The subsequent removal of thalidomide from worldwide markets in the 1960s resulted in a substantial reduction in incidence of these birth defects.
Another example of positive (+) population dechallenge occurred in the United States during the pediatric salicylates-Reye’s syndrome saga from the 1970s through the 1990s. Exposure to salicylates during a varicella or influenza-type illness led to many reports of Reye’s syndrome.3 Over 500 cases of Reye’s syndrome after pediatric exposure to aspirin were reported to the Centers for Disease Control (CDC) in 1980. After public notification of the cases, a Surgeon General’s Advisory, and modification of aspirin-containing medication labeling over the next 15 years, cases of Reye’s syndrome in the United States became extremely rare.3 This was due to decreased pediatric exposure to salicylates during varicella or influenza-like illness (population dechallenge).
In the two examples discussed above, the drug-associated adverse events may be considered signature diseases in that the reactions usually only occur after a specific exposure. This most likely allowed for easier identification in the evaluated populations. It should be mentioned that the examples above represent populations of vulnerable individuals; thalidomide with children exposed in utero, and aspirin with virally-infected children. These population dechallenges were easily detectable even though the pool of events were small compared to the overall exposure numbers.The occurrence of an individual patient rechallenge is extremely unlikely or impossible in these two examples because of the exposure situation and injury permanence – in utero exposure with thalidomide and often permanent sequelae or death in cases of Reye’s syndrome. This difficulty with individual rechallenge adds increased value to the consideration of a population dechallenge for causation assessment.
It is important to note that longitudinal observational assessments of exposure in large populations, such as those mentioned above, may be studied with various modifications of designs often referred to as interrupted time series analysis. These studies may help to determine the overall effect an intervention such as policy or public health action has on the population, and are often implemented by regulatory authorities, governmental institutions, and other health assessment groups. A detailed discussion of these methods is not the subject of this commentary.
When discussing epidemiological studies and reports of any kind, population exposures and results are analyzed and discussed. Descriptions of exposure-adverse event relationships are often lengthy, misinterpreted, and the statistics may be difficult to describe to a general audience when exposures in populations change over time. Sometimes it is helpful to take a step back and look at the overall information and results pictorially (graphs and charts). In this way, the ebb and flow of the data and what is really occurring is often easier to see. All this information is critical when determining experimental evidence from a pharma co epidemiological perspective. As Hill stated, “Here the strongest support for the causation hypothesis may be revealed.”1
Because of the importance of observations from studies in situations where a true “experimental” design is not appropriate or possible, and the impact that this information may have on a significant portion of the population, we propose that these types of studies use the terminology “population challenge”, “population dechallenge”, and when it may occur, “population rechallenge” with appropriate descriptions of “positive (+)” and “negative (-)”.
BRG can assist you with your epidemiology-related regulatory affairs and FDA compliance, submission, or response needs. This may include projects such as adverse event or drug and medical devices usage analyses using the FAERS, MAUDE, and various other healthcare databases. Additionally, BRG can help with FDA adverse event reporting. BRG regulatory experts have extensive experience designing and coordinating various FDA-required and non-required studies. These have included experimental (pivotal and non-pivotal clinical trials), observational, data-mining, and other pharmacovigilance and market vigilance assessments. Do not hesitate to contact us with questions or a quote request.
- Hill AB. The Environment and Disease: Association or Causation? Proc R Soc Med
- Vargesson N. Thalidomide-induced teratogenesis: history and mechanisms. Birth Defects Res C Embryo Today 2015;105(2):140-156.
- Belay ED, Bresee JS, Holman RC, Khan AS, Shahriari A, Schonberger LB. Reye’s syndrome in the United States from 1981 through 1997. N Engl J Med 1999;340(18):1377-1382.
BRG is a global scientific & regulatory consulting firm with extensive experience in the strategic development of drug products, biologics, medical devices, combination products and in FDA regulatory affairs. As such, BRG is an integral part of client decision-making processes.
The opinions and statements in this blog are those of the authors and do not necessarily reflect those of BRG. This blog is based on personal experience and reviews of information publicly available or identified in other database searches.