Study Data Exchange Standards R&D
Over the past few years, CDER has increased its support for standardized study data submissions using CDISC standards, and will continue to do so in the future.
- Currently, CDER supports the submission of CDISC content using SAS Transport File version 5.
- As adoption of CDISC standards increases in CDER and throughout the regulated industry, we recognize the inherent limitations of using SAS transport files for data exchange. These include:
- 8-character limitation for variable names
- 40-character limitation for labels for variables
- 200-character limitation for character fields
- Flat, two-dimensional file structure for what we now recognize is multi-relational, multi-dimensional data
- Within our overall drug review modernization efforts, CDER recognizes the need to identify a pathway for moving away from SAS Transport to a more modern, robust, and flexible transport mechanism for CDISC content and certain other study data currently submitted in PDF format.
- Ongoing work on the CDISC-HL7 Study Data Standards project indicates that HL7 version 3 xml may be a promising solution for the exchange of study data. The potential benefits include:
- No variable or data field size limitations
- More multi-relational representation of the data
- Ability to transport more content (e.g. unstructured text such as patient narrative information)
- Improved harmonization with other HL7 v.3 data standards for healthcare and regulatory information (e.g. Individual Case Safety Report, Regulated Product Submission, Clinical Trials Registry and Results, Clinical Document Architecture)
- Improved interoperability with healthcare and other regulatory information systems
- In parallel with our ongoing robust CDISC implementation efforts, CDER intends to conduct a Proof of Concept (PoC) and test in a measured, iterative, and incremental fashion the use of HL7 v.3 xml as a possible exchange method for certain use cases. Testing will proceed incrementally for the following use cases:
- Submission of Patient Narrative and Clinical Investigator Qualification Information (e.g. FDA Form 1572)
- Submission of Structured Protocol Information
- Submission of Subject Data
- Testing will include an opportunity for interested stakeholders to participate in an open and transparent manner.
- Careful evaluation of the test and an assessment of the costs, benefits and impacts of potential migration to a new exchange method—on both FDA and regulated industry-- will be conducted to inform possible next steps. A determination of whether and how to proceed will be based on that thorough evaluation of the pilots and proposals for further development or adoption.
- CDER is committed to working throughout the standards development and implementation processes with stakeholders to develop appropriate data standards adoption, testing and implementation plans with phase-in periods and to maintain clear communication of data submission expectations between CDER and sponsors.
- Per the FDA’s data standards proposal for PDUFA V, to be included in the draft recommendations for enhancements under PDUFA V, data standards development work, including any further pilot testing will be included in the agency data standards plan, and any specific adoption plans will be first included in a proposed guidance to industry for public comment, and later be articulated in a final guidance that includes a lag period of at least two years before requirement of the new standard, during which time CDER will accept the new exchange format as well as other file formats currently in use for the same information.
- CDER remains committed to working with CDISC to develop standardized study data content that meets all of CDER's data requirements.
- CDER will continue to participate in BRIDG/HL7 harmonization activities to address any gaps in the data standards.
- CDER remains committed to using CDISC standards for standardized study data content, regardless of the transport mechanism used for data exchange.
Please refer to the FDA Study Data Standards website for additional information.
Disease/Domain Content Standards R&D
CDER has identified the need to develop disease/domain-specific data standards to facilitate the evaluation of medical products. These clinical “content” standards consist of
- Clinical concepts for a specific disease or clinical domain area. (i.e. standard data elements)
- Their relationships
- Associated standard terminology (including standard value sets)
These content standards are primarily intended to support the assessments of new products in specific therapeutic areas, but, if properly developed, they may have other uses as well.
Ideally, data requirements for multiple use cases (e.g. healthcare, clinical research, public health reporting, regulatory review) are used to create a “superset” data standard that can support multiple uses of the data. This harmonization can help break down the information silos that adversely impact assessments across a medical product’s lifecycle.
It is desirable to incorporate well-designed standards into data collection instruments that are freely available (e.g. open source) and that can be used in a variety of use cases, such as clinical trials, routine clinical encounters, and for public health reporting activities. These data collection instruments can greatly facilitate the collection of more complete and higher quality data to support better analyses and improved regulatory and public health decision-making.
CDER wishes to identify and adopt disease/domain specific clinical content standards for use in clinical trials that may also have value in other use cases. Since many of these standards currently do not exist, CDER wishes to identify an efficient, predictable, and repeatable development process for disease/domain standards according to the following guiding principles:
- The standards can meet reviewer’s analytic needs
- The standards can be implemented in the near-term using currently supported submission standards for clinical trial data (i.e. CDISC SDTM/ADaM, SAS transport files version 5)
- The standards promote interoperability between clinical research and healthcare information systems
- The standards support the development and use of existing and emerging data exchange standards, specifically HL7 v3 templates for use in Clinical Document Architecture (CDA) and other v3 exchange standards.
As noted above, CDER is committed to working with clinical research and healthcare data standards development organizations (e.g. CDISC and HL7) to develop content standards and supportable transport mechanisms for exchanging data according to those standards.
Over the last decade, data standardization activities of clinical trial data have primarily focused on standardizing safety data. By and large, standard representations of a large portion of safety data common to all clinical trials already exist. However, similar standard representations of efficacy data are lacking, and some clinical domains are not yet standardized (e.g. medical imaging or histopathological data) for the purposes of reporting to FDA.
Efficacy data, by their very nature, are specific to the disease being studied, and therefore are less amenable to standardization across the entire clinical research enterprise and must be addressed individually disease by disease or domain by domain.
CDER has prioritized the key diseases/domains where data standardization can most effectively support our regulatory review activities. Prioritization efforts considered several factors: (1) areas of particular need, (2) areas with existing data standardization projects underway, and (3) areas with greater drug development pipeline activity. An initial list of 55 topic areas was identified and prioritized into 3 tiers. We encourage interested stakeholders to engage in and, whenever possible, sponsor these data standardization efforts.
Grouping the identified therapeutic areas into tiers recognizes that developing area-specific specifications for similar areas may be difficult to accomplish simultaneously and may benefit from sequential or delayed starts in order to leverage resources and learning. We also recognize that CDER resource limitations may limit our involvement in any given disease/domain standards development activity at any given time. We welcome input from other groups who may have accomplished work in these areas to avoid duplication of efforts.
The list is not exhaustive. Rather, it suggests possible initial priority areas for data standardization among the many diseases and domains for which CDER receives study data. In general, Tier 1 areas might be logical first order priorities; however, due to available resources and interests, projects in areas listed in Tier 2 and Tier 3 or in areas that are not listed might proceed ahead.
Please refer to the Priority Therapeutic Areas for Development webpage for additional information.
Questions and comments should be forwarded to the CDER eDATA team at firstname.lastname@example.org.