The Drug Safety Lifecycle · Section 3.3
~8 min read · The Drug Safety Coach — Global PV Career Course
Key points
Full text
Clinical trials are not only efficacy studies. They are the most controlled pharmacovigilance environment that exists — every patient is known, every dose is recorded, every adverse event is systematically collected. The safety data generated in clinical trials forms the foundational safety database that regulators evaluate at the time of marketing authorisation.
Phase I trials, conducted primarily in healthy volunteers (or patients where healthy-volunteer dosing is unethical, as in oncology), establish safety, tolerability, and PK/PD. This is when the SUSAR reporting obligation begins — any Suspected Unexpected Serious Adverse Reaction must be reported within 7 days (fatal/life-threatening) or 15 days (other serious unexpected), per ICH E2A. A Data Safety Monitoring Board (DSMB), operating independently of the sponsor, reviews unblinded interim safety data against pre-specified stopping rules.
Phase II enrols actual patients with the target disease, generating the first substantial dataset of adverse reactions in the intended population — qualitatively different from Phase I’s healthy volunteers, since real patients carry comorbidities and concomitant medications. Adverse reaction pattern recognition, dose-response safety analysis, and risk-factor identification here directly inform Phase III protocol design and the preliminary RMP.
Phase III confirms efficacy but also serves as the primary vehicle for establishing the safety profile presented to regulators — standardised CRFs, MedDRA coding, investigator causality assessment, and subgroup safety analysis (elderly, renally/hepatically impaired, paediatric, pregnant where applicable) feed directly into the Clinical Study Reports, the Summary of Clinical Safety, and the RMP.
Important
A fundamental limitation of Phase III safety data must be understood by every PV professional: pivotal trials are powered to detect efficacy, not rare adverse events. A trial of 3,000 patients has approximately 95% power to detect an adverse event with an incidence of 1/1,000 — but will likely miss events occurring at 1/10,000 or rarer. This is precisely why post-marketing pharmacovigilance is non-negotiable: the safety profile at approval is always incomplete.
Quick check
Test yourself before moving on — no pressure, just click an answer.
1. What is the primary role of a DSMB (Data Safety Monitoring Board)?
2. Why is post-marketing pharmacovigilance described as "non-negotiable" even after a successful Phase III program?