The genetic tests used for more than a decade to identify patients or family members who carry genetic mutations linked to hypertrophic cardiomyopathy are seriously flawed.
The tests have been erroneously flagging people as genetically positive for hypertrophic cardiomyopathy (HC) when they actually carried benign genetic variants, according to a new reassessment of the genetic linkages by researchers using a more genetically diverse database. Five genetic variants now reclassified as benign were collectively responsible for flagging 74% of people flagged at genetic risk for HC in the more than 8,500 cases examined.
The results call into question any genetic diagnosis of HC made since genetic testing entered the mainstream in 2003, especially among African Americans who seem to have been disproportionately affected by these mislabeled genetic markers because of inadequate population diversity when the markers were first established.
In addition, the results more broadly cast a shadow over the full spectrum of genetic tests for disease-linked variants now in routine medical practice because of the possibility that other linkage determinations derived from an inadequately-representative reference population, reported Arjun K. Manrai, PhD, and his associates (N Engl J Med. 2016 Aug 18;375:655-65). The researchers call the HC experience they document a “cautionary tale of broad relevance to genetic diagnosis.”
The findings “powerfully illustrate the importance of racial and genetic diversity” when running linkage studies aimed at validating genetic markers for widespread clinical use, Isaac S. Kohane, MD, senior author of the study, said in a written statement. “Racial and ethnic inclusiveness improves the validity and accuracy” of genetic tests, said Dr. Kohane, professor of biomedical informatics and pediatrics at Harvard Medical School in Boston.
“We believe that what we’re seeing in the case of hypertrophic cardiomyopathy may be the tip of the iceberg of a larger problem that transcends a single genetic disease,” Dr. Manrai, a biomedical informatics researcher at Harvard, said in the same statement. “Much genetic assessment today relies on historical links between a disorder and variant, sometimes decades old. We believe our findings illustrate the critical need to systematically reevaluate prior assertions about genetic variants,” Dr. Manrai added in an interview.
The two researchers and their associates reexamined the link between genetic variants and HC in three genetic databases that involved a total of more than 8,500 people. One database included 4,300 white Americans and 2,203 black Americans. A second database included genetic data from 1,092 people from 14 worldwide populations, and the third had genetic data from 938 people from 51 worldwide populations.
The analysis showed that although 94 distinct genetic variants that had previously been reported as associated with HC were confirmed as linked, just 5 met the study’s definition of a “high-frequency” variant with an allele frequent of more than 1%. These five variants together accounted for 74% of the overall total of linkages seen in these 8,533 people. Further analysis classified all five of these high-frequency genetic variants as benign with no discernible link to HC.
These five high-frequency variants occurred disproportionately higher among black Americans, and the consequences of this showed up in the patient records the researchers reviewed from one large U.S. genetic testing laboratory. They examined in detail HC genetic test results during 2004-2013 from 2,912 unrelated people. The records showed seven people had been labeled as carrying either a pathogenic or “presumed pathogenic” variant when in fact they had one of the five high-frequency variants now declared benign. Five of the seven mislabeled people were of African ancestry; the other two had unknown ancestry.
The researchers called for reevaluating known variants for all genetic diseases in more diverse populations and to immediately release the results of updated linkage assessments. This has the potential to meaningfully rewrite current gospel for many genetic variants and linkages.
“Our findings point to the value of patients staying in contact with their genetic counselors and physicians, even years after genetic testing,” Dr. Manrai said. Reassessments using more diverse populations will take time, he acknowledged, but tools are available to allow clinical geneticists to update old variants and apply new ones in real time, as soon as a new assessment completes.
Dr. Manrai and Dr. Kohane had no disclosures.
On Twitter @mitchelzoler