Expectedness

The determination of expectedness is strictly relative to the product's Reference Safety Information (RSI). For investigational drugs, this is the Investigator's Brochure (IB); for marketed products, it is the local prescribing information or Summary of Product Characteristics (SmPC). An event is 'expected' only if its nature, severity, specificity, and outcome are consistent with the descriptions in these specific documents. Any event absent from the RSI, or occurring with greater severity or specificity than listed, is classified as 'unexpected'.

Expectedness is not a simple keyword match. It requires a clinical comparison of the observed event against the RSI listing:

• Nature: The medical diagnosis or syndrome (e.g., 'aplastic anemia' vs. a listed term of 'anemia').
• Severity: A Grade 4 event may be unexpected if the RSI only lists Grade 1-2 events.
• Specificity: A specific diagnosis like 'Stevens-Johnson Syndrome' is unexpected if the RSI only lists a general term like 'rash'.
• Outcome: An event resulting in death may be considered unexpected if the RSI only describes non-fatal occurrences.

The expectedness classification is the primary gatekeeper for expedited safety reporting. According to ICH E2A guidelines, only Suspected Unexpected Serious Adverse Reactions (SUSARs) must be reported to regulatory authorities on an accelerated timeline (7 or 15 calendar days).

• Expected + Serious: Does not qualify for expedited reporting; handled in periodic aggregate reports.
• Unexpected + Serious: Qualifies as a SUSAR and triggers immediate unblinding and rapid regulatory submission.

An event cannot be automatically classified as 'expected' based solely on a class effect or its occurrence with other drugs in the same pharmacological class. The event must be explicitly documented in the specific product's own RSI. Similarly, a sponsor's prior knowledge of an event from a different clinical trial does not make it expected for the current protocol unless the IB has been formally updated to include it. The RSI is the sole authoritative source.

In blinded clinical trials, determining expectedness for a serious adverse event often necessitates emergency unblinding of the individual patient's treatment assignment. If the event is unexpected per the IB, the investigator or sponsor must break the blind to confirm whether the patient received the active drug or placebo. Only if the patient was on the active drug is the event a SUSAR requiring expedited reporting. Placebo events are not reportable as SUSARs.

AI-driven pharmacovigilance systems automate expectedness assessment by:

• Semantic Matching: Comparing extracted adverse event terms against the RSI using medical ontologies like MedDRA.
• Severity Classification: Parsing CTCAE grades from unstructured text to compare against the RSI's severity profile.
• Rule Engines: Applying deterministic logic that flags any event not found in the RSI term list as 'unexpected' for immediate human review.
• Contextual Analysis: Detecting negation and uncertainty to avoid misclassifying historical or hypothetical events.