Session 3: The New NIH Genetic Testing Registry: Peering Into the Black Box of Genetic Tests

Slide 1 of 35

The New NIH Genetic Testing Registry: Peering Into the Black Box of Genetic Tests
Public Health Genomics Interest Group Seminar Series
NCI Division of Cancer Control and Population Sciences

February 15, 2012

Wendy Rubinstein, MD, PhD, FACP, FACMG
Senior Scientist, NIH/NLM/NCBI
Director, NIH Genetic Testing Registry

The following relationship(s) exist related to this presentation:

No relationships to disclose.

Slide 2 of 35

Outline:

Recognizing the black box is a problem, and deciding to do something about it
Finding out what’s in the black box (GTR live demo)
Thinking outside the box

Slide 3 of 35

The Continuum of Translation Research in Genomic Medicine: How Can We Accelerate the Appropriate Integration of Human Genome Discoveries Into Health Care and Disease Prevention?

T1: From Gene Discovery to Health Application
T2: From Health Application to Evidence-based Guideline
T3: From Guideline to Health Practice
T4: From Practice to Health Impact

Gene discovery → Genetic tests

Khoury MJ et al. Genetics in Medicine. 2007 Oct;9(10):665-74.

Slide 4 of 35

Some problems

Pervasive belief that T1 = T4
Manifest destiny of genetics and genomics

Slide 5 of 35

Some problems

Extrapolation from a single example to all of health care

[Image of two men debating with each other from podiums]

Ransohoff & Khoury v. Gulcher & Stefannson

Eur J Clin Invest 2010

Slide 6 of 35

Some problems

Extrapolation from a single example to all of health care, when disease prediction matches disease diagnosis

Therefore, all predictions are true
All opportunities to intervene are effective and lifesaving
All individuals will rationally pursue avenues toward better health
Numeracy and risk perception are ideal
No potentially adverse effects need to be considered
Every observed association will translate into better public health

Slide 7 of 35

Genetic test is a basic unit or 'analyte' of genomic medicine

Subject of a study
Component of a study

[Image of T1 to T4 research translation continuum]

Slide 8 of 35

Some more problems

Information void
How many genetic tests are clinically available?

Slide 9 of 35

How many genes? How many tests?

[Image of graph showing growth by year, from 1993-2011, of the number of diseases for which genetic testing is available against the number of laboratories].

Source of data is the Gene Tests Database from 2011.(www.genetests.org)

http://www.ncbi.nlm.nih.gov/sites/GeneTests/

Slide 10 of 35

Some more problems

Information void
How many genetic tests are clinically available?
How are these tests performed (methods)?
What is the purpose?
Are there health benefits?
What is the evidence?

Slide 11 of 35

Yet more problems

Which genes have clinically valid tests?
Which conditions have clinical utility for genetic testing?
Which variants are pathogenic and which are not?

Slide 12 of 35

Yet more problems

Which genes have clinically valid tests?
Which conditions have clinical utility for genetic testing?
Which variants are pathogenic and which are not?

Thinking outside the box

Slide 13 of 35

One piece of the puzzle

Recognizing the black box is a problem, and deciding to do something about it

[Image of a puzzle with one missing piece]

Slide 14 of 35

The Call for a Test Registry

Currently, there are tests for more than 2,000 genetic conditions but no single source of information about these tests.

A 2008 Secretary's Advisory Committee on Genetics, Health, and Society report recommended that HHS establish a test registry to increase the transparency of genetic testing.

Other policy and advocacy groups have called for a registry that includes tests across the risk continuum and comprehensive information to enable informed decision making regarding genetic testing.

Javitt G et al. Developing the blueprint for a genetic testing registry. Public Health Genomics. 2010;13(2):95-105.

Zonno K. Call for Action from Genetic Alliance. Registry of Genetic Tests: A Critical Stepping Stone to Improving the Genetic Testing System. Genet Test Mol Biomarkers. 2009; 13(2): 153–154.

Slide 15 of 35

Clinical Genomics

Need database anchored on tests, not diseases
Must accommodate complex information
- Arrays
- Whole genome and whole exome tests

Slide 16 of 35

NIH Responds

NIH developed a voluntary genetic testing registry to:

Encourage providers of genetic tests to enhance transparency by publicly sharing information about the availability and scientific basis of their tests;
Provide an information resource for health care providers, researchers, and patients to locate laboratories that offer particular tests; and
Facilitate research and new scientific discoveries.

Slide 17 of 35

GTR Development Steps

March 18, 2010: NIH announcement to develop the GTR
June-August 2010: RFI public comment period
November 2, 2010: Public stakeholder meeting
April 2011: Initiated Paperwork Act Reduction process
January 18, 2012: 604 labs participating in GeneTests notified of plan to transition to the GTR. None opted out
February 7, 2012: Soft launch of the GTR website. Tests transitioned from GeneTests to GTR, GTR navigation features, resource links integrated.
Feb-March 2012: Open submission site pending approval by the Office of Management and Budget to collect information.
Mid-2012: Realize full potential of the GTR as labs migrate and fully register GTR tests

Slide 18 of 35

Stakeholder Input

Input from diverse stakeholders (e.g., test developers, health care providers, industry) throughout GTR development

84 public comments from 3 Federal Register notices
17 public comments from public stakeholder meeting
95 comments through "Contact GTR"
19 meetings/teleconferences with stakeholder groups
3 meetings with other government agencies
Feedback from presentations at 7 professional meetings
10 consultations with two clinical advisory groups

Slide 19 of 35

GTR Team

NIH Office of the Director – Policy oversight
Jim Ostell, Ph.D.
Chief, NCBI Information Engineering Branch
Directs NCBI's suite of genome tools and resources: GenBank, dbSNP, dbGaP, RefSeq, PubMed, PubMed Central, etc.
Wendy Rubinstein, MD, Ph.D.
GTR Director
Donna Maglott, Ph.D.
Lead, Database development
Jennifer Lee, Ph.D.
Lead, Web development
Brandi Kattman, M.S., C.G.C.
Genetic counselor
Adriana Malheiro, M.S.
Genetic counselor
Team of programmers, web developers, usability experts

Slide 20 of 35

GTR Advisory Groups

NCBI Medical Genetics Working Group:

Christine Seidman, M.D.
Leslie Biesecker, M.D.
Wendy Chung, M.D., Ph.D.
David S. Konecki, Ph.D.
Robert Nussbaum, M.D.
Charmaine D.M. Royal, Ph.D.
Wendy R. Uhlmann, M.S.
Marc S. Williams, M.D.

NIH Clinical Advisory Group:

Ivona Aksentijevich, M.D.
Leslie Biesecker, M.D.
Thomas A. Fleisher, M.D.
Daniela Gerhard, Ph.D.
Katrina Gwinn, M.D.
Stephen C. Groft, Pharm.D.
Emily Harris, Ph.D.
Suzanne Hart, Ph.D.
Rochelle M. Long, Ph.D.
Francis McMahon, M.D.
Catherine McKeon, Ph.D.
Winifred K. Rossi, M.A.
Robert Shamburek, M.D.
Bryan Traynor, M.D., M.M.Sc.
Tiina Urv, Ph.D.

Slide 21 of 35

Phased Approach

Initial phase:

Single-gene tests for heritable mutations, including pharmacogenetic tests
Multiplex panels and arrays

Subsequent phases:

Tests for somatic mutations (e.g., solid tumors, hematological malignancies)
Direct-to-consumer tests
Whole exome sequencing / whole genome sequencing

Slide 22 of 35

Data Elements

Designed to collect the maximum amount of information while taking into consideration
- Burden on the submitters
- Input from a variety of stakeholders
Distinctions between minimal, recommended, and optional fields
Not included
- Test price
- Patents and licensing agreements
- Turn-around time
- Proprietary information

Slide 23 of 35

Overview of GTR Information

Test information:

Intended use
Target population
Assay Limitations
Methodology
Analytical validity
Clinical validity
Clinical utility
Ordering information
Test credentials (e.g., FDA approved/cleared)

Laboratory information:

Contact information
Certification(s), license(s)

Additional resources:

Other NLM resources and professional practice guidelines well integrated in the GTR

Slide 24 of 35

Intended Audience

Initial target audience is health care providers
GTR aims to serve a wider audience and to increase usability for
- Non-genetics health care providers
- Patients/consumers

Slide 25 of 35

Quality of Information

Code of Conduct

Information that is accurate and not misleading

Professional organization 'stamp of approval' invited

Slide 26 of 35

Code of Conduct

[Image of GRT Code of Conduct Webpage from NCBI at http://www.ncbi.nlm.nih.gov/gtr/docs/code/]

Slide 27 of 35

Key features of GTR navigation

Global search and tabbed searches
Autocomplete dictionary
Quickly limit disease search results
- Tests, OMIM, or GeneReviews available
Quickly filter test results
- Condition/Phenotype
- Clinical or Research test
- Test purpose
- Test method
- Certifications e.g., NY CLEP-certified
- Laboratory Location
Compare labs and their methodology menus
- Plan sequential testing in proband and family

Slide 28 of 35

Key features of GTR navigation

Detailed Test pages with overview and tabs
Discovery Panel – context-specific
- Clinical practice guidelines e.g., ACMG, EGAPP, CPIC
- Automated searches e.g., GeneReviews, OMIM, Orphanet, PubMed
Locate a genetics professional
- ACMG, NSGC, GeneTests, NCI, ABMG, ABGC
Consumer Resources
- Print information for your patient from Genetics Home Reference
Access to NCBI's suite of molecular tools and resources
- Variation
Login to MyNCBI to save preferences for displays and retrieval sets
Stable accession and version history

Slide 29 of 35

Nomenclature and Standards

Challenging!
- Many overlapping nomenclatures and varied conventions
The GTR strongly endorses use of established standards
- GTR will continue GeneTests practice of using the HGNC standard for gene symbols and UniProt for protein names
- Whenever available, GTR will use disease names from SNOMED CT, and HGVS expressions for variations. Searches by alternate terms will continue.
We use SNOMED CT - a standard vocabulary for use in Electronic Medical Records
- Support other standard health record keys such as ICD codes, and LOINC codes
- Forward-looking to direct connections between electronic health systems and the GTR

Slide 30 of 35

Finding out what's in the black box

Live demo of GTR

http://www.ncbi.nlm.nih.gov/gtr/

Slide 31 of 35

Solving some problems

Secondary (incidental) variants:

Which to disclose (and in which settings)?

∑ → Whole genome tests

Slide 32 of 35

Solving some problems

Which genes have clinically valid tests?

Which conditions have clinical utility for genetic testing?

Slide 33 of 35

Solving some problems

Which genes have clinically valid tests?
Which conditions have clinical utility for genetic testing?
Tests, genes, and variants in GTR
- Claims and evidence provided by test submitters
- Stable versioning, can reference unique tests and review as if an analyte
- Aggregate info per similar tests for professional group
- ∫ Integrate information by professional group
ACMG workgroup taking on this task?
- GTR will display professional stamp of approval

Slide 34 of 35

Solving some problems

Which variants are pathogenic and which are not?

Slide 35 of 35)

[Image of ClinVar graphic and tagline: “ClinVar aggregates information about sequence variation and its relationship to human health]

ClinVar represents the relationship of genotype, phenotype, and clinical interpretation based on supporting evidence
Aggregating information about medically important human variation
Structured observations are recorded to facilitate aggregation, comparison, search, and re(evaluation)
ClinVar is an archive, not an interpretation tool
Layers of assertions: Confidence in any assertion is indicated as a range from a single source submission to practice guidelines
Attribution: Sources are acknowledged, with gateways to publications and external databases
Terminology consistent with community standards
Unrestricted availability: Data can be downloaded and integrated into external databases and local analysis pipelines
Will be a distinct web resource later this year
Companion resource with Genetic Testing Registry