Transcript CDISC - UCT Clinical Research Centre

DEVELOPING A
DATA STANDARD
FOR MALARIA
Clinical Data Interchange Standards Consortium (CDISC)
WorldWide Antimalarial Resistance Network (WWARN)
Lesley Workman
CRC, 16 February, 2016
www.wwarn.org
Twitter: @WWARN
Why do we need data standards for malaria?
• Data Sharing has shifted from an “add on” to an essential part
of the data cycle continuum – required by more and more
journals and medicines regulatory authorities.
• Meaningful data sharing requires its standardization
• Data standards also enable Data exchange / sharing for:
– Comparing data
– Pooling data
– Data re-use
“In short, establishing common standards for
data reporting will provide new opportunities
to transform the massive amount of data
from drug studies on specific diseases into
useful information to potentially speed the
delivery of new therapies to patients.”
Janet Woodcock, M.D.
FDA Works with Partners to Establish Important Therapeutic Area Data
Standards, October 24, 2012. FDA_Voice
Why CDISC?
• CDISC has established worldwide industry standards
to support the electronic acquisition, exchange,
submission and archiving of clinical research data
and metadata that are platform independent
• CDISC has been adopted by FDA as the mandatory
standard for data submissions after December 2016
• EMA and Japan are currently reviewing the adoption
of CDISC
• NIH and CDISC have recently started discussing the
need for similar standards for epidemiology studies
WWARN’s Experience
Over 5 years developing tools and platforms for curating, pooling and
archiving data: using data dictionaries, study protocols and data sets
relating to antimalarial efficacy contributed from a broad group of
researchers from over 230 institutions in > 50 countries on >110,000
individual patient and ~10 000 healthy volunteer records
Difficulties encountered:
• Lost/unavailable data
• No common standards
• Data not anonymized
• Poorly structured data
Gaps:
• Data only on licensed antimalarials
• Phase III-IV only
Draft Project Mission:
To support efficient, scientifically valid generation and reporting of clinical
malaria data to streamline antimalarial development, regulatory submission
and post-marketing research, as well as enable data sharing, comparison and
aggregation.
Draft Aim:
Consensus-based development of a single, freely available data standard to
ensure the consistent use of existing CDISC standards, and to facilitate
alignment and development of new standards for the electronic acquisition,
exchange, submission and archiving of clinical malaria data collection,
analysis and reporting
Draft Scope (Year 1):
– Uncomplicated falciparum malaria
– Uncomplicated vivax malaria
Preventive treatment (e.g SMC, IPT, MDA), severe malaria to follow, as needed
CDISC Malaria Project Snapshot
Focus: Uncomplicated Malaria
•
•
Status
Diagnosis
Clinical and laboratory findings
(baseline and follow-up) to assess
therapeutic efficacy
Project Scope
Review : to ensure existing CDISC
standards for the following are
complete
•Medical history
•Antimalarial Drug administration
•Pharmacokinetic sampling
•Adverse Events of special interest.
• Essential core clinical data items, with
definitions, data types and SDTM
mappings
• Concept maps of malaria research
concepts
• CDASH metadata for selected research
concepts
• Annotated CRFs (with CDASH and
SDTM-based annotations)
• ADaM: Analysis Data Model
• TAUG: Therapeutic Area User Guide
7
DATA versus METADATA
• When standardising data it is critical to have the
metadata
• Data is a single term of factual information
– EG. If the CRF section is vital signs and the measurement
required is “Temperature”
– The data is the actual recorded Result of temperature for
that person at that time
• Metadata describes the characteristics of the data
– Metadata for “Temperature” includes method of
recording and units of measurement
DATA ELEMENTS
• Data Elements are attributes about the data
– Stored in metadata repositories (for reuse)
• Used to be referred to as “data dictionaries”
• A simple data element will contain:
– Description (EG CRF FIELD name “What is your age?”)
– VARIABLE name (EG. CDISC name Age)
– Data type (character, numeric (also needs to include
decimal places/integer), date, binary
– Allowed values for the data item (EG. Controlled
terminology for SEX Male/Female)
CONTROLLED TERMINOLOGY and DOMAINS
• When many different terms are used to describe similar data
concepts it becomes difficult to reuse and prevents data
interchange without curation/mapping
• Controlled Terminology
– Defines allowed values for each SDTM
• Dates
– YYYY-MM-DD/DD-MMM-YYYY
• Unique study identifier: USUBJID
• [PATNO; SCR NUM; ENROLNUM; PID; STUDY NUMBER] etc. etc. etc.….
• DOMAINS: Collection of observations that relate to a particular topic
– EG: Demographics, Adverse Events, Vital signs, Medical History, Laboratory
• Standard variables in each domain EG. Laboratory Test Results
Chains of Connected Concepts
Subject
Study
Subject
Specimen
Collection
Result /
Subject
Specimen
Result /
Subject
Lab Test
Lab Test
Result
Result
Clinical
Significance
Assessment
Clinical
Significance
Result
Slide courtesy of Diane Wold (GSK)
11
Draft Concept Map
NEXT STEPS..CDASH
The standardization effort is planning to
1. Define each data element relevant to uncomplicated malaria
2. Create a concept map (defines how data elements are
related)
3. Develop the meta data
4. Define and describe the controlled terminology
FINAL OUTPUT
THERAPEUTIC AREA USER GUIDE:
UNCOMPLICATED MALARIA FOR P FALCIPARUM
1. Annotated CRF template (CDASH)
2. Examples of SDTM
3. All metadata with descriptions
4. Concept maps
STUDYID
WWARN01
WWARN01
WWARN02
WWARN02
WWARN03
WWARN03
WWARN04
WWARN04
DOMAIN
LB
LB
LB
LB
LB
LB
LB
LB
USUBJID LBSEQ LBCAT
W001
1 MALDIAG
W001
2 MALDIAG
W001
3 MALDIAG
W001
4 MALDIAG
W002
1 MALDIAG
W002
2 MALDIAG
W002
3 MALDIAG
W002
4 MALDIAG
LBSCAT
P falciparum
P falciparum
P falciparum
P falciparum
P falciparum
P falciparum
P falciparum
P falciparum
LBTEST
SEXUAL
ASEXUAL
SEXUAL
ASEXUAL
SEXUAL
ASEXUAL
SEXUAL
ASEXUAL
LBORRES
LBORRESU VSDY
0 µL
0
4840 µL
0
0 µL
1
160 µL
1
120 µL
0
34688 µL
0
24 µL
1
3240 µL
1
CDISC Standards specify how to structure the data
to support efficient data sharing for regulated clinical
trials.
CFAST is an initiative of CDISC and the Critical Path
Institute to accelerate clinical research and medical
product development by facilitating the creation and
maintenance of data standards, tools, and
methods for conducting research in therapeutic areas
important to public health.
The role of stakeholders
Current Stakeholders include:
•
•
•
•
CDISC, CPATH
WWARN members
WHO GMP / TDR
Pharmaceutical Manufacturers:
o GlaxoSmithKline
o Medicines for Malaria Venture
o Merck
o Novartis
o Sanofi
o Shin Poon
o Sigma Tau
o Takeda
o UCB
Others interested, please contact
[email protected]
Review draft data standards
Share relevant experience:
•
•
•
•
Recent CRF templates (~CDASH)
Data Specifications (~SDTM)
Statistical Analysis Plans (~ADaM)
Identification of critical issues in
regulatory submissions.
Pilot test CDASH
Critical regulatory issues for consensus building, e.g.:
1. Antimalarial efficacy
• Study population definition (PP vs. mITT vs. ITT)
• Parasite / Fever Clearance Time, its measurement and analysis
• Duration of follow up and the time points for assessment (and relationship
to half-life of the antimalarial/s tested).
• Role of PCR (recrudescence / reinfection / )
• Contribution of each partner drug to efficacy of a FDC
• Measurement of gametocyte carriage
2. Clinical data support other comparative advantages
• Improved compliance / adherence
• Transmission blocking
• Less vulnerable to resistance
3. Antimalarial safety
• Differentiating malaria disease effects from drug induced liver injury, ECG
(particularly QTc interval) changes, anaemia, haemolysis.
Sharing and Archiving Clinical Data to Strengthen EBM
• We need to share data
• We need to archive in order to share
• We need an archival format that is standardised and
human readable in order to share meaningfully
• Organisations to take long term responsibility for:
– Maintaining the standard archival format e.g. CDISC
– Curating and approving data submissions
– Providing the storage infrastructure and maintaining the
platform
Archiving the Phenome: Clinical Records Deserve Longterm Preservation – JAMIA paper
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC260559
2/
RECOMMENDATION 1: Stakeholders in clinical trials should foster a culture in which data
sharing is the expected norm, and should commit to responsible strategies aimed at
maximizing the benefits, minimizing the risks, and overcoming the challenges of sharing
clinical trial data for all parties.
RECOMMENDATION 2: - Timelines for sharing data
RECOMMENDATION 3: - Data sharing agreements
RECOMMENDATION 4: The sponsors of this study should take the lead, together with or via
a trusted impartial organization(s), to convene a multi-stakeholder body with global reach
and broad representation to address, in an ongoing process, the key infrastructure,
technological, sustainability, and workforce challenges associated with the sharing of
clinical trial data.
WHAT DOES CDISC DATA LOOK LIKE?
USEFUL LINKS
• http://www.cdisc.org/
• http://www.cdisc.org/Video-Library
• http://www.wwarn.org/tools-resources/clinical-datamanagement-and-analysis-plan
http://califesciences.org/live-webinar-cdisc-sdtmconversion-made-easy-with-cdisc-express/