Back

Quantifying risk modifiers of hereditary hemochromatosis using genomic and electronic health record data from FinnGen and UK Biobank

Toivonen, J.; Clancy, J.; FinnGen, ; Aberg, F.; Ritari, J.; Arvas, M.

2025-10-02 gastroenterology
10.1101/2025.09.30.25336973 medRxiv
Show abstract

Hereditary hemochromatosis is an autosomal recessive disorder characterized by excessive iron accumulation in the body. Early diagnosis of hemochromatosis allows starting treatment before severe organ damage has occurred. The C282Y variant in the HFE gene is the most common cause of hemochromatosis. However, its penetrance of only 20% limits its utility for population-wide screening. We aimed to identify and quantify the predictive value of novel genetic variants and non-genetic factors from electronic health care records that could modify the effect of the C282Y variant on hemochromatosis, thereby partly explaining its incomplete penetrance. We carried out a cohort study on data from 420,543 individuals in the FinnGen project, for whom genotype information and health care records were available. We performed a regular and an interaction GWAS of hemochromatosis and fitted statistical models of hemochromatosis with multiple genetic and non-genetic predictors. We validated our results using data from the UKBB. Our analyses identified one novel statistically significant variant (rs181949568) in CASC15 gene, within 4 Mb from the HFE gene, and we were able to quantify the strong protective role of blood donation. We found that donating at least twice a year is likely to be sufficient to drop the risk of C282Y homozygote to that of a C282Y-H63D compound heterozygote. To facilitate clinical application, we present the results as individual-level risk tables. Also, we found that in Finland hemochromatosis related diagnosis predicted later hemochromatosis diagnosis suggesting that in Finland hemochromatosis is not well recognized.

Matching journals

The top 10 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.