Complexities of Penetrance and Expression of Monogenic Disorders

A Scalable Approach for Genomic-first Rare Disorder Detection in a Healthcare-Based Population This is an interesting study which looks at the penetrance of disorders by evaluating genomic data and phenotypes from electronic health records (EHRs) of 218,680 participants (who consented) of Geisinger’s MyCode Community Health Initiative. The study analyzed for variants in the curated list of 2701 high-confidence, single-disorder-associated genes which are different from the American College of Medical Genetics and Genomics Secondary Findings list spanning 81 genes.[1] Variant inclusion criteria were stringently defined. More than 5000 (5484/218,680 [2.5%]) participants had a positive molecular finding (single nucleotide variants [SNV] or indel) spanning 490 genes. EHRs were evaluated if a corresponding diagnosis is a clinical fit for a given positive molecular finding “DxFit.” Overall, 213 of 490 positive molecular finding genes had at least one molecular-positive participant with DxFit. An additional 0.7% (n = 1,455) had possible molecular findings ranging from compound-heterozygous or novel loss-of-function (LoF) variants. Of the high-confidence molecular positives, 15.0%–21.1% had evidence of a corresponding clinical fit from existing diagnosis codes. The remainder lacked a corresponding clinical diagnosis code. There are two main reasons for this: one being that genomic testing may be more sensitive than clinical evaluation, and the other is that many so-called “monogenic disorders” have low penetrance, which we do not perceive when we evaluate symptomatic patients. This study is an eye-opener, especially for ordering genomic studies in asymptomatic individuals or reporting incidental findings, and highlights the fact that the magnitude of nonpenetrance for many genetic variants remains unclear. Exploring Penetrance of Clinically Relevant Variants in Over 800,000 Humans from the Genome Aggregation Database This article interrogates the underlying mechanisms of incomplete penetrance in clinically relevant variants. The investigators conducted a comprehensive case-by-case assessment of 734 predicted LoF variants found in gnomAD genomes in 77 genes associated with severe, early-onset, highly penetrant haplo-insufficient disease and found that most of them were located in the last exon or last 50 bp of the penultimate exon, resulting in a variant not predicted to undergo nonsense-mediated decay. Some of the variants had low expression score, some LoF variants were rescued by other modifying variants (frame-restoring indels), others were in non-coding transcripts, some represented somatic variants, and a proportion were rescued by different types of splice-modifying variants predicted by SpliceAI. The unexplained LoF variants prompted further investigation, which showed that some variants originated from the 1000 Genomes Project and may be cell culture-acquired variants, while some others were the result of age-related clonal hematopoiesis, and a subset of them had less biological relevance. Such a detailed investigation left only 33 out of 734 (4.5%) LoF variants without explanation.[2] The study also reports a rare example of incomplete penetrance of a pathogenic heterozygous variant p.Gly45Glu in the GJB2 gene, which, when found in cis with another LoF variant, does not cause any disease phenotype as the combined effect of the two variants nullifies the dominant-negative effect of the heterozygous p.Gly45Glu variant. Such a haplotype was identified in individuals belonging to the East Asian genetic ancestry group, and there was no evidence of this being a common mechanism in the general population. This study emphasizes the importance of deep assessment of a variant, especially in the absence of clinical features or positive family history, far beyond the standard criteria, highlighting that LoF variants are not invariably disease-causing and may exhibit reduced penetrance due to some underlying mechanisms. Low Population Penetrance of Variants Associated with Inherited Retinal Degenerations The authors have used two large biobanks (All of Us and UKB) with linked genomic and clinical data to quantify the population-level penetrance of inherited retinal degeneration (IRD)-associated variants. A cohort of 481 individuals with definite IRD-compatible genotypes was evaluated for disease-associated phenotypes, namely IRD, retinopathy, and a screening group of disorders such as myopia, refractive errors, and other ophthalmic disorders. The study showed that within the genotyped cohort, the disease-associated frequency (DAF) for IRD was 4.9%, the DAF for retinopathy was 16.8%, while that for the screening set was 28.1%, which suggests that IRD-causative genotype penetrance is low. They validated this in another cohort of the biobank, which linked ocular imaging data of the participants with their whole-exome sequencing results. Only 16.1%–27.9% of participants had definitive or possible IRD features. The study shows that the population-level penetrance of inherited retinal degeneration (IRD)-associated genotypes is markedly lower than the assumption of nearly 100% and suggests that a pathogenic variant may not be sufficient to cause disease, and some background genetic or environmental mechanisms may modify the risk.[3-4] Genetic testing of patients with retinal degeneration introduces a phenotypic ascertainment bias, which can overestimate the disease penetrance and severity. Molecular Genotype-phenotype Correlation in ACTB- and ACTG1-related Nonmuscle Actinopathies Long ago, it became obvious that the one gene- one disease hypothesis is not true. ACTB1 and ACTG1 genes are associated with the autosomal dominant disorder, Baraitser-Winter cerebrofrontofacial syndrome, but some other phenotypes related to these genes are also described. This article describes a cohort of 290 individuals with pathogenic/likely pathogenic variants in the ACTB1 and ACTG1 gene related nonmuscle actinopathies (NMA). The ACTG1 gene has a low pLI (probability of LoF intolerance) score, and there is a high prevalence of LoF variants in the population datasets, LoF variants in this gene are unlikely to be pathogenic. ACTB and ACTG1 missense and in-frame variants result in multiple NMAs. The authors categorised various phenotypes in clinically useful groups but could not find genotype-phenotype correlations inspite of conducting functional experiments. There is high variability among ACTB and ACTG1 missense variants, and the phenotypic outcomes may vary even in individuals within the same subtype. Understanding of genotype-phenotype correlation is important for personalized medicine, where individual genotype-based therapies can be utilized. However, this article highlights the clinical heterogeneity and complex pathobiology of genetic non-muscle actinopathies. Data availability Not Applicable. Author’s contribution Yashavi Gupta conceptualized the manuscript, extracted and compiled data from the included studies, interpreted the findings, and drafted the manuscript.