The National Institute of Health’s ClinVar public database of genetic variation is demonstrating good accuracy, and a handful of clinical labs are learning to share and review this relatively small genetic database
Accessible databases like ClinVar, which was launched by the National Institute of Health (NIH) in 2013, have emerged to aggregate genetic sequencing with acceptable results. ClinVar exists to meet the needs of the medical genetics community. It collaborates with organizations to make pertinent genetic information available.
ClinVar is an archive of compiled data relating to genotype and phenotype variations among humans. Through this database, individuals can present and peruse submissions regarding variants found in patient samples.
ClinVar is averaging about 6,000 submissions per month by both commercial laboratory companies and reference labs. Major contributors to the database include:
• Invitae Corp. (NYSE:NVTA);
• Quest Diagnostics Incorporated (NYSE:DGX); and
• Laboratory Corp. of America Holdings (NYSE:LH).
As of March, 2016, 180,000 submissions had been made to ClinVar since its inception.
Variations in Test Interpretation Has Consequences
A key challenge with a database like ClinVar is that different medical laboratories may interpret test outcomes in contrasting manners. This could result in patients being given incorrect information. Approximately 11% of submissions to ClinVar have two or more contributors, and 17% of those submissions have different interpretations. Since a patient’s treatment may be based on the interpretation, the variations can be significant and may result in serious consequences.
Heidi Rehm, PhD, is the Chief Laboratory Director at Laboratory for Molecular Medicine (LMM) at Partners HealthCare Personalized Medicine, and Medical Director, Clinical Research Sequencing Platform at the Broad Institute. She recently led an examination of some data located in the ClinVar database.
Rehm’s lab, along with the University of Chicago, Ambry Genetics, and GeneDx, analyzed over 6,000 variants submitted to ClinVar by two or more separate labs. After studying the data, the group discovered they were in concordance for classifications for 88% of the variants. They then reviewed the 724 variants about which there was disagreement and eventually came to a consensus on 86% of those variants. The study team was, however, unable to reach a definitive agreement on 33 variants within the study. At that time, there remained 492 variant discrepancies to be analyzed.
Crowdsourcing to Ensure Consistency in Reporting
Rehm sees the importance of creating a community-driven approach, supported by a massive effort in data sharing, as necessary to improve knowledge and consistency in classifying variants. This concept, which she compares to crowdsourcing, would include a process whereby researchers and labs within the medical community aggregate information to create a consistent baseline evaluation for the benefit of physicians and patients.
Rehm also has been striving to ensure consistency in reporting among clinical laboratories, gene testing companies, accrediting organizations, and journals, in regards to variant interpretations. She has suggested to the Food and Drug Administration (FDA) that it intervene and create regulations for labs that do not share variant interpretations to avert any threat to public health and safety. Rehm also would like insurers to require that medical laboratories share their data on variant interpretation before the insurer grants coverage.
Rapid Growth in Availability and Interest in Technology
The U.S. Department of Energy (DOE) and the National Institute of Health (NIH) launched the Human Genome Project in 1990 with the purpose of identifying and mapping all the genes within the human body. This study lasted for 13 years and determined that there are an estimated 20,000 to 25,000 genes in the human body.
Healthcare technology firm NextGxDx reports that there are currently more than 60,000 genetic tests available in the United States. The vast diversity of tests available range from carrier testing for genetic disorders, and diagnostic testing for genetic or chromosomal conditions, to prenatal diagnosis and newborn screenings, to genealogical and paternity testing to forensic analysis.
In the not-so-distant past, genetic testing was only performed in relation to specific diseases and conditions. Today, the genetic testing industry is experiencing rapid growth due to technological advances, an increase in the number of available tests, and expanding interest in the field. Data sharing will continue to be a critical tool for researchers, pathologists, clinical laboratory scientists, physicians, and patients as genetic testing continues to progress.