NEW YORK — Many pathogenic genetic variants may actually only lead to a small increase in disease risk, a new biobank-based study has found.
Researchers from the Icahn School of Medicine at Mount Sinai used biobank and linked electronic health data to determine how often individuals with pathogenic disease variants were diagnosed with those diseases, as the penetrance of pathogenic or loss-of-function variants is not always clear.
As they reported Tuesday in the Journal of the American Medical Association, the researchers found, on average, pathogenic or loss-of-function variants were associated with a small, 7 percent increase in disease diagnosis. Some variants, though, such as ones in BRCA1 and BRCA2, exhibited much higher penetrance. These could, they said, influence how clinicians interpret genetic test results.
"A major goal of this study was to produce helpful, advanced statistics which quantitatively assess the impact that known disease-causing genetic variants may have on an individual's risk to disease," Ron Do, an associate professor of genetics and genomic sciences at Mount Sinai, said in a statement.
The researchers analyzed exome sequencing data on 74,645 individuals from the BioMe and UK Biobank datasets to find 5,360 variants associated with more than 150 different diseases. These variants were all considered pathogenic variants by the ClinVar database or were predicted to be loss-of-function variants by bioinformatic assessments. Variants with benign, uncertain, or conflicting designations in ClinVar were excluded as were missense or synonymous variants and variants with recessive inheritance patterns in the Online Mendelian Inheritance in Man catalog.
For the individuals in the biobanks, the researchers determined their case or control status based on the presence or absence of ICD-10 diagnostic codes from their linked health records, and they used that to determine risk estimates for the 157 different diseases.
Overall, 89 percent of the pathogenic or loss-of-function variants had a risk difference — the difference between the prevalence of the disease in individuals with the variant allele versus in individuals with a normal allele — of 0.05. Additionally, the mean penetrance for pathogenic alleles — the chance the variant was linked to a diagnosis — was 6.9 percent.
"At first I was quite surprised by the results. The risks we discovered were lower than I expected," Do said.
Some pathogenic or loss-of-function variants, though, were associated with a higher penetrance. The penetrance of pathogenic or loss-of-function variants in genes on the ACMG 73 and tier 1 list was higher than those in other genes. In particular, pathogenic or loss-of-function variants in BRCA1 or BRCA2 associated with breast cancer risk had an average penetrance of 38 percent, though the individual variants harbored a range in penetrance.
Factors like age and ethnic background also influence penetrance, the researchers noted. The penetrance of variants linked to age-related disorders like type 2 diabetes was higher among people over the age of 70, as compared to people over the age of 20.
The findings indicated to Do that clinicians might have to rethink how variants are classified. He and his colleagues suggest that the current system of classification based on pathogenicity should be updated to also reflect penetrance.
"While more research is needed to be done, we feel that this study is a good first step towards eventually providing doctors and patients with the accurate and nuanced information they need to make more precise diagnoses," Do said.