One in every 20 people may inherit a genetic variant in key genes that make them prone to developing cancer, according to a major database involving hundreds of thousands of participants.
The research, in JAMA, showed that pathogenic or likely pathogenic variants (P/LPVs) for cancer susceptibility genes were much more common than previously thought.
The findings suggest that the typical practice of reserving germline genetic testing for cancer susceptibility among people with known risk factors will not identify all at-risk individuals.
This could lead to missed opportunities for personalized screening and surveillance.
The researchers also noted that prior research has shown that structured clinical management for patients with pathogenic variants is both cost-effective and lifesaving.
“Although this evidence does not support additional screening, future studies should evaluate the clinical utility, cost-effectiveness, and feasibility of broader genetic screening,” recommended Gideon Idumah, PhD, from the Cleveland Clinic in Ohio, and co-workers.
Most prior research has focused on the prevalence of P/LPVs in patients with cancer and less is known about their frequency in large, unselected populations.
To investigate further, the team set out to estimate the population-level prevalence of pathogenic variants in key cancer susceptibility genes among 414,830 participants in the All of Us study, a major U.S. initiative funded by the National Institutes of Health to further precision medicine.
Data from short-read whole genome sequencing and demographic data such as sex and ethnicity were analyzed.
The researchers identified 3454 unique P/LPVs across 77 transcripts and 72 genes, in 20,968 unique individuals, representing approximately 5.05% of the total population. Among these individuals, 469 had variants in more than one gene.
The prevalence of pathogenic variants did not differ according to sex, the authors report.
However, there were significant differences across racial groups, with Asian participants having the lowest prevalence and White participants the highest at 5.72%.
Of the individual genes, MUTYH had the highest prevalence at 1.33%, followed by BRCA2 at 0.42%, and MITF at 0.37%.
Twelve genes significantly differed in prevalence depending on whether the participant was Hispanic.
Overall, racial differences were observed in 18 genes with MUTYH showing the greatest variability.
The researchers note that people who are heterozygous for the autosomal recessive genes MUTYH, NBN, BLM, WRN, MSH3, DIS3L2, and RAD50 have no known increased risk for cancer, albeit carrying a reproductive risk.
The analysis also showed that 8932 participants have an autosomal recessive condition. In addition, 450 participants had possible mosaic P/LPVs, using a threshold of variant allele frequency of less than 30%.
There were significant differences between carriers and noncarriers in the prevalence of cancer, age at first cancer diagnosis, and in the number of family members with a history of cancer diagnosis.
In terms of the age at which cancer was first diagnosed, STK11 carriers and DICER1 carriers had the lowest median age, at a corresponding 31.4 and 35.4 years, while AIP carriers the highest at 70.8 years.
The carrier frequencies for BRCA1 and BRCA2 were 0.22% and 0.42%, respectively, which was far higher than previous estimates for non-Ashkenazi Jewish populations of between 0.13% and 0.25%.
The researchers note that a previous study suggested rates of thyroid cancer–associated variants approximately 10 to 20 times higher than current population estimates.
“Combined, these findings suggest that current genetic testing guidelines may not identify all at-risk individuals, although the clinical impact of broadening screening has not been evaluated,” the team wrote.
