Controlled Terminology Best Practices

March 3, 2021

Controlled Terminology have you confused? We got you.

When starting a new clinical study and preparing the data collection design, you may ask yourself: How important is Controlled Terminology (CT)? How do I prepare the data collection process to be CDISC compliant and avoid rework for submissions? We often receive questions such as these. Whether you are newer to clinical studies or a seasoned veteran, CT can often be an elusive and confusing topic. Having a better understanding of CT standards and requirements, as well as CT-related issues found in validation checks, can improve your submission data and processes.

Pinnacle 21’s Sarah Angelo recently hosted a webinar to provide some insight into and best practices for Controlled Terminology. She received almost two hundred questions, so this blog post serves as a follow-up with the key responses. For the full webinar, you can watch the video recording above and download the slide deck.

Lab Synonyms

Studies that obtain results for multiple types of lab tests have not only numerous records, but also an extensive list of units associated with the results. So, one of the most popular questions asked at the webinar was in regard to changing a Lab unit synonym value to display the CDISC submission value. As stated in the SDTMIG, the variable LBORRESU uses the UNIT codelist. This means that sponsors should submit a term from the column "CDISC Submission Value" in the published Controlled Terminology List maintained for CDISC by NCI EVS.

If sponsors have units that are not in the CDISC Submission Value column, check if the provided unit is mathematically synonymous with an existing unit and submit lab values using that unit. For example: "g/L" and "mg/mL" are mathematically synonymous. However, only "g/L" is in the CDISC UNIT codelist. If this is not the case, and no synonymous unit applies, then submit a New-Term Request Form.

To sum it up:

  • CDISC Submission values should be displayed rather than the synonym.
  • The synonym does not need to be stored in the SUPPQUAL dataset.
  • Update TLF shells or make a note if they display a synonym.
  • There is no impact on the Result variable. --ORRES does not change.
  • If Lab vendors supply raw data using synonyms, update --ORRESU to use the correct CDISC submission value. Best practice is to update Data Transfer Agreements at the start of the study, to confirm the correct unit values are being collected.

Controlled Terminology: Regulatory Agency Requirements

Although CT requirements have become more universal, the FDA and PDMA have separate Controlled Terminology requirements listed in their respective Technical Conformance Guides (TCG). It’s important to recognize the differences not only between FDA and PMDA, but also among the new, ongoing, and pooled studies. Below is a quick rundown of the guidance from each agency.

New Studies

  • FDA Guidelines: Section 6.1.2 Use of Controlled Terminologies, “FDA recognizes that studies are conducted over many years, during which time versions of a terminology may change. Sponsors should use the most recent version of the dictionary available at the start of a clinical or nonclinical study.”
  • PMDA Guidelines: “When creating the dataset and the definition document conforming to the CDISC standards, refer to the PMDA’s website (http://www.pmda.go.jp/) for versions of the CDISC standards, controlled terminologies, and dictionaries that are accepted by the PMDA. Acceptance of versions is judged based on the date of submission described in the new drug application by the applicant.”

Ongoing Studies

  • FDA Guidelines: “If a new version becomes available after the start of the study, sponsors may use the most current version of the dictionary for that clinical or nonclinical study. It is common to have different studies use different versions of the same dictionary within the same application (e.g., NDA, BLA). A submission of study data should describe (e.g., in the SDSP or relevant RG) the impact, if any, of the use of different versions on the study results."
  • PMDA Guidelines: Follow the guidance in the Technical Conformance Guide, which states the same version of CT be used within the same clinical study.

Pooled Studies

  • FDA Guidelines: The TCG states, "Regardless of the specific versions used for individual studies, pooled analyses (e.g., for an ISS) should be conducted using a single version of a terminology. The current version should be used at the time that data across studies are pooled. This will ensure a consistent and coherent comparison of clinical and scientific concepts across multiple studies. Sponsors should specify the terminologies and versions used in the study in the relevant RG."
  • PMDA Guidelines: The TCG states, “Datasets of integrated analyses of multiple clinical studies should be created using the same version, even if the version used to create the dataset of each clinical study was different. When standardizing the version, the reason that the specific version was chosen and considerations for converting to a different version should be described in the data guide.”

Links to TCGs

FDA: https://www.fda.gov/media/136460/download

PMDA: https://www.pmda.go.jp/files/000229468.pdfhttps://www.pmda.go.jp/files/000229468.pdf

Extensible Codelists

The FDA TCG discourages the creation of custom terms (i.e., so-called extensible code lists) for a submission, because this does not support semantically interoperable study data exchange. Furthermore, the use of custom or extensible code lists does not imply that sponsors may substitute their own nonstandard terms in place of existing equivalent standardized terms.

For example: If "Informed Consent Withdrawn" is collected on the CRF as a disposition event, which is associated with Extensible codelist (NCOMPLT), it is recommended to map to the corresponding CDISC term "WITHDRAWAL BY SUBJECT".

CT2002 – Fix Tips

  • Verify value in raw data
  • Check CDISC CT for possible submission values
  • Ensure synonyms are not used
  • Ensure consistency in your study and submission
  • Avoid free text in data collection
  • Submit your extended terms to CDISC (as early as possible)
  • Document in the Reviewer’s Guide the reason why the additional term was needed

If we missed any burning questions or topics, check out the full webinar and slide deck to see if we covered it. If you still have other questions, please feel free to contact us!