ACT Oncology Clinical Data Review provides rigor and quality assurance to the data clean-up process to enhance the predictability of finalization of end data and “lock” of the database for filing.
Initial and ongoing review of eCRF data is conducted by an experienced team of cross-trained clinical and data management experts in concert with other core team members (medical monitoring, biostatistics and project management personnel). Quality issues that may not be readily apparent to data management or monitoring team members are identified early and corrective actions are able to be implemented without delay. Clinical Data Review ensures adherence to the protocol (including toxicity grading and response evaluation criteria) and provides early detection or correction of site and/or CRA training and compliance issues. Such issues are not left until the end of the study which can result in analysis and overall timeline delays.
Effective use of Clinical Data Review enables data clean-up to be done on an ongoing basis by looking at the patient data across time and ensuring that the final data makes clinical sense. We find this is especially helpful in providing appropriate insights and corrections in the following areas:
- Safety data: the protocol indicates that an adverse event should be recorded at its worst severity. Given the challenge of ready views across forms, we find that there are often multiple listings of the same AE term with overlapping dates and various severities which require correction. Additionally, AE terms and grading need to be consistent with the CTCAE criteria and the Clinical Data Review process ensures that the appropriate grades are applied based on co-existing symptoms (when possible).
- Interim response: Each patient’s clinical response data over time is reviewed to ensure the evolution and/or progress of the disease makes clinical sense per documented guidelines to assess response
- Cross-patient trends: the Clinical Data Review process is limited to a focused team allowing for data trend identification that can be discussed with the statistician (to determine potential impact on end analysis) and with the project team (for retraining of the CRA), if appropriate