Efficient Management of Integrated Summary of Safety Packages for Oncology Medications with Multiple Ongoing Submissions

Photo by CDC on Unsplash

Introduction

The Integrated Summary of Safety (ISS) is a critical component of any drug application submission to regulatory authorities, providing a comprehensive summary of the safety profile of a new drug or biologic. In the case of oncology medications submitted for multiple tumor types or indications, the ISS process can become more complex due to the need to harmonize data from multiple trials and ongoing submissions. This article discusses ways to address the unique challenges for ISS packages for such compounds and suggests processes and techniques to efficiently build and maintain a reusable library of reference safety datasets (safety stack) that reflect the compound’s safety profile. Additionally, it covers methods to harmonize datasets to support integrated analysis with multiple data sources, ensuring harmonized variable attributes across multiple trials, compliance of integrated ADaM datasets, alignment with pivotal trial Clinical Study Report, and common dictionaries.

Table of Contents

1. Reference Safety Dataset Library Overview
2. High-Level Process for Reference Safety Dataset (RSD) Library Creation and Maintenance
3. Reference Safety Dataset Library Uses
4. Analysis Populations
5. Best Practices for ISS Datasets
6. Reference Safety Dataset Library Templates and Standards
7. Integrated Analysis Standards and Templates
8. Programming Important Points and Checks
9. MedDRA Leveling and Dictionary Conversion
10. Unique Challenges and Solutions
11. QC of Outputs
12. Conclusion
13. Reference Safety Dataset Library Overview

A reference safety dataset (RSD) library is an essential tool for the efficient management of ISS packages for compounds with multiple ongoing submissions. This library consists of submission-ready ADaM datasets for each individual trial, providing a consistent and reusable source of safety data that can be updated as needed to support various regulatory requirements.

High-Level Process for Reference Safety Dataset (RSD) Library Creation and Maintenance

To create and maintain an RSD library, the following high-level process is recommended:

• Define the RSD population based on the submission and approval status of each clinical trial, as determined by regulatory authorities.
• Create and validate submission-ready ADaM datasets for each individual trial by a central ISS team following programming standard operating procedures (SOP).
• Store RSD datasets in a central area, organized into sub-folders by MedDRA version for efficiency and consistency.
• Update the RSD library as needed to support additional trials, ADaM standard updates, and new dictionary versions.

Reference Safety Dataset Library Uses

The RSD library serves as a foundation for efficient ISS package preparation, providing:

• A reusable source of safety data that can be easily updated and customized for individual submissions.
• A consistent and harmonized set of safety data for integrated analyses.
• A standardized format for safety data that simplifies the process of comparing safety profiles across multiple trials and indications.

Analysis Populations

Analysis populations are crucial for ensuring accurate and meaningful safety analyses. The following populations should be considered for ISS datasets:

• Safety Population: Includes all patients who received at least one dose of the study drug and had at least one safety assessment.
• Efficacy Population: Includes all patients who met the study’s primary efficacy endpoint and received the study drug according to the protocol.

Best Practices for ISS Datasets

To ensure the quality and consistency of ISS datasets, the following best practices are recommended:

• Standardize variable names, labels, and formats across all datasets.
• Use consistent data conventions and coding across trials.
• Validate datasets using standard validation checks and custom checks specific to the study drug and indication.

Reference Safety Dataset Library Templates and Standards

To facilitate the creation and maintenance of RSD libraries, standard templates and guidelines should be developed, including:

• Dataset specifications and metadata
• Dataset specifications and metadata documentation, detailing variable names, formats, and labels.
• Standard coding and conventions for adverse events, laboratory results, and other safety data.
• Guidelines for handling missing or incomplete data.
• Procedures for data conversion and dictionary updates.

Integrated Analysis Standards and Templates

To support integrated analyses of safety data from multiple trials, it is essential to develop and maintain standards and templates that ensure consistency and harmonization, such as:

• Integrated analysis dataset specifications, outlining the structure and content of the combined safety data.
• Standardized analysis methods, including pooled analyses, meta-analyses, and subgroup analyses.
• Templates for tables, listings, and figures to facilitate clear and concise presentation of safety results.

Programming Important Points and Checks

Effective programming of ISS datasets requires attention to several key aspects, including:

• Ensuring compliance with regulatory requirements and ADaM standards.
• Implementing robust data validation procedures, including standard checks and study-specific checks.
• Developing efficient and modular programming code that can be easily updated and reused for different trials and indications.

MedDRA Leveling and Dictionary Conversion

MedDRA leveling and dictionary conversion are essential for harmonizing safety data across multiple trials, ensuring consistent reporting of adverse events, and facilitating integrated analyses. Key considerations include:

• Using the most current MedDRA version for all trials and updating the RSD library as needed to accommodate new versions.
• Implementing standardized procedures for converting data from other dictionaries, such as WHO-Drug, to MedDRA.
• Ensuring consistent application of MedDRA terms and levels (e.g., System Organ Class, High-Level Group Term, Preferred Term) across all datasets.

Unique Challenges and Solutions

Several unique challenges arise when managing ISS packages for compounds with multiple ongoing submissions, including:

• Ensuring consistent data quality and harmonization across multiple trials and data sources.
• Efficiently updating and maintaining the RSD library to support new trials, indications, and regulatory requirements.
• Managing complex programming requirements and validation procedures.

To address these challenges, it is essential to establish robust processes, standards, and best practices, as discussed in previous sections, and to maintain a strong collaboration between the ISS team, clinical trial teams, and regulatory authorities.

QC of Outputs

Quality control (QC) of ISS outputs, including tables, listings, and figures, is critical for ensuring accurate and reliable safety results. Effective QC procedures should include:

• Independent programming and review of all outputs.
• Cross-checking of outputs against source data and analysis datasets.
• Verification of outputs against regulatory requirements and internal standards.

Conclusion

In conclusion, managing ISS packages for oncology medications with multiple ongoing submissions presents unique challenges, requiring a systematic approach to data harmonization, analysis, and reporting. By developing and maintaining a reusable library of reference safety datasets, implementing best practices and standards for integrated analyses, and ensuring robust programming and QC procedures, it is possible to efficiently and effectively support the safety evaluation of these complex compounds throughout the drug development process.