Status
As of September 2024, the BioCelerate SEND (Standard for Exchange of Nonclinical Data) Implementation for Cross-Study Analysis initiative has successfully achieved all program objectives. The initiative produced a suite of open-source tools, guidance documents and research publications that are freely available for voluntary use by CROs, regulators and sponsors. With all planned deliverables completed, the initiative has successfully laid the groundwork for improved implementation and analysis of CDISC (Clinical Data Interchange Standards Consortium) SEND data packages.
Rationale
The introduction of CDISC SEND data packages enabled a common format for presenting nonclinical study data. With the requirement that all FDA preclinical submissions utilize the SEND format, adoption of the standard was increasing. However, certain implementation inconsistencies were limiting SEND’s potential value in enabling broader single- and cross-study analysis.
Through a public-private partnership with FDA’s Center for Drug Evaluation and Research (CDER), the team set out to examine the existing challenges and develop recommendations and tools to enable progress.
By identifying differences in standards use that negatively impacted the ability to conduct cross-study analysis, the initiative aimed to expand access to early-stage R&D knowledge, enable faster decision-making, and help organizations streamline processes while reducing redundant work—ultimately accelerating the delivery of new treatments to patients.
Process
The initiative pursued several parallel workstreams to address the challenges of SEND implementation with respect to cross-study analysis:
- The team first conducted research to identify and document the difficulties with analyzing cross-study data, resulting in published findings that highlighted key areas for improvement.
- Working closely with PHUSE (Pharmaceutical Users Software Exchange), a dedicated sub-team developed the sendigR package — an open-source R toolbox that enables users to construct relational databases from SEND datasets and perform cross-study analyses.
- The initiative collaborated with industry stakeholders to develop and implement recommendations for SEND data harmonization, including a proposal for populating Control Type (TCNTRL) with CDISC SEND Controlled Terminology.
- Through collaboration with the FDA, the team explored and documented data harmonization strategies for potential use in comparing toxicity profiles between multiple SEND datasets.
Solutions
Publications and White Papers
The initiative produced an array of deliverables, including the following:
- SEND harmonization & cross-study analysis: A proposal to better harvest the value from SEND data (Regulatory Toxicology and Pharmacology, 2020): A manuscript highlighting the difficulties with analyzing cross-study data
- Leveraging the Value of CDISC SEND Data Sets for Cross-Study Analysis: Incidence of Microscopic Findings in Control Animals (Chemical Research in Toxicology, 2020): A proof of concept manuscript for cross-study analysis co-authored in collaboration with the FDA
- Recommendations for Populating Control Type (TCNTRL) with CDISC SEND Controlled Terminology (2021): A white paper developed in collaboration with PHUSE, providing recommendations for populating Control Type with CDISC SEND Controlled Terminology
- Cross-study analyses of SEND data: toxicity profile classification (Toxicological Sciences, 2024): A collaborative publication with the FDA on data harmonization strategies for comparing toxicity profiles between SEND datasets
Tools and Resources
- sendigR Package: An open-source R package enabling users to construct relational databases from SEND datasets and perform cross-study analyses. The package includes:
- An integrated Python package (xptcleaner) for harmonizing SEND dataset terminology
- An R Shiny web application for toxicologists to perform historical control analyses
- Complete reference manual
- Demo application
- Accompanying manuscript (Frontiers in Toxicology, 2024) explaining the value and application of these tools
Impact
Organizations implementing these solutions have already reported significant operational improvements. With consistent SEND study data packages, organizations can:
- More easily compare and analyze data across multiple studies, leading to more informed decision-making
- Generate more efficient data delivery processes, resulting in reduced cycle times for both CROs and sponsors
- Reduce resources needed for maintaining internal SEND standards
- Create higher quality datasets with fewer errors, enabling more efficient regulatory review
For a practical example of implementation, see “Utilizing sendigR to Increase the Amount of Data and Confidence in a Cross Study Vehicle Effect Analysis by Using Data from Other Sites.”
The SEND Harmonization work has laid the foundation for the team to transition efforts to other initiatives, including Enabling Translational Safety.
Feedback and Impact
Have any feedback on these resources or experiences to share regarding implementation? Please use our feedback form.
All BioCelerate and TransCelerate initiatives are geared towards providing practical solutions to reduce inefficiencies in clinical trials, nonclinical studies, and drug safety. Learn about other BioCelerate and TransCelerate initiatives and solutions.
Disclaimer
These materials are provided ‘AS IS’ WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, OR NON-INFRINGEMENT. BioCelerate and its members do not accept any responsibility for any loss of any kind including loss of revenue, business, anticipated savings or profits, loss of goodwill or data, or for any indirect consequential loss whatsoever to any person using these materials or acting or refraining from action as a result of the information contained in these materials. Any party using these materials bears sole and complete responsibility for ensuring that the materials, whether modified or not, are suitable for the particular use and are accurate, current, commercially reasonable under the circumstances, and comply with all applicable laws and regulations.
Nothing in these documents should be construed to represent or warrant that persons using these documents have complied with all applicable laws and regulations. All individuals and organizations using these documents bear responsibility for complying with the applicable laws and regulations for the relevant jurisdiction.