PRAC Signal Management

A safety signal becomes a PRAC signal not because a statistical threshold has been crossed, but because information suggesting a new potentially causal association—or a new aspect of a known association—has progressed into the EU regulatory signal-management process and requires scientific assessment. The Pharmacovigilance Risk Assessment Committee (PRAC) is the EMA committee responsible for evaluating and recommending action on pharmacovigilance issues, including safety signals.

Purpose and Regulatory Framework

EU signal management is governed by Regulation (EC) No 726/2004, Directive 2001/83/EC, Commission Implementing Regulation (EU) No 520/2012 and GVP Module IX — Signal management (Rev. 1). EMA also maintains procedural questions and answers on signal management; the current published Q&A is Rev. 5, updated 20 January 2026.

These sources define a networked system. Signal detection and assessment are not performed by PRAC alone. Relevant responsibilities are distributed across:

• marketing-authorisation holders (MAHs);
• national competent authorities in Member States;
• EMA;
• PRAC and its rapporteurs or other designated regulatory assessors; and
• other EU regulatory committees or decision-making routes when a recommendation requires formal implementation.

The practical value of PRAC is that a safety concern affecting medicines authorised through different routes can be considered using a common EU scientific forum and, where necessary, lead to harmonised regulatory action.

A Signal Is a Hypothesis, Not a Confirmed Risk

GVP Module IX defines a signal as information arising from one or multiple sources that suggests a new potentially causal association, or a new aspect of a known association, judged sufficiently likely to justify verificatory action.

Three distinctions follow from that definition.

First, a signal is not proof of causality. The purpose of signal assessment is to determine whether the hypothesis is supported strongly enough to change the understanding or management of the medicine.

Second, a disproportionality statistic is not itself a signal conclusion. Statistical screening can identify product-event combinations that deserve clinical review, but case quality, exposure, confounding, biological plausibility, product information and other evidence determine whether further regulatory evaluation is justified.

Third, not every MAH signal becomes a PRAC agenda item. MAHs operate their own signal-management processes and may close, monitor or further assess hypotheses according to the evidence and applicable regulatory requirements. Regulatory authorities independently perform signal activities and may request information from MAHs when a concern enters EU assessment.

For the broader company process, see [[evdas-and-signal-detection]] and [[eudravigilance-signal-detection]].

Where PRAC Signals Come From

A regulatory signal can originate from many evidence streams, including:

• EudraVigilance ICSRs and case series;
• statistical screening and EVDAS outputs;
• scientific and medical literature;
• clinical trials and post-authorisation studies;
• PSUR/PBRER assessments;
• information from MAHs;
• national competent authorities;
• non-clinical or mechanistic evidence;
• information from regulators outside the EU; and
• other regulatory procedures that reveal a new safety question.

The signal hypothesis may therefore begin with a single unusual case, a pattern of reports, an epidemiological association or a mechanistically plausible concern. The evidentiary route into assessment varies, but the regulatory task is the same: determine what the totality of evidence means for the medicinal product.

EudraVigilance in the Regulatory Signal System

EudraVigilance provides a major shared source of post-authorisation safety data. Regulators can review individual cases, case series, reporting patterns and statistical outputs. The system is especially valuable because it brings together reports across the EEA and enables broader analysis than a single company's database may permit.

EudraVigilance nevertheless has the same fundamental limitations as spontaneous-reporting systems. Reporting is selective, denominators are usually uncertain, stimulated reporting can change case counts, duplicates and incomplete information can distort patterns, and confounding by indication or concomitant medicines may be substantial.

For that reason, PRAC signal assessment is a clinical and scientific synthesis, not a vote on a database statistic.

From Detection to Regulatory Assessment

The terminology used in company signal systems and EU regulatory procedures is related but should not be collapsed into a single universal workflow. At a conceptual level, regulatory assessment moves through:

detection → validation/confirmation that further evaluation is warranted → scientific assessment → recommendation → regulatory implementation and follow-up.

The exact administrative route depends on the medicine's authorisation status, the origin of the signal and the regulatory procedure used.

Validation and confirmation

Early review establishes whether the information contains enough new and potentially causal evidence to justify formal assessment. This involves checking whether the issue is already known, whether the cases or other evidence are sufficiently informative, and whether alternative explanations substantially weaken the hypothesis.

Validation does not require a universal number of cases, a particular reporting odds ratio or QPPV sign-off. Those may be internal procedural controls, but they are not generic EU thresholds for signal validity.

Prioritisation

Signals differ in urgency. Factors such as seriousness, preventability, public-health impact, exposure, vulnerability of the affected population, novelty and strength of evidence can affect prioritisation. An uncommon but potentially fatal preventable event may require faster regulatory attention than a more frequent mild event.

Prioritisation determines how urgently the hypothesis must be assessed; it does not predetermine the eventual conclusion.

How PRAC Assesses a Signal

Once a signal enters PRAC assessment, the scientific question is no longer simply whether an association is possible. The committee must decide what the evidence means for the medicine's safety profile and whether regulatory action is warranted.

A high-quality assessment normally integrates several domains.

Clinical case evidence

Case-level review considers the clinical phenotype, timing, dose, dechallenge or rechallenge where informative, competing causes, co-medication, medical history and consistency between cases. For very rare or distinctive events, a small number of well-documented cases can be highly informative. For common events with high background incidence, spontaneous reports alone may be much less discriminating.

Epidemiology and exposure

Where available, epidemiological data can help estimate incidence, compare risks and evaluate confounding. Exposure information provides context for the number and pattern of observed events, although crude reporting rates should not be interpreted as incidence without an appropriate denominator and ascertainment framework.

Clinical-trial evidence

Randomised and other clinical-study data can provide controlled comparisons, but trials may be too small or too short to evaluate rare or delayed harms. Trial evidence is therefore interpreted together with post-authorisation data.

Mechanism and class evidence

Biological plausibility, pharmacology, toxicology and evidence from related products can strengthen or weaken a causal hypothesis. Class evidence is informative only when the relevant biological or pharmacological features are sufficiently shared.

Existing product information and risk management

PRAC considers whether the suspected reaction is already described, whether the current wording adequately reflects the evidence, and whether existing pharmacovigilance or risk-minimisation measures remain proportionate.

Requests for Information From MAHs

During a regulatory signal assessment, EMA or the competent authority may request additional information from one or more MAHs. The request can seek focused analyses rather than a generic safety summary.

Depending on the signal, useful evidence may include:

• cumulative and interval case searches;
• detailed case narratives and line listings;
• exposure estimates;
• clinical-trial summaries;
• literature reviews;
• epidemiological analyses;
• mechanistic or non-clinical information;
• information on class effects;
• current and proposed product information;
• relevant RMP safety concerns and pharmacovigilance activities; and
• the MAH's integrated medical assessment.

The MAH should answer the regulatory question directly. A large data package without a coherent interpretation can be less useful than a focused submission that explains what the evidence supports, what remains uncertain and which action is scientifically justified.

QPPV oversight of regulatory signal responses

The QPPV should have appropriate visibility of important signal assessments and significant regulatory interactions. This does not mean that the QPPV must personally author every response or sign every signal document. The pharmacovigilance system should ensure that material safety issues, deadlines, conclusions and proposed actions reach the QPPV at a level appropriate to the role.

PRAC Recommendations

After assessment, PRAC may recommend different outcomes depending on the evidence. These can include:

• no change to the product information or risk-management system;
• continued monitoring or further evaluation;
• changes to the SmPC and package leaflet;
• changes to the RMP;
• additional pharmacovigilance activities;
• additional or modified risk-minimisation measures;
• communication to healthcare professionals or patients where appropriate; or
• use of another EU regulatory procedure when the issue requires broader benefit-risk action.

The recommendation should be understood as the scientific conclusion of the signal assessment. The legal route by which it becomes binding product information or another regulatory obligation depends on the applicable procedure and authorisation route.

Published PRAC Signal Recommendations

EMA publishes PRAC recommendations on safety signals. As of 8 September 2026, the most recently published set on EMA's signal-recommendation page is from the 6–9 July 2026 PRAC meeting, published on 3 August 2026. EMA also publishes extracts containing new product-information wording where relevant.

These publications are important regulatory-intelligence sources for MAHs because they show current EU conclusions on particular product-event associations and the wording or actions expected after PRAC assessment.

Monitoring PRAC publications is good operational practice for companies whose products or classes may be affected. The company should then determine whether the recommendation applies to its medicinal product, authorisation and regulatory obligations.

Product Information Changes

A recommendation to update product information does not mean that the MAH should independently edit marketed materials outside the applicable regulatory process. The new wording must be implemented through the appropriate variation or other regulatory route for the product concerned.

The operational chain is therefore:

PRAC scientific recommendation → applicable regulatory procedure → approved product-information change → implementation in controlled company systems and market materials.

This distinction matters in inspections. Evidence of a PRAC recommendation is not the same as evidence that the required regulatory change has been implemented correctly.

Relationship With the RMP

Signal assessment can alter the RMP when the new evidence changes the safety specification or creates a need for different pharmacovigilance or risk-minimisation activities.

Examples include:

• a potential risk becoming an important identified risk;
• a new important potential risk being added;
• an existing concern being removed or refined;
• a new PASS being justified to resolve remaining uncertainty; or
• additional risk-minimisation measures becoming necessary.

An RMP change should follow the current scientific understanding and applicable procedural requirements. It should not be made automatically merely because a signal was discussed by PRAC.

For the broader RMP framework, see [[risk-management-plans]].

Relationship With PSUR Assessment

Some safety issues are identified or further evaluated through PSUR/PBRER assessment. The signal-management and periodic-reporting processes therefore interact. A PSUR may strengthen an emerging concern, demonstrate that a previous signal is no longer persuasive, or reveal a need for regulatory action.

The key principle is consistency: the company's signal records, PSUR conclusions, RMP and product information should not present contradictory scientific positions without an explainable reason.

When a Signal Requires a Broader Regulatory Procedure

A signal can reveal a problem whose implications extend beyond a simple labelling change. If the evidence raises significant questions about the overall benefit-risk balance, conditions of use or need for EU-wide action, a referral or other formal regulatory procedure may be appropriate.

Not every signal becomes a referral. The procedural route depends on the seriousness and scope of the issue, the medicines affected and the legal basis available to regulators.

Practical MAH Implementation

A marketing-authorisation holder needs a process that connects regulatory signal activity with its internal safety system. The objective is not to reproduce EMA's administrative workflow. It is to ensure that important external developments are recognised, assessed and implemented correctly.

Useful controls include:

• monitoring relevant EMA and national authority signal communications;
• identifying which company products and authorisations are affected;
• assigning regulatory requests to appropriate scientific owners;
• maintaining traceability between the request, source data, assessment and response;
• ensuring important conclusions reach the QPPV and relevant governance routes;
• updating signal records, PSUR/PBRER content, the RMP and product information where required; and
• tracking implementation of regulatory outcomes.

No universal EU requirement mandates a particular company signal committee, RACI matrix, signatory chain or review calendar. Those are internal governance choices. The regulatory test is whether the MAH can meet its obligations and demonstrate effective signal management.

Potential Failure Modes

The following are illustrative failure modes, not published inspection findings.

Inspection Considerations

A pharmacovigilance inspector may examine whether the MAH's signal-management system can recognise, assess and act on regulatory signals effectively.

Relevant inspection questions could include:

• How does the organisation know that a PRAC signal recommendation affects one of its products?
• Can a regulatory request be traced to the data and medical assessment used in the response?
• Are signal conclusions consistent with the safety database, literature, PSUR and RMP?
• How are significant regulatory signal issues escalated to the QPPV?
• Can the company show that approved product-information changes were implemented?
• Where further pharmacovigilance was requested, can the organisation show that the activity was initiated and tracked?
• Are external signal developments incorporated into internal regulatory intelligence and signal governance?

The underlying themes are scientific reasoning, traceability, timeliness and implementation rather than possession of a particular checklist or meeting structure.

Practical Review Checklist

This checklist is recommended operational practice, not an EMA template.

1. Is the product-event hypothesis clearly defined?
2. Is the strongest supporting and opposing evidence identified?
3. Is the issue new, or a new aspect of an already known risk?
4. Are case data interpreted in the context of exposure and background occurrence where possible?
5. Are class and mechanistic arguments scientifically transferable to the product?
6. If EMA or a competent authority requested information, does the response answer the specific question?
7. Are important limitations and uncertainties stated explicitly?
8. Does the proposed action follow logically from the evidence?
9. Has QPPV visibility been appropriate to the significance of the issue?
10. Are signal conclusions reflected consistently in the PSUR/PBRER, RMP and product information where relevant?
11. If PRAC recommended a change, has the applicable regulatory implementation route been identified?
12. Is completion of the resulting action traceable?

Key Takeaways

PRAC signal management is the EU regulatory process through which important safety hypotheses are scientifically assessed and, where necessary, translated into regulatory action.

A signal is a hypothesis requiring evaluation, not a confirmed adverse reaction. EudraVigilance statistics, individual cases, clinical trials, epidemiology, literature, mechanism and class evidence all contribute to assessment.

MAHs and regulators have related but distinct responsibilities. PRAC does not replace the MAH's own signal-management system, and an MAH should not wait for PRAC before acting on safety information that requires action under its own obligations.

PRAC recommendations can lead to product-information changes, further pharmacovigilance, RMP changes, risk minimisation or broader regulatory procedures. The company must track the recommendation through the applicable regulatory route to actual implementation.

As of September 2026, GVP Module IX Rev. 1 remains the published signal-management module, while EMA's signal-management Q&A was updated to Rev. 5 on 20 January 2026.

References

1. European Medicines Agency. Guideline on good pharmacovigilance practices (GVP) Module IX — Signal management (Rev. 1). EMA/827661/2011 Rev. 1. https://www.ema.europa.eu/en/documents/scientific-guideline/guideline-good-pharmacovigilance-practices-gvp-module-ix-signal-management-rev-1_en.pdf
2. European Medicines Agency. Questions and answers on signal management. EMA/261758/2013 Rev. 5, updated 20 January 2026. Available from EMA's signal-management pages.
3. European Medicines Agency. PRAC recommendations on safety signals. Current EMA repository of adopted signal recommendations. https://www.ema.europa.eu/en/human-regulatory-overview/post-authorisation/pharmacovigilance-post-authorisation/signal-management/prac-recommendations-safety-signals
4. European Medicines Agency. PRAC recommendations on signals adopted at the 6–9 July 2026 PRAC meeting. EMA/PRAC/155652/2026, published 3 August 2026.
5. European Union. Commission Implementing Regulation (EU) No 520/2012, consolidated version current at 12 February 2026.
6. European Union. Directive 2001/83/EC, as amended.
7. European Union. Regulation (EC) No 726/2004, as amended.

Regulatory Note

This article distinguishes EU legal obligations, GVP guidance, EMA/PRAC procedure and recommended MAH operating practice. As of 8 September 2026, EMA continues to list GVP Module IX Rev. 1 as the current published module and Signal Management Q&A Rev. 5 as the current procedural Q&A. Because the EU pharmacovigilance framework was amended by Commission Implementing Regulation (EU) 2025/1466 and GVP revisions are planned, current EMA procedural material should be checked before relying on this article for a live regulatory response.