Pharmacovigilance Essentials for Early-Phase Trials

Pharmacovigilance (PV) in early-phase trials operates under a unique combination of pressure and uncertainty. Your safety database is small. Your understanding of the drug's toxicity profile is evolving in real time. The regulatory framework is rigid, with deadlines measured in hours, not days. And the consequences of getting it wrong range from regulatory action to patient harm.

For first-in-human studies, the stakes are even higher. The EU Clinical Trials Regulation (CTR 536/2014), now fully applicable in the UK through the Medicines for Human Use (Clinical Trials) Regulations 2004 (as amended), imposes specific obligations on sponsors that cannot be delegated away. Understanding these obligations — and building systems to meet them reliably — is non-negotiable.

The Regulatory Framework You Must Understand

UK pharmacovigilance for clinical trials is governed by a stack of overlapping requirements. The key instruments:

• The Medicines for Human Use (Clinical Trials) Regulations 2004 (as amended) — transposes the EU CTR into UK law, covering SAE reporting, SUSAR classification, and annual safety reporting
• MHRA GCP Guide — the inspectorate's interpretation of what good PV looks like in practice
• ICH E2A — definitions and standards for expedited reporting
• ICH E6(R2) Section 5.17 — sponsor responsibilities for safety assessment and reporting
• Development Safety Update Report (DSUR) guidelines (ICH E2F) — annual safety reporting to MHRA and ethics committees

Post-Brexit, the UK has retained the core PV framework but sponsors must now report to the MHRA via the CEID (Clinical Trials Regulation Information System, or the MHRA's own portal) rather than EudraVigilance. The timelines and expectations remain broadly aligned with EU requirements.

SAE Reporting: Timelines That Do Not Forgive

Day 0 starts when any member of the sponsor's organisation — including the CRO acting as delegate — becomes aware that a case meets the criteria for expedited reporting. Not when the report is entered into the safety database. Not when the medical monitor reviews it. When someone at the sponsor or CRO first learns about it.

This is the timeline that catches sponsors out. If a site reports an SAE to the CRO project manager on Friday evening, the clock starts then. If it sits in someone's inbox over the weekend, you may already be in breach by Monday morning.

What Counts as a SUSAR

A Suspected Unexpected Serious Adverse Reaction (SUSAR) requires all three elements:

1. Suspected — there is a reasonable possibility the investigational product caused the event
2. Unexpected — the event is not listed in the Investigator's Brochure (IB) at the specificity or severity observed
3. Serious — results in death, is life-threatening, requires hospitalisation, causes significant disability, or is an important medical event

The classification decision requires medical judgement. This is why your PV setup needs qualified medical review — not just administrative processing. A misclassified SUSAR (treated as an expected SAE when it should have been expedited) is a regulatory violation with real consequences.

Building a PV System That Works

The Safety Management Plan

Before the first dose, you need a documented Safety Management Plan that covers:

• SAE collection and verification procedures — who collects, who verifies, who enters into the safety database
• Medical review and causality assessment process — who makes the suspected/expected determination
• Expectedness assessment — reference document (IB version, reference safety information section), who updates it, how IB amendments trigger re-assessment
• Reporting pathways — site → CRO → sponsor → MHRA, with defined handoff points and escalation triggers
• Signal detection methodology — how you review aggregate safety data, how often, and who signs off
• DSUR preparation process — data lock point, authorship, review timeline, submission pathway

The Data Safety Monitoring Board

For Phase I and dose-escalation studies, a DSMB is not just best practice — it is often a regulatory expectation. The DSMB charter should define:

• Membership and independence requirements (no investigators, no sponsor staff with commercial interest)
• Meeting schedule (after each cohort, after each dose level, or calendar-based)
• Stopping rules and decision criteria
• What data the DSMB receives and in what format
• How recommendations are communicated to the sponsor
• Whether the DSMB can recommend protocol modifications beyond stopping (e.g., cohort expansion, dose modification)

In early-phase oncology, DSMBs are increasingly expected to review not just safety data but also preliminary efficacy signals that may inform dose-selection decisions. Build this flexibility into the charter from the start.

The Investigator's Brochure: Your PV Reference Document

The IB is the reference document for expectedness assessment. If it is out of date or incomplete, your PV system will over-report expected events as SUSARs (wasting regulatory resources) or under-report genuinely unexpected findings (creating compliance risk and patient safety risk).

Best practice for IB management in early-phase trials:

• Update after every cohort or significant safety finding — do not wait for the annual revision cycle
• Clearly delineate Reference Safety Information (RSI) — the section used for expectedness assessment should be distinctly formatted and version-controlled
• Include all preclinical toxicology findings — even if they have not been observed clinically yet
• Track IB version against safety database entries — each SAE assessment should reference the IB version used

Signal Detection in Small Data Sets

Traditional pharmacovigilance signal detection — disproportionality analysis, Bayesian methods — relies on large databases. Early-phase trials do not generate enough data for statistical signal detection. Instead, signal detection in Phase I/II is fundamentally clinical judgement supported by systematic review.

Practical approaches:

• Cumulative safety review after every cohort — compare observed adverse events against preclinical data and the IB
• Laboratory trending — not just flagging out-of-range values, but tracking shifts from baseline across the cohort
• Time-to-event analysis — if a particular AE occurs within a consistent time window post-dose, that temporal relationship is a signal regardless of statistical power
• Cross-study comparison — if you have multiple studies with the same compound, review safety data in aggregate

In early-phase trials, every serious adverse event is a potential signal. The question is not whether you have enough data for statistical significance — it is whether the clinical picture suggests a relationship that changes the benefit-risk profile.

PV Outsourcing: What You Can and Cannot Delegate

Many small sponsors outsource safety management to their CRO or a specialist PV vendor. This is efficient, but the sponsor retains regulatory accountability. Key principles:

• You can delegate task execution, not accountability. The MHRA will hold the sponsor responsible for late SUSAR reports, even if the CRO was at fault.
• Define delegation in a specific Safety Letter of Delegation or relevant section of the CRO contract. Generic delegation logs are not sufficient.
• Retain medical oversight. Causality assessment and expectedness determination require medical qualification. If your CRO's PV team does not include a qualified physician, you need a separate medical reviewer.
• Audit the process, not just the deliverables. Review cycle times, not just report quality. A perfectly formatted SUSAR report delivered three days late is a compliance failure.
• Ensure sponsor access to the safety database. You should be able to run your own queries and generate your own reports at any time, not wait for the CRO to send you a monthly summary.

DSUR: The Annual Safety Story

The Development Safety Update Report is the most important safety document you produce each year. It is not a data dump — it is a narrative that tells MHRA and ethics committees what you have learned about your drug's safety profile, what it means, and what you are doing about it.

Common mistakes in early-phase DSURs:

• Listing SAEs without interpretation — the DSUR requires analysis, not just tabulation
• Failing to compare clinical findings against nonclinical data
• Not updating the overall benefit-risk assessment based on the year's findings
• Omitting significant findings from other ongoing or completed studies with the same compound
• Late submission — the 60-day window from DLP is firm

What DEOX Does Differently

We have built pharmacovigilance into our operating model from the ground up, not bolted it on as an afterthought. For every early-phase programme we manage:

• Safety is led by qualified physicians — not administrative staff with a medical sign-off step at the end
• Real-time safety dashboards — sponsors see safety data as it comes in, not in a monthly PDF
• Automated Day 0 tracking — our systems timestamp receipt at every handoff point, so you never lose track of reporting deadlines
• Integrated signal review — safety review is part of our regular project governance, not a separate siloed activity
• DSUR preparation starts on Day 1 — we build the safety narrative incrementally, not in a 60-day scramble

For first-in-human studies, we recommend our sponsors engage an independent DSMB from the outset. We can facilitate the setup — charter development, member identification, meeting logistics — so the safety oversight is genuinely independent and rigorous.

The Bottom Line

Pharmacovigilence in early-phase trials is not a back-office compliance function. It is the system that protects participants, generates the safety evidence regulators require, and provides the data your programme needs to make dose-selection and go/no-go decisions. Getting it right requires medical expertise, operational discipline, and systems designed for speed and accuracy.

The sponsors who struggle with PV are usually the ones who treat it as a reporting obligation rather than a clinical intelligence function. The ones who get it right build safety review into their study governance from the protocol design stage and invest in systems that make compliance the path of least resistance.