Lifetime risk of autosomal recessive mitochondrial disorders calculated from genetic databases

Abstract

Background: Mitochondrial disorders are a group of rare diseases, caused by nuclear or mitochondrial DNA mutations. Their marked clinical and genetic heterogeneity as well as referral and ascertainment biases render phenotype-based prevalence estimations difficult. Here we calculated the lifetime risk of all known autosomal recessive mitochondrial disorders on basis of genetic data.

Methods: We queried the publicly available Genome Aggregation Database (gnomAD) and our in-house exome database to assess the allele frequency of disease-causing variants in genes associated with autosomal recessive mitochondrial disorders. Based on this, we estimated the lifetime risk of 249 autosomal recessive mitochondrial disorders. Three of these disorders and phenylketonuria (PKU) served as a proof of concept since calculations could be aligned with known birth prevalence data from newborn screening reports.

Findings: The estimated lifetime risks are very close to newborn screening data (where available), supporting the validity of the approach. For example, calculated lifetime risk of PKU (16·0/100,000) correlates well with known birth prevalence data (18·7/100,000). The combined estimated lifetime risk of 249 investigated mitochondrial disorders is 31·8 (20·9-50·6)/100,000 in our in-house database, 48·4 (40·3-58·5)/100,000 in the European gnomAD dataset, and 31·1 (26·7-36·3)/100,000 in the global gnomAD dataset. The disorders with the highest lifetime risk (> 3 per 100,000) were, in all datasets, those caused by mutations in the SPG7, ACADM, POLG and SLC22A5 genes.

Interpretation: We provide a population-genetic estimation on the lifetime risk of an entire class of monogenic disorders. Our findings reveal the substantial cumulative prevalence of autosomal recessive mitochondrial disorders, far above previous estimates. These data will be very important for assigning diagnostic a priori probabilities, and for resource allocation in therapy development, public health management and biomedical research.

Funding: German Federal Ministry of Education and Research.

Keywords: Autosomal recessive mitochondrial disorders; Lifetime risk; POLG; Population genetics; Prevalence; SPG7.

Fig. 1

Analysis diagram showing the experimental design. A comprehensive set of genes for autosomal recessive mitochondrial disorders was defined based on a review of the literature. ClinVar and HGMD were queried to collect the total number of disease causing variants for each gene. In addition, loss of function variants in our in-house database and gnomAD were considered pathogenic. These variants were evaluated for their pathogenicity according to the ACMG guidelines. The lifetime risk of mitochondrial disorders was calculated based on the allele frequencies of these variants in the gnomAD dataset as well as in our in-house database.

Fig. 2

Calculated life time risk for the most prevalent autosomal recessive mitochondrial disorders. Comparison of the lifetime risks of different monogenic nuclear mitochondrial diseases according to the in-house database and gnomAD dataset calculated independently for the European population and the overall dataset. Error bars represent 95%-confidence intervals. (a) depicts the lifetime risk for genes >1/100,000 in any population as well as for the remaining 220 genes. (b) shows the lifetime risk for genes with a lifetime risk >0.02/100,000 in any population.