Signal Detection and Management Procedures
In this article, we describe the process for detection, monitoring, evaluation and reporting of safety signals arising from individual case safety reports (ICSR) or from aggregate data for any medicinal product. It is difficult to talk about signal detection and management without mentioning pharmacovigilance analytics and how they are applied at different stages of the signal management process.
In general, this process applies to all medicinal products, covering their entire life cycle, specifically including clinical development and post-market phases, for any type of adverse event, serious or non-serious.
Personnel in charge of patient safety and involved in aggregate or individual safety data review, are normally affected by these procedures.
The objective of the ongoing monitoring is to identify case reports or case report series of adverse events (AE) or adverse drug reactions (ADR), that is worthy for further exploration and potentially requires safety actions such as a safety signal investigation.
This ongoing monitoring of adverse event reports comprises the retrieval of data from the global safety database at monthly intervals for all monitored products, and review of the data with the purpose of timely identification of (potential) new safety signals requiring further investigation. The monitoring also comprises data retrieval from the clinical databases for the analysis of non-serious adverse events.
Retrieval strategies (case selection criteria, format, and periodicity of retrievals from the global safety database, for example) should be defined in advance on a per-product basis. Retrieval strategies and results of all ongoing monitoring activities must be documented.
The signal management process is based on the examination of individual case safety reports (ICSR), aggregate data from active surveillance systems or studies, and literature information or other data sources.
The signal management process includes the following steps:
- Signal detection
- Signal validation
- Signal prioritization
- Signal assessment
- Recommendation for action
- Exchange of information
Detected potential safety signals are processed within patient safety department of the company. Other company functions and external resources may be involved as required.
Generally accepted procedures for signal investigation, objective thresholds for accepting safety signals, or generalizable rules for subsequent action do not exist. Expert knowledge and medical judgment are always required.
Likewise, the timelines for processing a potential signal depend on the severity and potential public health impact on the population concerned. Potential safety signals that, if accepted, may pose a significant public health threat (e.g. an unexpected adverse event that is both serious and frequent) should be processed with priority until timelines are defined by the safety management team and/or regulatory authorities.
The signal management process is always based on the information available at the time of the review and may change over time.
Events which do not fall under the definition of a reportable (valid) ICSR but may affect the benefit-risk balance of a medicinal product and/or impact on public health shall be notified as Emerging Safety Issue, per applicable regulatory requirements.
Signal Detection: Input, Periodicity and Materials
The Pharmacovigilance Scientist defines the retrieval strategy on a per-product basis depending on the maturity of the safety profile and number of case reports for the concerned product. At a minimum, retrieves no less frequently than monthly a standard dataset comprising the following:
- Summary tabulation of AEs / ADRs for the monitoring period including, but not limited to, designated medical events (DME) and targeted medical events (TME), relating to new cases over the time period under study.
- Cumulative summary tabulation listing all AEs / ADRs on the database for the product.
- Aggregate reports (monthly requests) of adverse events including serious, non-serious, and any other events of interest, obtained from:
- Clinical studies
- Regulatory reports
- Commercial complaints
- Preclinical in vitro and in vivo studies
- Epidemiologic data
- Media. Internal and external websites, and social media
- Medical literature
- Data from off-label use
Aggregate safety data may include the following elements: MedDRA coded terms, by System Organ Class (SOC) and Preferred Term (PT); frequency of events; nature and type of events (serious, non-serious, events of special interest, expectedness, relatedness).
Also, it will be important to:
- Define additional retrievals (additional datasets and/or higher monitoring frequency depending on, for example, regulatory commitments, an existent risk management plan (RMP), or known safety issues.
- Document the retrieval strategy on a per-product basis via a “Product Safety Signal Monitoring Plan”.
Ongoing Monitoring Activities for Signal Detection
Following are the steps PV scientist will perform, regarding ongoing monitoring:
- Retrieve tables and listings as defined in the Product Safety Monitoring Plan.
- Review the retrieved data set within 1 week after retrieval and document in the Product Safety Signal Monitoring Tracking Sheet.
- Note: generally accepted standards for ongoing monitoring of thresholds for identifying safety signals do not exist. Expert knowledge and medical judgement are always required.
- Consider the following elements during ongoing monitoring of case reports for signal identification:
- Information from summary tabulations, e.g. overall number of cases, number of fatal or serious cases, number of unlisted cases, number of cases per MedDRA SOC or of certain PTs, for the monitoring period and for cumulative data.
- Line listing information, as demographics (age, gender), dose of suspect product, temporal relationship, information on de-challenge and re-challenge.
- Obvious alternative explanations for the AE, like underlying diseases; or technical reasons as for example recent data migration, or solicited reporting systems.
- Causality assessment.
- Specific topics / medical concepts to be monitored, if applicable.
- Document the ongoing monitoring activities and any findings via the Product Safety Signal Monitoring Tracking Sheet.
PV personnel may identify potential safety signals through various pharmacovigilance activities, including the following:
- Ongoing monitoring of AEs / ADRs.
- Individual medical review of ICSRs and customer product quality complaints.
- Preparation of aggregate reports (as for example the Periodic Benefit Risk Evaluation Report, PBRER) and RMPs as applicable to the medicinal product.
- Review of scientific and medical literature.
- Data obtained from company-sponsored clinical and non-clinical studies, including surveillance systems.
- Information obtained from a health authority.
- Other, such as data on quality, systematic reviews, meta-analyses, internet and digital media under the management or responsibility of the MAH.
PV personnel may employ some combination of statistical and clinical methods for the evaluation of a potential signal including the following:
- Careful review of individual case details.
- Comparing rates to an historical period of reporting rates.
- Using more reliable data sources such as incidence rates from previous clinical studies.
- Preclinical studies from biologic effects, or pharmacokinetics or pharmacodynamics effects.
- Search for additional cases that meet similar criteria (similar events) of the signal in the product-specific exposed population of interest or from experience with similar products in the class.
PV Scientist will determine if a new or potential safety signal exists that warrants further investigation including, but not limited to:
- New AEs, not currently documented in the RSI or in the package insert, specially if they are serious and have occurred in rare sub-populations.
- An apparent increase in the severity of an AE that is included in the RSI or in the package insert.
- Occurrence of serious adverse events (SAE) known to be extremely rare in the general population.
- Previously unrecognized interactions with other products, supplements or food.
- Identification of a previously unrecognized at-risk patient population or subgroup of patients, such as patients with specific medical conditions, comorbidities, or with specific racial or genetic predispositions.
- Adverse events arising from the way a product is being used either on or off-label (e.g. adverse events seen at doses higher than those normally prescribed or in sub-populations not recommended in the label.
- Adverse events arising from user errors, or from medication errors.
- Concerns arising from potential inadequacies of a currently implemented risk minimization action plan (RiskMAP), as for example reports of a serious adverse event that appears to indicate failure of a risk minimization goal.
- Other concerns that may be identified by PV department or a regulatory agency.
PV scientist notifies the PV physician immediately of all potential signals detected.
PV scientist adds the detected -potential- safety signal to the product safety signal monitoring tracking sheet.
Normally, the PV scientists performs the following steps for signal validation:
Evaluate the data supporting a detected signal in order to verify that the available source documentation contains sufficient evidence demonstrating the existence of a new potentially causal association or a new aspect of a known association, and therefore justifies further prioritization and assessment of the signal.
Depending on the nature and context of the signal investigation, the below describes aspects that will be considered when validating a signal:
On clinical relevance:
- Strength of the association with the product
- Evidence of dose-response effect
- Frequency, that is, for example the number of spontaneous reports in comparison to earlier periods and/or in relation to estimated patient exposure; same type of information in the context of clinical trial data.
- Quality of the reports. Completeness of data, plausibility of the information, availability of data to substantiate reported diagnosis.
- Reporter and company causality assessment of individual cases.
- Temporal relationship of the product use and event, including information on de-challenge and re-challenge.
- Consistency of data patterns indicating potential risk groups.
- Consistency of findings across available data sources.
- Specificity of a case series (for example, same histopathology or subtype of a disorder is reported in all cases of a series of reports.
- Alternative medical or technical explanations.
- Seriousness and severity of the reaction and its outcome, relative to the disease being treated.
- Drug-drug interactions.
- Potential to mitigate the risk in the population.
- Feasibility of a further study using controlled or observational designs.
- Degree of benefit the product provides, including availability of other therapies.
On previous awareness:
- Biological plausibility of the event in light of the known or assumed pharmacological properties of the suspect drug or the drug class.
- Extent to which information is already included in the RSI or package insert.
- Association has already been addressed in an aggregate report, or has been subject to a regulatory procedure.
In principle, only new signals for which there is no previous awareness should be validated. However, an already known association may give rise to a new signal if its apparent frequency of reporting, duration, severity or a change in the previously reported outcome (e.g. deaths) suggests new information on the already known association.
Availability of other relevant sources of information providing a richer set of data on the same AE /ADR:
- Literature findings
- Experimental findings or biological mechanisms
- Screening of databases with larger datasets (e.g. FAERS database)
After the investigation of the safety signal, the conclusion can be:
- Validated and accepted: a causal association between the product and the event is assumed
- Not validated and rejected: no causal association between the product and the event is assumed, or
- Pending, not confirmed signal: no clear conclusions regarding causality can be drawn. The signal is further monitored and re-evaluated at a defined time point.
Document this conclusion in the product safety monitoring tracking sheet. Set deadline or threshold for re-investigation of pending signals and document in the product safety monitoring tracking sheet.
In collaboration with the PV physician, document each safety signal investigation depending on its context using the product safety signal investigation report, and sign the form. Include the documentation background, information sources and methods, results and conclusion of the investigation. Examples of additional sources of documentation include the company comments on the ICSRs, aggregate reports, or other documents submitted to health authorities.
The following steps are to be performed by the PV scientist:
The PV scientist will promptly identify validated signals with important public health impact or that may significantly affect the benefit-risk profile of the medicinal product in treated patients. These signals require urgent attention and need to be prioritized, that is, evaluated without delay.
For prioritization of a signal, consider the following aspects, if applicable:
- Impact on patients depending on the severity, reversibility, potential for prevention and clinical outcome of the association
- Consequences on treatment discontinuation on the disease and the availability of other therapeutic options
- Strength and consistency of the evidence supporting the association as, for example, biological plausibility, a high number of cases reported in a short period of time, identification of the signal in different settings, data sources or countries
- Clinical context, that is, whether the association suggests a clinical syndrome that may include other reactions
- Public health impact, including the extent of utilization of the product in the general population and in special populations and the patterns of the medicinal product utilization
- If a marketing authorization application for a new medicinal product is still under evaluation
- Enter the outcome of the signal prioritization process in the product safety signal monitoring tracking sheet
Signal Assessment – Evaluation of Risk
The objective of signal assessment is to further evaluate the significance and potential risk of a validated signal so as to identify the need for additional data collection, risk mitigation or minimization activities in a timely manner, or for any regulatory action.
The following steps will be performed for signal assessment:
- Review appropriate internal and external sources to obtain further information (e.g. literature articles, application dossier, spontaneous reports, expert consultation, or information held by regulatory authorities).
- Assess the significance of a signal at a broader level, at the therapeutic or at the System Organ Class (SOC) level, or at the level of a Standardized MedDRA Query (SMQ) to obtain a potential link to a complex disease, to a prior stage or a reaction or to clinical complications of the adverse reaction of interest.
- Document the risk assessment of the signal per product safety signal investigation report, and recommend no further action, or further action to prevent or minimize patient risk as described in the next section, Recommendation for Action.
Recommendation for Action
The next steps are performed by the PV physician:
Based on the level of safety risk or patient impact, recommend action and review recommendations with the product safety board. Recommended actions may include, but are not limited to, any of the following with appropriate time frame for action consistent with the nature and severity of the signal:
- Initiation of a Health Hazard Assessment (HHA) for potential field action (field alert reporting evaluation)
- Request quality complaint investigation for further product evaluation
- Expedited reporting to regulatory agencies
- Direct healthcare professional communication / Dear Doctor Letters
- Updating safety related labeling or prescribing information
- Clinical expert statements
- Reporting to investigators, Institutional Review Boards (IRB), Ethics Committees, updating study documents, or holding or stopping ongoing studies early
- Continued assessment of the product benefit-risk balance
- Further investigation of the safety risk through additional studies
- Development of a pharmacovigilance plan focused on evaluating the identified risk
- Reporting via periodic report submission
- Risk management document updates
- Additional educational materials or training
For all validated signals, and in accordance with final recommendations from the committee:
- Modification of the ongoing monitoring strategy of the product
- Initiation of label change and/or other external communication activities
- Initiation of recall/correction procedure
- Information to concerned health authorities
- Issuing or updating a Risk Management Plan
- Introduction of enhanced pharmacovigilance activities
- Introduction of additional risk minimization activities
- Conducting a post-authorization safety study
- Periodic review of the signal
The product safety review committee will determine appropriate timelines for initiation and completion of suitable actions for all validated signals, which will be documented in the meeting minutes and in the individual product safety signal investigation report.
The PV scientist documents all detected signals including validation, prioritization, assessment, actions including timelines and reporting dates, as well as current status in the product safety signal monitoring tracking sheet.
Exchange of Information
The following steps will be performed by the PV physician:
- Communicate immediately to regulatory affairs (to generate prompt regulatory action as communication to applicable regulatory authorities and interested business partners) as an Emerging Safety Issue all validated signals pointing towards an implication for public health or the benefit-risk profile of the specific product.
- Depending on the severity of the signal, communicate validated signals representing a new potential signal or a new aspect of a known risk and not having implications for the benefit-risk profile to applicable regulatory authorities.
- Communicate the outcome of signal assessment involving new or changed risks to the public including health care professionals and patients as well as to the concerned marketing authorization holders.