Archetypal NOTCH3 mutations frequent in public exome: implications for CADASIL

Abstract

ObjectiveTo determine the frequency of distinctive EGFr cysteine altering NOTCH3 mutations in the 60,706 exomes of the exome aggregation consortium (ExAC) database.MethodsExAC was queried for mutations distinctive for cerebral autosomal dominant arteriopathy with subcortical infarcts and leukoencephalopathy (CADASIL), namely mutations leading to a cysteine amino acid change in one of the 34 EGFr domains of NOTCH3. The genotype‐phenotype correlation predicted by the ExAC data was tested in an independent cohort of Dutch CADASIL patients using quantified MRI lesions. The Dutch CADASIL registry was probed for paucisymptomatic individuals older than 70 years.ResultsWe identified 206 EGFr cysteine altering NOTCH3 mutations in ExAC, with a total prevalence of 3.4/1000. More than half of the distinct mutations have been previously reported in CADASIL patients. Despite the clear overlap, the mutation distribution in ExAC differs from that in reported CADASIL patients, as mutations in ExAC are predominantly located outside of EGFr domains 1–6. In an independent Dutch CADASIL cohort, we found that patients with a mutation in EGFr domains 7–34 have a significantly lower MRI lesion load than patients with a mutation in EGFr domains 1–6.InterpretationThe frequency of EGFr cysteine altering NOTCH3 mutations is 100‐fold higher than expected based on estimates of CADASIL prevalence. This challenges the current CADASIL disease paradigm, and suggests that certain mutations may more frequently cause a much milder phenotype, which may even go clinically unrecognized. Our data suggest that individuals with a mutation located in EGFr domains 1–6 are predisposed to the more severe “classical” CADASIL phenotype, whereas individuals with a mutation outside of EGFr domains 1–6 can remain paucisymptomatic well into their eighth decade.

Clinical features and mutation spectrum in Chinese patients with CADASIL: A multicenter retrospective study

“…In our Han Chinese cohort, which contained patients largely from eastern and southern China, the exon 11 was considered to be a “hot region,” in line with previous reports . Interestingly, a recent research based on ExAC database indicated that mutations causing CADASIL revealed an obvious distribution discrepancy and predominately located outside EGFR 1‐6 . Nevertheless, the mutation p.Arg607Cys, rather than p.Arg544Cys, ranked in the first place of the mutational spectrum in our subjects and occupied a ratio of nearly 20%.…”

“…9,10 Interestingly, a recent research based on ExAC database indicated that mutations causing CADASIL revealed an obvious distribution discrepancy and predominately located outside EGFR 1-6. 22 Nevertheless, the mutation p.Arg607Cys, rather than p.Arg544Cys, ranked in the first place of the mutational spectrum in our subjects and occupied a ratio of nearly 20%. scores.…”

Estimation of minimal disease prevalence from population genomic data: Application to primary familial brain calcification

“…In CADASIL, another dominantly‐inherited microvascular disorder of the brain, the burden of EGFr cysteine‐altering NOTCH3 variants in ExAC data—the type of variants known to be pathogenic or probably pathogenic in patients with CADASIL—was surprisingly 100‐fold higher than prevalence estimations of 2–5 p. 100,000 in epidemiological studies (Moreton, Razvi, Davidson, & Muir, ; Narayan, Gorman, Kalaria, Ford, & Chinnery, ; Razvi, Davidson, Bone, & Muir, ; Rutten et al, ). Observing a different distribution of these variants in the EGFr domains of NOTCH3 in ExAC compared to Dutch patients with CADASIL, the authors hypothesized that variants located in certain EGFr domains could have a milder effect leading to incomplete penetrance (Rutten et al, ). Conversely, PKD2 ‐Related Autosomal Dominant Polycystic Kidney Disease prevalence estimates was 0.63 (95%CI, 0.54–0.72) per 10,000 in the population of Brittany based on patients included in a cohort and 2.31 (95%CI, 1.10–3.51) per 10,000 in ExAC (Cornec‐Le Gall et al, ), revealing a much lower difference between the two estimations than for CADASIL.…”

“…Given the fact that (i) a significant proportion of PFBC patients do not exhibit neuropsychiatric symptoms; (ii) most of the symptomatic patients have a disease onset during adulthood; (iii) the few patients with symptomatic PFBC starting during childhood rarely present severe symptoms; and (iv) the disease penetrance is full when considering abnormal brain calcification as the major diagnostic criterion, it is reasonable to estimate a minimal prevalence from genomic data. The power of genome and exome sequencing in large populations gathered by ExAC makes this strategy now applicable through access to publicly available data (Appadurai et al, ; Cornec‐Le Gall et al, ; Minikel et al, ; Rutten et al, ; Sleat, Gedvilaite, Zhang, Lobel, & Xing, ).…”

Cerebrovascular disorders associated with genetic lesions

“…To date, more than 500 families with the disease have been identified worldwide (the majority in Europe), with more than 230 unique mutations reported (Human Genetic Mutation Database website, www.hgmd.cf.ac.uk), but a clear genotype-phenotype correlation has still to emerge [196, 197]. Recently, mutations mapping to EGF repeats 1-6 have been associated with higher lesion load on MRI, possibly predisposing to a more severe form of the disease [198]. …”