CDISC Standards: A Complete Guide for Clinical Trial Sponsors
CDISC (Clinical Data Interchange Standards Consortium) standards are the global framework for structuring clinical trial data for regulatory submissions. Understanding how SDTM, ADaM, CDASH, and SEND work together is essential for any sponsor preparing an FDA or EMA submission.
What Are CDISC Standards?
CDISC is a non-profit organisation that develops global, vendor-neutral data standards for clinical research. The FDA has required CDISC-formatted data for new drug and biologics applications since 2016 (NDA/BLA) and 2017 (IND). The EMA and other regulatory agencies increasingly expect CDISC-compliant submissions. The four primary CDISC standards are CDASH (data collection), SDTM (study data tabulation), ADaM (analysis datasets), and SEND (non-clinical data).
CDISC SDTM: Study Data Tabulation Model
SDTM defines how clinical trial data should be organised into standardised domains for regulatory submission. Each domain represents a category of data — demographics (DM), adverse events (AE), laboratory results (LB), vital signs (VS), and so on. SDTM datasets are accompanied by a define.xml (a machine-readable data dictionary) and a reviewer's guide that explains the study-specific implementation. Correct SDTM mapping is critical: FDA reviewers use SDTM datasets directly in their analysis tools, and mapping errors can trigger review queries or delay approval.
CDISC ADaM: Analysis Data Model
ADaM datasets are derived from SDTM and are designed to support statistical analysis. The key ADaM datasets are ADSL (subject-level analysis dataset), ADAE (adverse event analysis dataset), and ADTTE (time-to-event analysis dataset). ADaM datasets must be traceable back to SDTM — every derived variable must have a documented derivation. The ADaM define.xml and reviewer's guide accompany the datasets in the submission package.
CDISC CDASH: Data Collection Standards
CDASH defines how data should be collected on eCRFs. Collecting data in CDASH format from the outset simplifies SDTM mapping and reduces the risk of transformation errors. CDASH provides standard question text, field names, and controlled terminology for common data elements. Sponsors who implement CDASH at study start-up typically see faster database lock and fewer SDTM mapping issues downstream.
Controlled Terminology and Codelists
CDISC controlled terminology (CT) defines the permitted values for coded fields in SDTM and ADaM. Using the correct CT version — and documenting which version was used — is a common source of submission queries. The CDISC CT is updated quarterly; sponsors should specify the CT version in their define.xml and ensure consistency across all datasets.
FDA Submission Validation Rules
The FDA publishes Pinnacle 21 validation rules that are applied to CDISC datasets during submission review. Errors and warnings from Pinnacle 21 must be addressed or explained in the reviewer's guide. A specialist CRO with CDISC expertise will run Pinnacle 21 validation as part of the submission preparation process, resolving issues before the package is submitted to the agency.
Key Checklist
Frequently Asked Questions
What is CDISC and why is it required for FDA submissions?
CDISC (Clinical Data Interchange Standards Consortium) develops global standards for clinical trial data. The FDA has required CDISC-formatted data (SDTM and ADaM) for NDA/BLA submissions since 2016 and for IND submissions since 2017. CDISC standards enable FDA reviewers to use standardised analysis tools and compare data across submissions.
What is the difference between SDTM and ADaM?
SDTM (Study Data Tabulation Model) organises raw clinical trial data into standardised domains for regulatory submission. ADaM (Analysis Data Model) contains derived datasets designed to support statistical analysis. SDTM is the source; ADaM is derived from SDTM and must be fully traceable back to it.
What is a define.xml in a CDISC submission?
A define.xml is a machine-readable data dictionary that accompanies SDTM and ADaM datasets in a regulatory submission. It describes every dataset, variable, codelist, and value-level metadata in the submission package. FDA reviewers use define.xml to navigate and validate the submission data.
How long does CDISC conversion take for a clinical trial?
CDISC conversion timelines depend on the complexity of the study, the number of domains, and the quality of the source data. A straightforward Phase II study might take 8–12 weeks for SDTM and ADaM conversion; a complex Phase III programme with multiple studies could take 6–12 months. Starting CDISC planning early — ideally at protocol design — significantly reduces conversion time.
What is Pinnacle 21 and how is it used in CDISC submissions?
Pinnacle 21 (now Certara's P21 Community and Enterprise tools) is the industry-standard software for validating CDISC datasets against FDA and PMDA conformance rules. It generates an issues report listing errors, warnings, and notices. Sponsors must address or explain all errors and critical warnings before submitting to the FDA.
Need specialist support?
IDDCR Global Research provides specialist CRO services for pharmaceutical, biotechnology, and medical device sponsors. Contact us to discuss your requirements.