OR WAIT null SECS
© 2023 MJH Life Sciences™ and Applied Clinical Trials Online. All rights reserved.
Processes for collecting data of serious adverse events (SAEs) and events of special interest (ESI) during clinical trials have been underwhelming for sponsors. Improving these processes using the latest digital advances can result in more complete analyses and more effective decision-making.
Current processes for the collection of serious adverse events (SAEs) and events of special interest (ESI) during clinical trials are not providing sponsors the appropriate levels of completeness and accuracy.1 While these data quality issues are not unique to clinical studies,2 within this experimental environment of clinical research, when the knowledge of safety should be evolving rapidly based on very limited numbers of reports, these issues hamper the improved understanding of product profiles. Ultimately missed content, unavailable for expert adjudication by Data and Safety Monitoring Committees or pharma company experts creates ambiguity and may impair decision-making on study conduct. Improving the process of SAE/ESI capture within trials through the latest digital advances enables more robust analyses and, therefore, identification of individuals or groups carrying greater risk profiles vs. the broader population.
The increasing number of SAEs twinned with a static approach to both pharmacovigilance (PV) data capture and processing will lead to unsustainable, escalating costs. In a time of cost-containment pressure within the pharma industry and beyond, this will lead to an unjustifiable budget environment and a demand for increased levels of control. Business process optimization reviews carried out by PV heads should undertake the following holistic review of resources and their application:
Process: Is the clinical process for safety reporting under control?
ACTION: Ensure dedicated management control over timelines, data visibility and data classification.
Content: Is the content correct? Is your data of high quality & completeness?
ACTION: Relevant, complete data needs to be captured first time at the point of reporting.
Cost: Are your costs increasing? Are your existing processes and technologies scalable?
ACTION: A scalable, automated process should be deployed that can be managed appropriately. This provides the added benefit of follow-up costs being minimized.
Safety data collection in clinical studies challenges
The design of a clinical trial is defined by a protocol and precise study plan, requiring investigators to conduct the study according to the protocol and participating patients to attend a site or clinic for specified study visits. Traditionally, AE collection is triggered through data collected via EDC or the clinical data system utilized by the pharma or CRO, however these systems are updated typically on a visit-by-visit basis, with the eCRF designed to reflect the specified visit schedule. While there are often opportunities to collect some adverse event related data, the reality is that adverse events do not appear at specific times. SAEs most often occur while the patient is not at a routine visit and if medical treatment is required this will be given by a medical practitioner outside of the clinical trial community. In these cases, key data collection for SAEs/ESIs may be overlooked; once missed, this current data cannot be re-created (e.g. ECGs, cardiac enzymes, renal function etc.) and no concerted follow-up can retrace the history vital to understanding such events.
Reportable SAEs must be processed and distributed to regulatory authorities, investigators and IRBs within an expedited timeline. In addition, predefined reports must be routed for expert adjudication to Safety and Data Monitoring Committees. Their assessment is wholly dependent on the quality and completeness of the data provided to them, determining not only the continuation of studies but ultimately the robustness of study results.
The obvious near-term safety impact of missing data is clear, but the downstream compliance effects are also seen during regulatory inspections. A recent article3 states, “Clinical trials have two main endpoints, one is efficacy and the other safety, being both of them of equal importance from the regulatory standpoint. Unreported or under-reported adverse events are a major problem during regulatory inspections. The monitor has the responsibility to capture any issue with reporting before the inspector does, and ensure compliance.” While monitor follow-up may be able to identify some additional data points taken at the time of the event, patients and healthcare professionals (HCPs) should have the ability to record these events at the time of occurrence to ensure collection and recording of relevant data at the first interaction.
Much has been written about the inadequacy of causality assessments in clinical trials. Differing medical opinions can be reached; initially by the investigator, then the study monitor, pharmacovigilance team and ultimately the Safety and Data Monitoring Committees. The classification of an event’s relationship to the study medication is critical to its definition as an adverse event. Misinterpretations by investigators can lead to under-reporting and impact the timeliness of safety decisions and be flagged in an inspection later in the process. As an example of the disagreements identified in causality assessments, we can review the results of a prospective monitoring of fatal SAEs in a Phase III trial in patients with advanced colorectal cancer.4 In this example, the study team ‘recorded a disagreement between the assessment of the local investigators and the IDMC5 on the relationship between the study drugs and death in 65% of the patients whose charts were reviewed. Local investigators frequently underestimated the relation between the administration of study drugs and death.’
Any misclassification of whether an adverse event is serious or not has the potential to skew data analyses. A recent instance can be reviewed in the FDA briefing document for NDA 21071 Avandia6 where investigators failed to apply the criteria of ‘medically serious’ to classify the cases correctly: ‘Because endpoints had to involve a death or an overnight hospital stay, investigators appear to have been confused that SAEs also had similar requirements. Because strokes are arguably always serious, we checked for stroke AEs. We found three cases for which the investigators classified strokes as AEs rather than SAEs. For two of the cases, the monitor queried the investigators that the strokes should be SAEs but the investigators insisted they were AEs because the patients were not hospitalized’.
Current processes are still often paper-based and require notable site staff time, particularly within studies with a moderate or large number of SAEs. Significant SAE reconciliation efforts are required due to the collection of redundant data and unnecessary follow-up tasks are often generated from EDC requiring the drug safety scientist to update the SAE report with data not relevant to the SAE.
Benefits of AE event-driven data
Fit-for-purpose, digital safety data collection solutions provide simple, user-friendly safety data capture at initial reporter interaction. Such solutions deliver increased accuracy and improved timeliness of information, mandatory capture of data relevant to the SAE in question, a reduction in the collection of redundant data and savings in site staff time, streamlining of the query resolution process, a reduction in the SAE reconciliation effort required before database lock as no redundant information is collected, enable the company to meet reporting deadlines with greater ease and provide a time reduction spent on unnecessary ‘follow up’ information generated from on-going EDC updates.
They should include:
The founders of MyMeds&Me identified the gaps in the end-to-end process of drug safety during their careers leading global organizations within pharma. The Reportum solution was conceived and developed to address each of these requirements and, in particular, to access better safety data at source while simultaneously enabling greater effectiveness of pharmacovigilance operations.
The benefits of digital automation of the safety data collection process in clinical studies are far reaching. Organizations already on this transformational path may have begun this business process optimization with their initial focus on cost containment and regulatory reporting obligations, however the visibility, speed and completeness of data enables much deeper improvements.
Regulators and patients are pushing for greater benefit-risk information and this can begin early in the process. Safety and Data Monitoring Committees and the pharma companies pose the questions, ‘Does the drug work? In whom does it work? Does the drug cause risk? Who is at risk? and How do we maximize benefit-risk?’ The digitization of this process enables consistency of safety data capture across studies and indeed across a pharma company linked to the prompt review of that data.
Safety and clinical teams must be equipped to identify safety issues with consistency and speed so that there is an opportunity to amend the protocol, amend consent or even stop the study, depending on the findings. Obviously pharma company PV teams are required by regulations to be transparent on any new information that impacts benefit-risk with the various stakeholders within a clinical trial: HCPs, ethics committees, patients and regulatory authorities. This updated information ensures informed decisions to be made by both investigators and patients particularly if consent forms are updated. Decisions can then be made by each party informed by full benefit-risk information.
Increased data integrity and transparency provides:
Deploying technology that drives the capture of relevant safety data at the point of reporting will transform the way clinical trials are conducted and analyzed. The richer and more granular data will lead to a greater understanding of risk profiles at the individual level as well as clearer understanding of sub-groups that carry higher risks. Future treatment regimens could be optimized based on the detailed individual profile knowledge and treatments can finally be geared to individuals and not just to the broad population.
Andrew Rut, MD is CEO and Founder of MyMeds&Me