GVP Module VI: Medical Literature Monitoring and Literature ICSRs

Introduction

Medical literature is an important source of information on suspected adverse reactions. It can contain individual case reports, case series, study-derived cases and other safety information that may need to enter the pharmacovigilance system.

For an MAH, literature monitoring is therefore not simply a bibliographic exercise. It is a controlled pharmacovigilance process involving search strategy design, retrieval, screening, medical assessment, ICSR validation, follow-up, duplicate management, reporting and quality control.

The European Union framework also contains an important division of responsibility. EMA operates a Medical Literature Monitoring (MLM) service for specified active substances and designated literature. MAHs remain responsible for monitoring other relevant medical literature within the scope of their pharmacovigilance obligations. EMA states that the MLM service was established under Article 27 of Regulation (EC) No 726/2004 and that MAHs can access the resulting ICSRs through EudraVigilance. citeturn0search1turn0search6

This article therefore covers both sides of the system: what EMA's MLM service does and what an MAH must do when literature remains within its own monitoring responsibility.

A second distinction is equally important. A literature ICSR is not automatically a spontaneous report merely because it was discovered through literature searching. Under ICH E2D(R1), the report type depends on how the case arose. If the publication describes spontaneous observations made in clinical practice, the ICSR is classified as a spontaneous report. If the publication reports cases identified through a study or another organised data-collection system, the ICSR is classified as a report from study, with the relevant study type identified separately. citeturn0search23turn0search24

1. Where Literature Fits in GVP Module VI

Literature is a source of suspected adverse reactions and must be considered within the ICSR framework.

The fundamental sequence is:

Literature source
      ↓
Search / retrieval
      ↓
Screening
      ↓
Potential safety information
      ↓
Clinical and regulatory assessment
      ↓
ICSR validity assessment
      ↓
Case processing / reporting
      ↓
Follow-up and duplicate management

Not every retrieved publication becomes an ICSR. The purpose of screening is to distinguish relevant safety information from records that do not require ICSR processing.

EMA's own MLM inclusion/exclusion document states that only valid ICSRs qualify for reporting and that literature records should therefore be screened, reviewed and assessed against the applicable criteria. citeturn0search21

2. The MAH's Responsibility

MAHs in the EEA are generally responsible for monitoring medical literature concerning their medicines and reporting individual cases of suspected adverse reactions as required.

The existence of the EMA MLM service does not create a general exemption from literature monitoring.

Instead, the MAH must determine whether its active substances and relevant literature fall within the scope of the EMA service. For substances and literature outside that scope, the MAH remains responsible for appropriate monitoring. EMA expressly states this division of responsibility on its MLM page. citeturn0search1

3. What the EMA MLM Service Does

EMA's MLM service was established to monitor selected medical literature for specified active substances and to identify and enter relevant ICSRs into EudraVigilance.

The service reduces duplicated literature processing where multiple MAHs market products containing the same active substance. EMA explains that the service covers selected substances with multiple marketing authorisations and multiple MAHs in the EEA. citeturn0search0turn0search1

The service is therefore best understood as a centralised regulatory literature-monitoring service for a defined scope, not as a replacement for every MAH's literature-monitoring system.

4. EMA MLM Substance Scope

EMA publishes the active-substance groups covered by its MLM service. The selection is linked to the number of marketing authorisations and MAHs and is subject to review according to the resources available to the Agency. citeturn0search1

An MAH should therefore maintain a controlled determination of whether each relevant active substance is covered by EMA MLM.

This determination should not be left to informal knowledge or assumed from the fact that a substance is authorised centrally.

5. Literature Databases Used by EMA

EMA's MLM service uses designated literature reference databases. The detailed EMA guide describes a comprehensive biomedical database, a database with broad pharmaceutical and drug-therapy coverage, and a database focused on complementary and alternative medicine. citeturn0search24

EMA subsequently expanded the MLM service to include MEDLINE from 1 April 2024. EMA explains that MEDLINE contains more than 36 million citations and that the additional search provides earlier availability of articles already represented through broader EMBASE coverage. citeturn0search1turn0search23

The precise databases and journal coverage used by EMA should therefore be taken from the current EMA MLM documentation rather than reproduced as a static list in an SOP or article.

6. EMA Search Frequency

The EMA detailed guide describes different search frequencies for the reference databases used by the MLM service. Its established process includes daily searching of the principal biomedical database, excluding weekends, and monthly searches of the other designated databases according to the database-provider update cycle. citeturn0search24

The current MEDLINE strategy document states that its daily searches refer to calendar days excluding Saturday and Sunday. citeturn0search23

These are EMA MLM service parameters. They should not automatically be treated as a universal prescription that every MAH must reproduce identically in every literature-monitoring system.

7. Search Strategy Is a Pharmacovigilance Control

A literature-monitoring search strategy should be sufficiently comprehensive to identify relevant safety information while maintaining a workable level of precision.

EMA describes its MLM strategies as customised for each substance group using key strings. The searches are applied to the relevant indexed journal dataset and are generally not restricted by language or safety-specific subheadings, because the objective is broad coverage. Search strings are reviewed and updated when necessary to improve precision or accommodate changes to substance variants and thesauri. citeturn0search22turn0search24

This demonstrates an important principle for MAHs: the search strategy itself is part of the control environment.

8. What Should a Search String Contain?

There is no single universal search string that is appropriate for every active substance.

A defensible strategy normally begins with the substance identity and its relevant variants. Depending on the database and monitoring objective, the strategy may need to account for:

• active-substance name;
• synonyms;
• spelling variants;
• salts and relevant chemical forms;
• common abbreviations;
• development or historical names where relevant;
• combination-product terminology;
• and database-specific controlled vocabulary.

The precise syntax must be adapted to the database being searched.

A search strategy should not be made artificially narrow merely to reduce the number of records requiring screening.

9. Search Strategy Documentation

A controlled literature-monitoring system should preserve the rationale and version history of its searches.

Useful records include:

• database searched;
• search date;
• search string;
• search syntax/version;
• substance scope;
• filters and limits;
• inclusion/exclusion logic;
• responsible reviewer or system;
• search results;
• changes from the previous strategy;
• and documented evaluation of search performance.

10. Literature Source Does Not Determine Report Type

A crucial concept under ICH E2D(R1) is that the fact that a case was found in the literature does not, by itself, determine the ICSR's report type.

The distinction is based on how the case arose.

Literature case arising from spontaneous observations

Suppose a physician publishes a case report describing a patient treated in ordinary clinical practice. The physician observed the patient's adverse reaction and subsequently wrote a case report.

The publication is a literature source, but the underlying observation was spontaneous.

Under ICH E2D(R1), the ICSR should therefore be classified as a spontaneous report for the ICH E2B(R3) "Type of Report" element. citeturn0search23turn0search24

Literature case arising from a study

Now consider a publication from a prospective clinical study in which investigators systematically collected adverse events from participants according to a study protocol.

The case was generated through organised data collection rather than a spontaneous observation.

Under ICH E2D(R1), the ICSR should therefore be classified as a report from study. The applicable study type is identified separately in the relevant ICH E2B(R3) study-type data element. citeturn0search23turn0search24

This is particularly important when a published study describes a clinical trial. The fact that the MAH discovers the case through literature monitoring does not convert the underlying clinical-trial case into a spontaneous report.

If the publication does not make the origin clear

ICH E2D(R1) also addresses situations in which it is unclear whether the published case arose from spontaneous observations or a study. In the ICH E2B(R3) "Type of Report" element, the appropriate value in that circumstance is Other, rather than making an unsupported assumption. citeturn0search24

This distinction should be taught explicitly to literature-screening personnel because source of discovery and type of report are different concepts.

11. Why the Distinction Matters

The classification affects how the case is represented in EudraVigilance and how downstream users interpret the data.

Consider two publications:

The important principle is:

Literature is the source through which the information was discovered; it is not necessarily the regulatory origin of the case.

This distinction is explicitly reflected in the revised ICH E2D/E2B framework. citeturn0search23turn0search24

The next chunk will examine inclusion and exclusion criteria in detail, screening stages, special situations identified through literature, full-text review, duplicate publications and how EMA's MLM processing moves from search results to ICSRs.

10. Screening Is Not the Same as Case Validation

EMA's MLM process separates identification of potentially relevant publications from the later assessment of whether individual case information is present and reportable.

This distinction is useful for MAHs as well. A publication can be relevant to the medicinal product but contain no individual case. Conversely, an apparently ordinary article can contain one or more individual cases requiring processing.

The reviewer should therefore ask sequential questions rather than treating publication relevance as synonymous with ICSR validity.

11. First Question: Is the Publication Within Scope?

The first screening decision concerns whether the record falls within the literature-monitoring scope.

Examples of records that may be excluded from the ICSR workflow include material that is not relevant to the medicinal product or substance being monitored, or publications that fall into categories specifically excluded by the applicable EMA MLM criteria.

The exact exclusion criteria should be taken from the current controlled source. EMA's published inclusion/exclusion document is unusually detailed and includes specific exclusions for animal, in-vitro and toxicology studies, among other categories. citeturn0search12

Example

A paper describes toxicity of a compound in cultured human cells but contains no patient exposure or clinical case.

Screening outcome: the paper may be scientifically relevant to the product-development knowledge base, but it is not an individual patient safety case merely because it concerns toxicity.

The EMA MLM criteria specifically identify in-vitro and toxicology studies among exclusion categories. citeturn0search12

12. Second Question: Is There an Individual Patient?

Once a publication is considered potentially relevant, the reviewer assesses whether it contains information concerning an individual patient or patients.

A population-level epidemiological study may provide important safety evidence without providing individual case information that can be processed as an ICSR.

Example

A cohort study reports that patients receiving a medicine had a higher incidence of a particular outcome than an unexposed comparator group, but no individual patients are described.

The publication may be highly relevant for signal detection or aggregate safety evaluation, but it should not automatically be converted into an individual ICSR for every participant.

13. Third Question: Is a Suspected Medicinal Product Identified?

The publication must provide sufficient information to associate the clinical event with a suspected medicinal product under the applicable ICSR framework.

Example

A case report describes severe liver injury in a patient receiving several medicines but the article does not identify which medicine was considered suspected.

The reviewer should assess the actual article and its clinical attribution rather than automatically assigning suspicion to the product being monitored.

14. Fourth Question: Is a Suspected Adverse Reaction Reported?

The publication must contain a reported clinical event that can be assessed as a suspected adverse reaction under the applicable criteria.

Example

A pharmacokinetic paper states that participants received Product X but reports only concentration measurements and no adverse event.

Exposure alone does not create an ICSR.

By contrast, a case report describing Product X followed by a clinically described reaction may require ICSR assessment.

15. Literature Can Contain Multiple Cases

One publication may describe several individual patients.

The reviewer should determine whether the information supports separate individual cases and whether the patients can be distinguished sufficiently for case processing.

Example

A case series describes five patients who developed a particular reaction after treatment. Each patient has a separate clinical history and outcome.

The publication may therefore contain multiple potential ICSRs rather than one case with five reactions.

The correct handling depends on the information actually available and the applicable duplicate and case-processing rules.

16. Full-Text Review

Abstract screening may identify a potentially relevant publication, but the abstract may not contain enough information for final assessment.

EMA's MLM service uses full-text retrieval for confirmed ICSRs and records the relevant full-text request information in its processing workflow. citeturn0search14

For an MAH, the same principle is important: a title or abstract should not be treated as equivalent to a reviewed full publication when material information needed for case assessment is missing.

Example

The abstract says:

"A patient developed severe renal impairment after treatment."

The full article reveals that the patient received three concomitant nephrotoxic medicines and the authors attributed the event to another medicine.

The full-text information may materially change the case assessment.

17. Inclusion and Exclusion Are Evidence-Based Decisions

The EMA MLM inclusion/exclusion framework contains defined categories rather than relying on a general judgement of whether a paper "looks relevant". It also records exclusion criteria in the MLM tracking process. citeturn0search12turn0search14

An MAH should similarly maintain controlled criteria and sufficient documentation to explain significant screening decisions.

The purpose is not to maximise the number of cases created. It is to identify the relevant safety information accurately and consistently.

18. Special Situations in Literature

Literature can contain special situations such as:

• medication errors;
• overdose;
• misuse or abuse;
• off-label use;
• pregnancy exposure;
• occupational exposure;
• lack of efficacy;
• and product-use problems.

The presence of a special situation does not automatically answer whether a valid ICSR exists. The reviewer should apply the relevant GVP VI principles to the actual information in the publication.

Example

A published case reports accidental double dosing but no adverse reaction.

The article may be safety-relevant, but the processing decision should follow the applicable special-situation and ICSR criteria rather than automatically creating an adverse reaction that was not reported.

19. Animal, In-Vitro and Toxicology Publications

Not every safety-related scientific publication belongs in the ICSR stream.

Animal experiments, in-vitro experiments and toxicology studies can contribute to the scientific safety assessment of a medicinal product, but they do not normally provide an individual human case.

EMA's MLM inclusion/exclusion document specifically identifies these categories within its exclusion framework. citeturn0search12

This is an important example of why literature monitoring should not be reduced to keyword matching.

20. Conference Abstracts

Relevant conference abstracts can contain safety information before a full journal article is available.

The MAH's literature-monitoring process should therefore address applicable meeting abstracts and other relevant published material within its monitoring scope.

EMA's inspection guidance specifically notes that MAHs should maintain awareness of relevant published abstracts from meetings, draft manuscripts and local medical journals as part of mandatory literature-monitoring activities. citeturn0search5

21. Local and Non-English Literature

Global literature monitoring should not be limited to English-language publications when the applicable requirements extend to literature in countries where the medicinal product is authorised.

EMA inspection guidance states that relevant local medical journals should be monitored and that the frequency should be aligned with the applicable literature-monitoring requirements. citeturn0search5

Where translation is needed, the process should preserve the original publication and document the translation used for safety assessment.

22. Duplicate Publications

The same patient may appear in several publications, for example as an initial case report followed by a detailed case report or follow-up publication.

A literature-monitoring system therefore needs duplicate controls at both publication and case level.

The existence of a second publication does not automatically mean that a second ICSR should be created.

The reviewer should determine whether it represents:

• the same patient;
• additional follow-up;
• a different patient;
• or a genuinely separate case.

23. Literature Case and Existing Spontaneous Case

A published case may describe a patient whose reaction was previously reported spontaneously.

The literature source should be assessed against the existing database records. Where it represents the same case, the information should normally be incorporated appropriately rather than creating an artificial duplicate.

This is particularly important because literature may contain considerably more clinical information than the original spontaneous report.

24. From Screening to ICSR

The complete workflow can be represented as:

Search result
    ↓
Scope screening
    ↓
Potentially relevant publication
    ↓
Patient/case assessment
    ↓
Product + reaction assessment
    ↓
Full-text review where required
    ↓
Validity assessment
    ↓
Duplicate assessment
    ↓
Case creation/update
    ↓
Reporting and follow-up

EMA's MLM user guide similarly records screening and reviewing outcomes, confirmed or potential ICSRs, seriousness and full-text requests in its tracking workflow. citeturn0search14

25. The Reporting Clock for Literature Cases

For literature ICSRs identified through the regular global reference-database search, EMA inspection guidance states that the regulatory reporting day zero is the date on which the search was conducted, when sufficient information for a valid ICSR is identified through that activity.

For other mandatory literature-monitoring activities, such as relevant meeting abstracts, draft manuscripts and local medical journals, day zero begins when the MAH identifies sufficient information to establish the minimum criteria for a valid ICSR. citeturn0search5

This distinction is operationally important and should be reflected in controlled procedures and training.

26. Practical Screening Examples

Example A — Relevant science, no ICSR

A meta-analysis reports an increased risk of arrhythmia across 40 studies but contains no identifiable individual patient cases.

Outcome: safety-relevant scientific evidence; not automatically 40 ICSRs.

Example B — Clear literature ICSR

A case report identifies a patient, identifies Product X as the suspected medicine and describes a specific adverse reaction with a clinical chronology.

Outcome: assess as an individual literature case under the applicable ICSR criteria.

Example C — Abstract insufficient, full text decisive

An abstract suggests a suspected reaction, but the full text identifies another medicine as the authors' suspected cause.

Outcome: final assessment should incorporate the full-text evidence.

Example D — Same patient, second publication

A short initial case report is followed six months later by a detailed publication concerning the same patient.

Outcome: assess for follow-up/duplicate relationship rather than automatically creating a new case.

The next chunk will cover search-string construction in greater technical depth, search validation and maintenance, EMA MLM versus MAH responsibilities, inspection findings, outsourcing, quality controls, practical end-to-end examples, References and the Regulatory Note.

27. How EMA Builds a Medical Literature Monitoring Search Strategy

EMA's Medical Literature Monitoring (MLM) service is not based on a single generic pharmacovigilance search string. The Agency defines substance groups, identifies the literature reference databases to be searched, develops database-specific strategies using the applicable thesaurus and validates those strategies before operational use. The search strategies are reviewed and updated when necessary. citeturn0search22turn0search23

For MEDLINE, EMA states that the strategies are designed to be comprehensive for the areas monitored and use the syntax and terminology of the MEDLINE search engine. The published strategy document provides the actual substance-group strategies rather than a universal string that can simply be copied for every medicine. citeturn0search23

28. What Goes Into a Search String?

A controlled literature search normally has several conceptual components:

Medicinal-product concept
        +
Clinical/safety concept
        +
Database-specific indexing / thesaurus terms
        +
Relevant synonyms and variants
        +
Defined scope and limits

The exact construction depends on the database and the monitoring objective.

A medicinal-product concept may need to account for the active substance, established synonyms and other searchable variants. The safety component should be designed to retrieve the types of safety information within scope without unnecessarily restricting the search to a narrow list of adverse-event terms.

EMA's published MEDLINE strategies illustrate this principle: the Agency states that its parameters aim to retrieve suspected adverse reactions from spontaneous and solicited reports as well as specified special situations. citeturn0search23

29. Why a Search String Should Not Be Copied Blindly

A search strategy developed for one database may not be valid in another.

Differences can include:

• controlled vocabulary;
• field names;
• indexing conventions;
• truncation rules;
• proximity operators;
• spelling and synonym handling;
• and date or publication filters.

EMA explicitly describes validation of the search strategy against the applicable thesauri and database before operational use. citeturn0search22

A MAH should therefore maintain database-specific, version-controlled strategies rather than treating a search string as a static paragraph of Boolean text.

30. Search Sensitivity and Precision

Literature monitoring involves a practical balance between retrieving enough relevant records and avoiding an unmanageable volume of irrelevant material.

A strategy that is too narrow may miss relevant publications. A strategy that is excessively broad may generate large numbers of records that consume screening resources without improving case detection.

For safety surveillance, a missed relevant publication can be more consequential than an additional irrelevant screening record. The strategy should therefore be designed and validated according to the applicable monitoring objective and documented risk assessment.

EMA's MLM strategy process explicitly includes validation and subsequent review of search strategies where changes are needed to improve precision or align with changes in the indexing thesaurus. citeturn0search22turn0search23

31. Inclusion and Exclusion Are Not the Same as Search Terms

Search terms determine which records are retrieved. Inclusion and exclusion criteria determine how retrieved records are assessed.

These should not be collapsed into a single Boolean query.

For example, a search may deliberately retrieve a broad set of records, after which screening excludes animal studies, in-vitro studies, toxicology studies or publications containing only aggregated patient information where the applicable EMA criteria call for exclusion from ICSR processing. citeturn0search21

This separation makes the process more transparent and allows the search strategy and screening criteria to be maintained independently.

32. Example: Why a Safety-Term-Only Search Can Fail

Suppose a company searches:

("Product X") AND ("adverse event" OR "adverse reaction")

This may retrieve some obvious case reports but can miss relevant publications that describe the clinical event without using those exact expressions.

A robust strategy should therefore be designed around the database's indexing system and the safety-information scope rather than assuming authors will use a standard pharmacovigilance vocabulary.

33. Example: Why the Product Concept Must Be Controlled

A substance may appear under:

• international nonproprietary name;
• development code;
• salt or ester form;
• established synonym;
• abbreviation;
• or other indexed representation.

The applicable search strategy should determine which variants are necessary. However, variants should be evidence-based and controlled rather than continuously added without validation.

34. EMA MLM and the MAH's Literature Responsibility

EMA's MLM service covers defined active substances and defined literature databases. It does not remove the MAH's general responsibility to monitor literature outside the scope of the Agency's service.

EMA explicitly states that MAHs remain responsible for literature monitoring and reporting for active substances not covered by the service and for literature not monitored by EMA, including relevant literature for substances otherwise covered by MLM. citeturn0search2

Therefore:

EMA MLM coverage
        ≠
Complete MAH literature-monitoring responsibility

The MAH should maintain a documented coverage assessment showing which literature sources are covered by EMA and which remain within the MAH's own monitoring system.

35. What EMA MLM Actually Delivers to MAHs

For cases identified through the EMA service, EMA transmits relevant ICSRs through the EudraVigilance Gateway and makes them available to concerned MAHs through the ICSR download functionality. citeturn0search2

This service reduces duplicate entry of the same literature case into EudraVigilance, but the MAH still needs to incorporate the information into its own pharmacovigilance system and fulfil applicable obligations outside the scope of the EMA service.

36. Search Strategy Governance

A mature MAH literature-monitoring system should maintain, at minimum:

• approved search strategies;
• database and platform definitions;
• search frequency;
• date of implementation;
• search execution records;
• strategy validation evidence;
• search-result records;
• screening decisions;
• exclusion rationale;
• full-text retrieval records;
• case-processing records;
• duplicate assessment;
• and change-control history.

The precise records required should follow applicable regulatory requirements and the organisation's quality system.

37. Change Control

Search strategies should not be changed informally.

A change may be triggered by:

• a new product synonym;
• a new salt or formulation;
• a database indexing change;
• a new thesaurus version;
• a change in the medicinal-product portfolio;
• a newly recognised safety issue;
• a regulatory change;
• or evidence that the existing strategy is generating unacceptable false-negative or false-positive results.

The change should be assessed, approved, implemented and, where appropriate, retrospectively evaluated.

38. Inspection Perspective

An inspector may reasonably ask the organisation to demonstrate not merely that literature searches are performed, but that the search system is capable of identifying relevant safety information and is adequately controlled.

Useful evidence includes:

• the current search strings;
• historical versions;
• validation records;
• search logs;
• screening records;
• exclusion decisions;
• missed-case investigations;
• quality-control results;
• vendor oversight records;
• and evidence that strategy changes were implemented correctly.

EMA's own MLM process provides a useful model of this controlled approach: search strategies are defined, validated, published, applied and reviewed for required updates. citeturn0search22

39. Outsourcing Literature Monitoring

An MAH may use a third party to perform literature searches or screening. ICH E2D(R1) expressly recognises that MAHs may conduct searches themselves or use external services. The responsibility for appropriate oversight and compliance nevertheless remains with the MAH. citeturn0search24

The quality agreement should therefore address, as applicable:

• search scope;
• databases;
• search strings;
• frequency;
• screening criteria;
• escalation;
• reporting timelines;
• quality control;
• business continuity;
• inspection support;
• and change control.

40. Literature Case Type: Study Versus Spontaneous Observation

The fact that an ICSR was discovered in the literature does not determine its E2B report type.

ICH E2D(R1) provides the critical distinction:

• where the case arose from a spontaneous observation described by the author, it is classified as a spontaneous report;
• where the case arose from a study or other organised data-collection system, it is classified as a report from study.

Thus, a published clinical-trial case discovered through literature monitoring does not become a spontaneous case merely because the MAH found it through a literature search. citeturn0search24

Example

A journal article describes five patients who experienced adverse reactions during a prospective clinical trial.

The MAH discovers the article through its literature-monitoring process.

The source of discovery is literature, but the origin of the cases is the clinical study. The cases should therefore be classified according to the study origin, not as spontaneous reports merely because they were published.

41. Literature Day Zero

ICH E2D(R1) states that the reporting clock for literature ICSRs begins when the MAH or its third party identifies sufficient information to determine that the ICSR reporting criteria are met. Where follow-up is necessary to establish those criteria, day zero is the date sufficient follow-up information is received, subject to regional requirements. citeturn0search24

For EMA MLM specifically, the Agency's operational processes have defined search and screening activities and make resulting ICSRs available through EudraVigilance. The MAH should therefore understand the difference between an EMA-generated MLM case and a case identified through its own literature-monitoring activity.

42. End-to-End Example

Consider a company monitoring Product X.

Step 1 — Search: the approved MEDLINE strategy retrieves a case report.

Step 2 — Screening: the publication concerns a human patient exposed to Product X and describes a suspected adverse reaction.

Step 3 — Full text: the article identifies the patient, reporter and suspected product and provides the clinical event.

Step 4 — Origin: the authors describe the patient from their clinical experience rather than a planned study.

Step 5 — Classification: the literature case is classified as a spontaneous report under ICH E2D(R1).

Step 6 — Duplicate check: the MAH checks whether the patient was previously reported through another source.

Step 7 — Processing: the case is entered or linked appropriately, with the literature citation retained as the source.

Now change Step 4: the article is a report from a prospective clinical trial. The case origin is then study-based and the E2B report type should reflect that study origin. citeturn0search24

43. Key Takeaways

• Medical Literature Monitoring is a controlled pharmacovigilance process, not simply a PubMed search.
• EMA MLM uses substance-group-specific, validated strategies and defined literature databases. citeturn0search22turn0search23
• Search strategy and inclusion/exclusion criteria are separate controls.
• EMA MLM does not remove the MAH's responsibility for literature outside EMA's coverage. citeturn0search2
• A literature source does not determine the ICSR's report type.
• A published case from spontaneous clinical experience is a spontaneous report; a case arising from a study is a report from study. citeturn0search24
• Duplicate management is essential because the same patient may appear in several publications or safety sources.
• Search strategies require validation, version control and change management.
• Outsourced literature monitoring remains an MAH oversight responsibility.

References

1. European Medicines Agency. Medical Literature Monitoring and entry of relevant information into EudraVigilance by the European Medicines Agency — Inclusion and Exclusion Criteria for processing of Individual Case Safety Reports. EMA/119265/2015.
2. European Medicines Agency. Detailed Guide regarding the monitoring of medical literature and the entry of relevant information into the EudraVigilance database by the European Medicines Agency. EMA/161530/2014.
3. European Medicines Agency. European Medicines Agency MLM Service MEDLINE Search Strategy. EMA/130770/2024.
4. European Medicines Agency. Medical Literature Monitoring — EMA service and business processes.
5. European Medicines Agency. ICH E2D(R1) Guideline on post-approval safety data: definitions and standards for management and reporting of individual case safety reports. EMA/CHMP/ICH/59123/2024.
6. European Commission. Directive 2001/83/EC, as amended.
7. European Commission. Regulation (EC) No 726/2004, as amended.
8. European Medicines Agency. GVP Module VI — Collection, management and submission of reports of suspected adverse reactions to medicinal products.

Regulatory Note

This article explains EU medical literature monitoring and literature ICSR processing using current EMA and ICH material. It distinguishes the EMA MLM service from the broader literature-monitoring responsibilities of marketing authorisation holders.

ICH E2D(R1) has a legal effective date in the EU of 18 March 2026. EMA's implementation strategy and the current GVP framework should be checked when applying the article to a live process because implementation requirements and GVP Module VI are subject to regulatory change. citeturn0search5

Examples in this article are illustrative and are not presented as real inspection cases unless a source is explicitly identified.