What is an Important Protocol Deviation (PD)?
TransCelerate’s Protocol Deviation team has applied ICH E6 R2 approaches to support the identification of important protocol deviations and three key principles to clarify the base definition of what constitutes a protocol deviation.
“Important protocol deviations are a subset of protocol deviations that may significantly impact the completeness, accuracy, and/or reliability of key study data or that may significantly affect a subject’s rights, safety, or well-being. For example, important protocol deviations may include enrolling subjects in violation of key eligibility criteria or failing to collect data necessary to interpret primary endpoints, as this may compromise the scientific value of the trial.”
As of January 2023, the Protocol Deviations Initiative is complete. There are no new solutions or updates planned.
Protocol Deviation or PD Management processes are complex, varied, and broadly impacted by a wide range of definitions for identification and classification. These definition differences are associated with process inefficiencies for both sides and sponsors, that can hinder rapid identification of important Protocol Deviations. TransCelerate launched an initiative and developed tools to address industry-wide problems related to interpretation of regulatory guidance on protocol deviations. This video describes these tools in more detail and shows the viewer how to use them.
This manuscript, published in a peer-reviewed journal, highlights the contents of the Protocol Deviation toolkit. To access “Protocol Deviations: A Holistic Approach from Defining to Reporting” please click here.
The ICH E3 Q&A R1 defines a protocol deviation as any change, divergence, or departure from the study design or procedures defined in the protocol. The current definition is often over-interpreted leading to the inclusion of a wide scope of items being reported. For this reason, we recommend adding the following clarifying points:
- An event occurred—to avoid theoretical situations;
- The event is related to the protocol or documents referenced in the protocol (e.g., laboratory manual)
- The event is independent of fault, blame or circumstance—to ensure an objective approach to identification (e.g., sample tube broke in route to the central laboratory, participant refused a procedure).
The Protocol Deviation Process Map can help with process management.
Click image to enlarge.
A template to assist in the identification of documentation of protocol specific “important deviations.” To download the Excel PDAP template please click here.
This Decision Tree can guide in the identification and classification of important and non-important protocol deviations.
Click image to enlarge.
Taking a holistic approach to Protocol Deviations Management
With feedback from Sponsors, CROS, Sites and Health Authorities, the Protocol Deviations Toolkit has been designed to provide a holistic approach to PD management. The process map above outlines the end-to-end process flow including REPORT and TRAIN activities that occur throughout the process.
It is important to use all three input levels as part of the plan. To support consistency, it is recommended that Organizational and Intermediate level definitions of importance are used whenever possible, and protocol level definitions are only introduced where necessary. This will reduce variability in classification and categorization and help to support consistency in analysis and reporting.
The Protocol Deviations Assessment Plan (PDAP) can be created as a stand-alone document or incorporated into existing Integrated Quality and Risk Management Plans (IQRMPs). Whatever the form, we recommend that it is created in conjunction with protocol development and maintained as a living document until the last study data has been reviewed. As a living document, suitable version control measures should be utilized.
Protocol Deviation Process Map
Protocol Deviation Management
The current definition is often over-interpreted leading to the inclusion of a wide scope of items being reported. For this reason, we recommend adding the following clarifying points:
- An event occurred—to avoid theoretical situations;
- The event is related to the protocol or documents referenced in the protocol (e.g., laboratory manual)
- The event is independent of fault, blame or circumstance—to ensure an objective approach to identification (e.g., sample tube broke in route to a central laboratory, participant refused a procedure).
Protocol deviations can be identified via programmatic or manual processes. Manual identification approaches can vary but should focus on information not captured in an electronic data capture system
The application of risk-based approaches may not identify all protocol deviations. However, it should identify important protocol deviations.
This Decision Tree can guide in the identification and classification of important and non-important protocol deviations.
Decision Tree
Protocol deviation classification and categorization should be confirmed. It is recommended that a study team member (or group of study team members), other than the identifier, perform this review.
Feedback and possible retraining should be provided to the identifier if the protocol deviation is misclassified, miscategorized, or otherwise erroneously reported.
Based on the collection method (programmatic and/or manual) important protocol deviations should be reconciled and all discrepancies should be addressed. Examples of reconciliation include:
- Removal of duplicate protocol deviations (e.g., those that were identified via both programmatic and manual methods)
- Consistency between data point(s) and protocol deviation (e.g., data point in the clinical database contained a transcription error—once corrected no protocol deviation existed).
Study teams typically conduct periodic reviews, monitor clinical study data, and conduct analyses on an ongoing basis during the conduct of the clinical study. The department/functional area involved and approach to the data vary based on organizational structure. These activities are typically focused on safety and efficacy elements at the individual participant level data, across study sites, countries, and the protocol as a whole.
In this ANALYZE phase, an assessment should be made to determine the impact of the protocol deviation on the data sets and populations included in the SAP. Important protocol deviations may be one of several factors used to determine which participants are excluded from the “per protocol” analysis study population and may affect individual data points used in statistical analysis.
Each company should determine when to finalize their PD process (close out the living PDAP or other documents used). We recommend that finalization occur after the last study data has been reviewed.
Important protocol deviations are summarized and included in the CSR and archived. In addition to the final CSR, completing the following tasks is also a best practice:
- Archive non-important protocol deviations in a validated repository or system to support the review and reporting (e.g., Clinical Trial Management System [CTMS], Electronic Data Capture [EDC], Trial Master File [TMF] or a custom system) and in supporting PD data sets (e.g., SDTM).
- Conduct a “lessons learned” session.
- Share protocol level important protocol deviations for consideration at an intermediate or organizational level (Inputs Feedback Loop on Map).
- Complete any outstanding documentation or archival activity.
Click on each component of the PD Toolkit to learn more
PD Process Guide
A proposed framework describing flexible PD management approaches, elements for consideration based upon proposed interpretation of the ICH E3 definition for important protocol deviations and other associated PD Guidance with links to the PD Process Map
PD Map
A map of the PD management process containing processes for both important and non-important deviations
PD Assessment Plan
A template to assist in the identification and documentation of protocol-specific ‘important’ deviations
PD Decision Tree
This Decision Tree can guide in the identification and classification of important and non-important protocol deviations