Monika Kawohl

1

New FDA Documents -Discussion of
Implementation Considerations

• Monika Kawohl
• Principal Statistical Programmer
• Accovion GmbH, Marburg, Germany

2
FDA Submissions - Experiences

• Questions
• Who has been involved in preparing SDTM for
  submission to FDA?
• Who has been involved in preparing ADaM for
  submission to FDA?
• Who has been involved in preparing CDISC mock
  submissions?
• Who has experiences with submitting in
  other/hybrid data formats?
• Who has been involved in discussions with the
  FDA?
• Experiences with submission of data pools in
  CDISC formats to support integrated analyses?
• Who has contacted the FDA with general questions
  orWho provided feedback on current FDA
  documents?
• For standardized data submission questions,
  contact EDATA_at_fda.hhs.gov.
• For center specific contacts seehttp//www.fda.g
  ov/ForIndustry/DataStandards/StudyDataStandards/de
  fault.htm

3
Chance to Get Involved in Discussions with FDA

• Support the FDA/PhUSE Collaboration
• see http//www.phuse.eu/css
• FDA/PhUSE Working Groups
• Data Validation and Quality Assessment
• Standardizing Data within the Inspection Site
  Selection Process
• Challenges of Integrating and Converting Data
  across Studies
• Standards Implementation Issues with the CDISC
  Data Models
• Development of Standard Scripts for Analysis and
  Programming
• Non-Clinical Road-map and Impacts on
  Implementation

4
CDER Common Data Standards Issues Document

• Version 1.1 (Dec-2011) Updates regarding SDTM
• Implementation of Amendment 1 to SDTM 1.2
  strongly preferred
• Subject Elements (SE) mandatory SDTM domain
• EPOCH in every subject-level domain
• ELEMENT/ETCD desired but not requested
• SUPPAE.AETRTEM
• Splitting domains (gt 1GB)
• provide split domains in separate subdirectory
  SPLIT???
• File size considerations
• use required maximum length of variableinstead
  of 200 for every character variable

5
CDER Common Data Standards Issues Document

• FDA Expected SDTM Variables
• Baseline flags (--BLFL) in findings domains
• Study day (-DY) for every -DTC
• derived based on RFSTDTC
• RFSTDTC start of treatment in most study
  designs
• Death diagnosis or comments in pathology reports
• SUBJID Subject identifier used in Study Report
• define.xml define.pdf (printable version, no
  need for hyperlinks)

6
CDER Common Data Standards Issues Document

• Version 1.1 Updates regarding ADaM
• Analysis Datasets should include
• appropriate numeric time variables(e.g. with
  value 8 for Week 8)
• basic demographic variables (e.g. sex)
• all covariates named in the protocol
• Analysis datasets should be derivable from SDTM
• To support traceability from results to collected
  data
• Not new but important!
• General Recommendation Contact the FDA

7
CDISC/FDA Webinar

• CDISC Standards In the Regulatory Submission
  Process, presented 26 January 2012
• recorded webinar
• presentation slides
• available at http//www.cdisc.org/webinars
• Follow-up on topics of special interest
• FAQ Database from CDER Data Standards Questions
  Team
• Accessible by the public?
• Importance of Reviewers Guide for adequate
  documentation
• Standard template?

8
FDA Standardized Study Data Draft Guidance

• http//www.fda.gov/downloads/Drugs/GuidanceComplia
  nceRegulatoryInformation/Guidances/UCM292334.pdf
  (Feb-2012)

9
FDA Study Data Standards Resources

• http//www.fda.gov/ForIndustry/DataStandards/Study
  DataStandards/default.htm

10
Implementing SDTM Amendment 1, FDA Docs.

• Information for new DM variables
• Store data in DM only vs. DM another Domain?
• Date/Time of Informed Consent
• Date/Time of Death
• Date/Time of End of Participation
• Derive AE Treatment Emergent Flag in SDTM
  (SUPPAE)?
• When to use SUPPQUAL?
• Generally provide SDTM variables Expected by
  FDA?
• Process to ensure appropriate variable length
• Process for splitting domains and how to document
  it

11
Implementing SDTM Amendment 1, FDA Docs.

• How to create the define.pdf?
• Always create a Reviewers Guide?
• Per submission datasets folder(i.e., separate
  ones per set of SDTM or ADaM data)?
• Describing SDTM/ADaM and/or multiple studies
  together
• Contents?
• OpenCDISC SDTM (/ADaM) validation
• Timing of compliance checks within the process?
• Describing/addressing issues
• Define.xml schema validation/OpenCDISC define.xml
  validation
• How to describe which SDTM versions and
  appendices are used?

12
Implementing Controlled Terminology (Updates)

• Excerpt from Draft Guidance Providing
  Regulatory Submissions in Electronic Format -
  Standardized Study Data (Feb-2012)When planning
  a study (including the design of case report
  forms, data management systems, and statistical
  analysis plans), the sponsor should identify
  which FDA-supported standard terminologies to use
  for submission.If a sponsor identifies a
  concept for which no standard term exists, we
  recommend that the sponsor submit the concept to
  the appropriate terminology maintenance
  organization as early as possible to have a new
  term added to the standard dictionary. We
  consider this good terminology management
  practice for any organization. The creation of
  custom terms for a submission is discouraged
  (i.e., so called extensible code lists)...
• Use of CDISC New Term Request Webpage?
• Experiences/Response Times?
• How to deal with CDISC CT Updates within a
  project?
• Documentation of CT version(s) used
• How to distinguish between CDISC CT values and
  sponsor extensions?
• Experiences with Updated Trial Summary (TS)
  domain?

13
Trial Summary (TS) Update

• Alignment with ClinicalTrials.gov
• New Trial Summary Parameters
• Indication of Use per Trial Summary Parameter
• Required
• Conditionally Required
• Expected
• Introduction of NULL FLAVOR Variable
• Coded Reason if TSVALmissing
• Introduction of underlying CT information
  variables

14
Trial Summary (TS) Update Sample