<img alt="" src="https://secure.lote1otto.com/219869.png" style="display:none;">
Skip to content

Raising Awareness for Additional FDA Data Standards Submission Recommendations (Part II)


In the first part of this article, I raised awareness of the availability of additional FDA guidances containing CDISC implementation examples for specific therapeutic areas or indications. Let’s now take a closer look at the key content of those additional guidances.

As mentioned in Part I, all these guidances are referenced in the “List of FDA Technical Specification Documents” section of the FDA Study Data Technical Conformance Guide1 and in the “Technical Guides” section of the Study Data Standards Resources FDA webpage. Their CDISC-related key content is summarized in the table below, and in particular:

  • FDA divisions concerned: Center for Drug Evaluation and Research (CDER) vs Center for Biologics Evaluation and Research (CBER)
  • CDISC standards covered
  • Study Data Tabulation Model (SDTM) vs. Analysis Data Model (ADaM)

  FDA Additional Data Standards Guidance References Content

Submitting Study Datasets for Vaccines to the Office of Vaccines Research and Review

v2.1 (December 2019)

  • CBER and Office of Vaccines Research and Review
  • SDTM
  • CRF design and mapping recommendations for reactogenicity data (flat vs. nested model)
  • Vaccines CDISC TAUG

Submitting Select Clinical Trial Data Sets for Drugs Intended to Treat Human Immunodeficiency Virus-1 Infection

v1.0 (March 2018)

  • CDER
  • ADaM, e.g., ADAE (non-OCCDS), ADLB (list of main laboratory parameters of interest), ADEFFOUT (ADSL-like focusing on efficacy endpoint)

Technical Specifications–Comparative Clinical Endpoint Bioequivalence Study Analysis Datasets for Abbreviated New Drug Applications

v1.0 (September 2018)
  • CDER
  • ADaM, e.g., ADSL and other ADaM specific for each type of bioequivalence study

Submitting Clinical Trial Datasets for Evaluation of QT/QTc Interval Prolongation and Proarrhythmic Potential of Drugs

v1.0 (June 2019)

  • CDER
  • ADaM, e.g., ADSL, ADEG, ADPC

Technical Specifications for Submitting Clinical Trial Data Sets for Treatment of Noncirrhotic Nonalcoholic Steatohepatitis (NASH)

v1.1 (January 2022)2
  • CDER
  • SDTM (14 domains) events adjudication recommendations (ZA)
  • ADaM (7 domains)

The Submitting Study Datasets for Vaccines to the Office of Vaccines Research and Review technical specification document is a CBER-OVRR3 guidance that provides detailed specifications for some SDTM topics. As an example, the way reactogenicity data4 are reported, that is, the “flat” vs. “nested” model (see also the CDISC Vaccine TAUG), with the agency preferring the flat model (it requires the daily assessment of a set of pre-specified adverse events). The guidance also provides other detailed implementation preferences, such as preferred terms to be used for the -CAT variable (CECAT= “reactogenicity” vs. CECAT=” to differentiate reactogenicity event vs. efficacy events).

SDTM Implementation

Overall, all the specifications are aligned with the SDTM implementation guidance, with the agency also making clear, for example, that immunogenicity data should be mapped to the Immunogenicity Specimen (IS) domain and not to Laboratory Test Results (LB) domain.

The guidance has been referenced by the CBER reviewer in a couple of submissions we did at Cytel when the agency provided feedback on the Study Data Standardization Plan (SDSP)5 we submitted as part of a briefing book;6 despite the CBER biostatistician’s request posing some challenges giving the fact some of the studies were either conducted by other vendors or closed before both FDA and the CDISC TAUGs were released, we were able to accomplish most of the request. Clearly, this is a guidance everyone working on vaccines should not ignore!

ADaM Recommendations

The guidance does not contain any ADaM recommendations, nor does the related CDISC Vaccine TAUG guidance. This is the opposite of many of the other FDA guidances where a lot of the content is related to ADaM; for most of these guidances, there is currently a CDISC ADaM sub-team reviewing and, I would say, validating the content. In fact, I also noted a lack of peer review in the guidance, with a few inconsistencies within and across guidance for related topics. Moreover, it would have been good if some examples were included in the guidance as most are kind of derivation specifications, listing variables for each suggested ADaM dataset; a bit of rationale and type of analyses to be supported by each ADaM dataset could have helped in better understanding some derivations.

This is, for example, the case for the Comparative Clinical Endpoint Bioequivalence Study Analysis Datasets guidance. The guidance provides only detailed ADaM specifications and general considerations on how certain comparative clinical endpoint bioequivalence study data should be submitted, as well as data on skin adhesions and irritation/sensitization, for example: total nasal symptom score, lesion count, treatment success based on Physician’s Global Assessment, Psoriasis Area Severity Index, Intraocular Pressure, and so on.

Despite the guidance not referring to any specific version of the ADaM standard, a better alignment with the standard ADaM variables and best practices is recommended. Again, this is the case for the Comparative Clinical Endpoint Bioequivalence Study Analysis Datasets guidance and the way windowing method are applied on one of the described ADaM datasets (ADIP).

In Submitting Select Clinical Trial Data Sets for Drugs Intended to Treat Human Immunodeficiency Virus-1 Infection, in addition to providing detailed ADaM specifications, the CDER division is also emphasizing that “the specifications should be seen by applicants as an opportunity to discuss with the reviewers issues with trial design or study conduct that may affect the content of the ADaM datasets discussed in the guidance,” a statement that is thereafter re-iterated in other guidances such as the NASH one. The HIV guidance describes three ADaM datasets: ADEFFOUT, where many variables potentially affecting the efficacy are described as well as the variable for efficacy measures of viral load; ADLB, where primary laboratory parameters of interest are listed; and ADAE, where the ADaM ADAE document is used as a reference rather than the more recent Occurrence Data Structure (OCCDS) ADaM Class.

NASH Guidance

The Noncirrhotic Nonalcoholic Steatohepatitis (NASH) guidance is the last released guidance (January 2022) and among all is the one with a satisfactory level of details and explanations. The guidance describes requirements for seven ADaM datasets, which include the “Drug Induced Liver Injury” ADDILI ADaM dataset, the “Non-Invasive Serum Biomarkers of Liver Fibrosis and NASJ” ADRS ADaM dataset, and some recommendations for fourteen SDTM datasets. The SDTM part also includes some recommendations on terminology to be used for certain variables and expected results, such as for TESTCD in Biospecimen Findings (BS) and Microscopic Findings (MI) domains, as well as recommended naming conventions for certain non-standard variables to be mapped in the supplemental qualifier dataset.

In the NASH guidance, like for IS cited above for the vaccine guidance, the MB is correctly recommended for mapping hepatitis serology parameters, such as Hepatitis A Virus Antibody, and not in LB as I can still see in some SDTM mapping I regularly review.

The use of RELREC is also recommended, for example, to connect AE with CM, a practice not consistently applied by SDTM “mappers.”

Furthermore, the guidance proposes some recommendations on how to map some adjudication data. With this regard, as of today, there is not yet a standard recommended approach on how to handle adjudicated events; there has been a proposal from PHUSE, but this is not yet acknowledged by CDISC even in the recent SDTM IG 3.4, where instead a basic example of adverse event data adjudications is made using some specific terms in the supplemental qualifier’s dataset. The NASH guidance follows a similar approach recommended by PHUSE and proposes a finding about datasets (Adjudication - ZA) domain to store the information about the adjudication of certain events such as drug-induced liver injury, liver-related death, hepatic decompression events, major adverse cardiac events, and cardiac-related death.

ADaM specifications are also well documented, in particular for ADSL, with the agency listings and providing expected derivations for certain key variables they wanted to see in the dataset, while for ADAE, the guidance includes the list of some key adverse events of interest they wanted to see clearly identified (flagged).

For ADLB, the agency recommends dual reporting for Standard results, using both SI units and U.S. conventional units when the two systems differ, this is to minimize conversion needs during review. While waiting for an official recommendation from the CDISC teams, e.g., SDTM, the dual unit reporting could be managed in ADaM. It is also recommended that the identification of units to be used and reported is discussed as early as possible in the process, such as during the submission of the SDSP. This is something we do now by default in the SDSP we create.

The guidance also lists some examples of time-to-event endpoints of interest for both safety and efficacy.

The Submitting Clinical Trial Datasets for Evaluation of QT/QTc Interval Prolongation and Proarrhythmic Potential of Drugs guidance describes three ADaM datasets: ADEG, ADPC, and ADSL. One aspect the agency emphasizes in this guidance is the consistent use of terminology for certain variables such as AVISIT and ATPT (and others) across ADaM datasets, e.g., between ADPC and ADEG.

In the first part of this article, I cited three other FDA guidances, though their content is not discussed here as they are not fully CDISC related (or not yet):

  • Data Standards for Drug and Biological Product Submissions Containing Real-World Data
  • Submitting Next Generation Sequencing Data to the Division of Antiviral Products
  • Submitting Nonclinical Datasets for Evaluation of Rodent Carcinogenicity Studies of Pharmaceuticals

In conclusion, as discussed throughout the two parts of this article, despite not having the force and effect of law as re-iterated by the FDA in these guidances, it is good for sponsors and vendors to periodically monitor and carefully review these guidances, first to raise the awareness in their organization, but also, most important, to assess the impact on existing data governance.



1 Center for Drug Evaluation and Research and Center for Biologics Evaluation and Research, Study Data Technical Conformance Guide (Washington, D.C.: U.S. Food and Drug Administration, 2022), section 5.

2 Cécile Cornou and Henning Pontoppidan, “Impact of FDA Technical Specifications on CDISC Implementation for NASH Trials,” presented at CDISC Europe Interchange, 2022.

3 Office of Vaccines Research and Review.

4 Specific expected or common reaction following vaccine administration.

5 PHUSE, “Study Data Standardization Plan (SDSP) Package,” last modified September 13, 2022.

6 Documentation submitted to facilitate an Agency meeting, for example, to request scientific advice in relation to a proposed drug development. It should contain company questions, justifications or company positions, and appropriate summary information.


About Angelo Tinazzi


Angelo Tinazzi is Senior Director, Statistical Programming, Clinical Data Standards and Clinical Data Submission at Cytel. He is a well-published and recognized expert in statistical programming with over 20 years' experience in clinical research. The application of CDISC standards in different therapeutic areas is part of his core expertise since 2003 in particular in the context of data submission to health authorities such as the FDA and PMDA.

Angelo is an authorized CDISC instructor and member of the CDISC ADaM Team as well as the CDISC European Committee where he also manages the Italian-speaking CDISC User Network.


What's new on the blog:

New call-to-action

New call-to-action

New call-to-action

contact iconSubscribe back to top