(M-006) Risk Based Validation Strategy for Pharmacometric Analysis Ready Dataset to improve efficiency and effectiveness

Kiran Kumar Kode – Clinical Pharmacology and Quantitative Pharmacology – Astrazeneca plc; Dheeraj Rupani, MS – Clinical Pharmacology and Quantitative Pharmacology – Astrazeneca plc

Objectives: Optimize risk scoring of pharmacometric analysis ready dataset validation process by using “Quality Risk Matrix” to boost efficiency and cut time and costs. This strategy streamlines the validation process while preserving data quality and integrity, ultimately benefiting the organization through more efficient and cost-effective validation procedures.

Methods: Risk scoring for an activity relies on three key factors: Severity, Likelihood, and Detection. To effectively lower the risk score, it's essential to enhance control measures that decrease the likelihood of negative events and improve detection capabilities. This approach not only minimizes potential risks but also strengthens the overall safety and reliability of the process. We utilize following methods to minimize risk likelihood and enhance detection, ensuring effective risk management strategies and improved safety outcomes.

DaRT (Data Request Tool): The data request tool enforces adherence to data standards and SAS programming templates, minimizing risk likelihood by improving consistency, accuracy, and reliability across data management processes, and promoting standardized practices throughout operations.
Data Standards and Standards around deriving the covariates:
Variable Mapping Specifications (VMS): This process engages study programmers from the beginning, ensuring alignment of derivations with the study team. It reduces the likelihood of discrepancies and facilitates easy comparison with study reports, aiding in the early detection of potential issues for improved accuracy.
Standard Programming Templates and Standard macros: Standard programming templates and macros play a crucial role in streamlining programming practices. They promote a uniform approach across projects, enhancing consistency, efficiency, and quality.
Data Visualizations: Data visualizations enable quick identification of inconsistencies, providing a clear view of data anomalies and facilitating prompt resolution for enhanced accuracy and data integrity.
Other important methods are Process, Trainings, Communication and Collaboration

Results: We achieved a robust risk management system that minimizes risks in data management and programming. The Data Request Tool (DaRT) and adherence to data standards improved consistency and accuracy. Variable Mapping Specifications ensured alignment with team derivations, aiding in early discrepancy detection. Standard programming templates and macros enhanced uniformity and efficiency across projects. Data visualizations quickly identified inconsistencies, ensuring greater accuracy. Supported by strong processes, continual training, and effective communication, these methods significantly improved safety and reliability. We are working on upgrading DaRT, customizing SAS programming templates and data visualizations for different analysis types to further enhance the effectiveness of pharmacometrics analysis ready datasets.

Conclusions: The strategies implemented have strengthened our validation strategy, enhancing safety, reliability, and accuracy in data processes and ensuring consistent, efficient, and dependable outcomes.

Citations: Risk-based approach for method development in pharmaceutical quality control context: A critical review - Riccardo Deidda, Serena Orlandini , Philippe Hubert , Cédric Hubert
Journal of Pharmaceutical and Biomedical Analysis
Volume 161, 30 November 2018, Pages 110-121

Keywords: Risk Based Validation, Validation Strategy and Techniques, Process improvement