Expectedness Assessment

During product development and throughout the life of the product, a company is constantly observing, listing, validating adverse events (AE) that occur during preclinical, clinical and the post marketing phases. These AEs are listed or mentioned initially in the Investigator’s brochure (IB), the Company Core Data Sheet (CCDS), and later in the Package Insert (PI) or Summary of Product Characteristics (SmPC). These documents are also called Reference Safety Information (RSI).

As per CIOMS V “The concept of expectedness refers to events that may or may not have previously been observed and documented. It does not refer to what might be anticipated (expected in a different sense) from the known pharmacological properties of the medicine. Nor does it refer to what may occur in the course of the treated disease such as in the case of disease progression and/or lack of drug effect”

An AE that does not find mention in the RSI is unexpected

Depending on the stage of the product in the life cycle, the RSI applicable will be different. For a product under investigation, it will be the IB. For a product marketed in Europe, it will be the SmPC, while for USA and other countries it will be the PI. CCDS is another document where all safety information is stored in detail.

Expectedness Synonyms

Listedness – For Events included in CCDS

Labelledness – For Events included in local labels (e.g. SmPC and USPI)

Seems simple, isn’t it? Check RSI and if not found there, AE is unlabelled. Not really. The same AE in different medical scenarios can be interpreted differently. In addition, the prevailing regulatory guidelines in the country have also to be interpreted and applied, making it a challenging task even for experts

An informal survey seeking views on some selected borderline AE cases was conducted at Drug Information Association (DIA) Safety Monitoring Workshops on Pharmacovigilance held in Europe and the United States. A list of 10 borderline examples relating to ‘‘labelledness’’ and six relating to ‘‘seriouseness’’ were identified. In addition to about 100 attendees, 22 physician monitors employed by either Glaxo or SmithKline Beecham Pharmaceuticals completed the exercise

There certainly was a transatlantic difference in the interpretations of the participants. For the outcome death, individuals in the United States would tend to report a fatality as an unlabelled event, whereas in Europe this is generally viewed as an outcome rather than a factor relevant to labelledness. For example, in Europe, 97% of the respondents accept that if myocardial infarction is to be labelled, death is labelled too since it could be the outcome of the event

The 22 medical monitors differed on all examples except one where they all agreed that greater anatomical specificity did not affect the labelledness of lung fibrosis

Guide to Expectedness Assessment

• A sign, symptom or diagnosis that already appears in the list of adverse reactions in an RSI could be reported using a term with a similar meaning. If so, it is not classified as “unexpected”
• If a sign, symptom or diagnosis is different from reactions already included in the RSI with respect to their nature, specificity, mechanism, severity, or outcome, it is not considered as “expected”.
• In the absence of sufficient documentation and in the face of uncertainty, a reaction should be regarded as unexpected.

Examples to Illustrate the Problems and Recommended Solutions

Further anatomical specification:

• If chest pain is expected, left-sided chest pain, which is equivalent to chest pain will also be expected
• If arteritis is expected, temporal arteritis should be considered unexpected due to the associated additional risks and poorer prognosis even though this is also further anatomical specification

Further histological specification does not per se make an expected ADR unexpected [e.g., a liver biopsy shows hepatic necrosis (expected) with the presence of eosinophils (not mentioned in labelling)

Greater diagnostic specification: Cerebral thromboembolism and cerebral vasculitis would both be unexpected (by virtue of greater diagnostic specificity) if the labelling only listed cerebral vascular accidents

Further specification regarding severity:

• if “liver injury” is mentioned in the RSI, fulminant hepatitis will be unexpected owing to the known high incidence of fatal outcome.
• If rash is listed, then Steven Johnson Syndrome is unexpected

Further specification regarding duration:

• If the label refers to acute elevated liver function tests, a raised level lasting three months would be unexpected. When acute cholestatic liver injury is mentioned in the RSI, prolonged cholestatic liver injury is unexpected, since prolonged forms may not be reversible

Do additional signs and symptoms necessarily infer unexpectedness?

• Petechia associated with labelled thrombocytopenia, or dehydration associated with labelled pseudomembranous colitis, are not unexpected.
• If an expected ADR is not usually accompanied by or complicated by a sign, the ADR (i.e., the complication) should not be considered expected. Melena, a complication of labelled gastrointestinal irritation, is unexpected because gastrointestinal irritation per se does not usually cause bleeding. On the other hand, melena would be expected if the label includes “gastrointestinal bleeding.”

How should signs and symptoms of a diagnosis or syndrome be handled?

• If a diagnosis is an expected ADR, then it’s signs and symptoms are also considered to be expected, when they are reported together; e.g., if RSI mentions anaphylactic reaction, combination of hypotension, wheezing, and urticaria together would be an expected event.
• However, the reverse is not true. If RSI mentions only isolated hypotension, wheezing or urticaria then anaphylactic reaction is unexpected.

Events with FATAL outcome

• If RSI specifies an event to be associated with fatal outcome, the event should be considered expected
• If preexisting underlying disease progresses to death (e.g., fatal malignant neoplasm progression), it is usually considered expected or if cardio-respiratory failure is listed fatal cardio-respiratory arrest is considered expected

What is the Role, if any, of “Class Labelling” in RSI ?

• “Class ADRs” should not automatically be expected for the subject /suspect drug unless the drug itself is implicated

Examples

• Drugs of this class are known to cause tremors
• Drugs of this class are known to cause tremors but no reports of tremors have been received till date with this drug
• Drugs of this class including this drug are known to cause tremors

Lack of Expected Clinical Effect and the RSI

• It is an undisputed fact that no drug can be effective in 100% of patients. So, a lack of effect is not unexpected. An anticancer medication may not cure a patient since in chronic, end stage cases there are many factors that come into play including natural progression of disease. But in this example, if we have multiple instances of lack of effect, that should be treated as a signal and handled as such.
• Contrary to the above example if a medication aggravates the effect it is supposed to treat like an asthma or a migraine, that would be unexpected

Overdose

• If an ADR is listed only under Overdose section, it should be considered unlabelled/unexpected if the ADR occurred at normal dose, but the reverse is not true.
• Overdose without any other ADR is usually considered as expected
• In case of overdose with associated ADRs, if all ADRs are labelled, overdose should be marked labelled
• If at least one ADR associated with overdose is unlabelled, overdose itself should be marked unlabelled

Note: Some companies mark overdose as labelled irrespective of listedness/labelling of associated ADRs

Labelling of Medication Errors

• Medication Errors with or without associated ADRs should be considered unlabelled
• Transmission of infectious agent/ contamination should be considered serious and unlabelled

How Should Various Sections of a Core Data Sheet or Other RSI Containing Document Inter-relate with Regard to Safety Information?

Contraindications and precautions section

• In the contraindications or precautions sections it may be mentioned that the subject medication should not be used, or used with caution in presence of certain concurrent medical disorders. This does not imply, that such concurrent conditions are ADRs, unless they are specifically mentioned as such in the adverse reaction section.
• If it is specified that dosage should be reduced in case of renal insufficiency, then renal insufficiency is not an expected ADR unless it is also included in the ADR section
• If it is mentioned that the drug is contraindicated in patients with renal insufficiency and patient is reported to have renal failure, it is still not an expected ADRs since it is not mentioned in the ADR section

ADRs in Drug Interactions section

• If an ADR is reported to have occurred due to interaction between drug X and Y, the ADR should be considered expected if it is mentioned in undesirable effects section or in drug interactions section that ADR is reported to occur if X and Y are given together
• If an ADR is reported to have occurred due to interaction between drug X and Y (metabolized by CYP3A), and in drug interactions section it is mentioned that ADR is observed if X is given along with CYP3A inhibitors, ADR is labelled, although drug Y is not specifically mentioned

One can conclude that in spite of all the possible rules and algorithms set for decision making, medical judgement plays a major role in expectedness assessment

References

1. https://allaboutpharmacovigilance.org/18-i-listedness-expectedness-assessment-introduction/ downloaded May 2020
2. https://allaboutpharmacovigilance.org/18-ii-listedness-expectedness-assessment-reference-safety-information-documents/ downloaded May 2020
3. https://allaboutpharmacovigilance.org/19-1-guidelines-and-examples-of-expectedness-assessment/ downloaded May 2020
4. Current Challenges in Pharmacovigilance:Pragmatic Approaches. Report of CIOMS Working Group V 2001.
5. Management of Safety Information from Clinical Trials. Report of CIOMS Working Group VI 2005

Dr. Vishwas Sovani (June 2021)