Lessons learned from unsolicited findings in clinical exome sequencing of 16,482 individuals.

Unsolicited findings (UFs) are uncovered unintentionally and predispose to a disease unrelated to the clinical question. The frequency and nature of UFs uncovered in clinical practice remain largely unexplored. We here evaluated UFs identified during a 5-year period in which 16,482 index patients received clinical whole-exome sequencing (WES). UFs were identified in 0.58% (95/16,482) of index patients, indicating that the overall frequency of UFs in clinical WES is low. Fewer UFs were identified using restricted disease-gene panels (0.03%) than when using whole-exome/Mendeliome analysis (1.03%). The UF was disclosed to 86 of 95 individuals, for reasons of medical actionability. Only 61% of these UFs reside in a gene that is listed on the "ACMG59"-list, representing a list of 59 genes for which the American College of Medical Genetics recommends UF disclosure. The remaining 39% were grouped into four categories: disorders similar to "ACMG59"-listed disorders (25%); disorders for which disease manifestation could be influenced (7%); UFs providing reproductive options (2%); and UFs with pharmacogenetic implications (5%). Hence, our experience shows that UFs predisposing to medically actionable disorders affect a broader range of genes than listed on the "ACMG59", advocating that a pre-defined gene list is too restrictive, and that UFs may require ad hoc evaluation of medical actionability. While both the identification and disclosure of UFs depend on local policy, our lessons learned provide general essential insight into the nature and odds of UFs in clinical exome sequencing.

Long-QT mutation p.K557E-Kv7.1: dominant-negative suppression of IKs, but preserved cAMP-dependent up-regulation.

AIMS: Mutations in KCNQ1, encoding for Kv7.1, the α-subunit of the IKs channel, cause long-QT syndrome type 1, potentially predisposing patients to ventricular tachyarrhythmias and sudden cardiac death, in particular, during elevated sympathetic tone. Here, we aim at characterizing the p.Lys557Glu (K557E) Kv7.1 mutation, identified in a Dutch kindred, at baseline and during (mimicked) increased adrenergic tone. METHODS AND RESULTS: K557E carriers had moderate QTc prolongation that augmented significantly during exercise. IKs characteristics were determined after co-expressing Kv7.1-wild-type (WT) and/or K557E with minK and Yotiao in Chinese hamster ovary cells. K557E caused IKs loss of function with slowing of the activation kinetics, acceleration of deactivation kinetics, and a rightward shift of voltage-dependent activation. Together, these contributed to a dominant-negative reduction in IKs density. Confocal microscopy and western blot indicated that trafficking of K557E channels was not impaired. Stimulation of WT IKs by 3'-5'-cyclic adenosine monophosphate (cAMP) generated strong current up-regulation that was preserved for K557E in both hetero- and homozygosis. Accumulation of IKs at fast rates occurred both in WT and in K557E, but was blunted in the latter. In a computational model, K557E showed a loss of action potential shortening during ß-adrenergic stimulation, in accordance with the lack of QT shortening during exercise in patients. CONCLUSION: K557E causes IKs loss of function with reduced fast rate-dependent current accumulation. cAMP-dependent stimulation of mutant IKs is preserved, but incapable of fully compensating for the baseline current reduction, explaining the long QT intervals at baseline and the abnormal QT accommodation during exercise in affected patients.

The unfolding clinical spectrum of holoprosencephaly due to mutations in SHH, ZIC2, SIX3 and TGIF genes.

Holoprosencephaly is a severe malformation of the brain characterized by abnormal formation and separation of the developing central nervous system. The prevalence is 1:250 during early embryogenesis, the live-born prevalence is 1:16 000. The etiology of HPE is extremely heterogeneous and can be teratogenic or genetic. We screened four known HPE genes in a Dutch cohort of 86 non-syndromic HPE index cases, including 53 family members. We detected 21 mutations (24.4%), 3 in SHH, 9 in ZIC2 and 9 in SIX3. Eight mutations involved amino-acid substitutions, 7 ins/del mutations, 1 frame-shift, 3 identical poly-alanine tract expansions and 2 gene deletions. Pathogenicity of mutations was presumed based on de novo character, predicted non-functionality of mutated proteins, segregation of mutations with affected family-members or combinations of these features. Two mutations were reported previously. SNP array confirmed detected deletions; one spanning the ZIC2/ZIC5 genes (approx. 100 kb) the other a 1.45 Mb deletion including SIX2/SIX3 genes. The mutation percentage (24%) is comparable with previous reports, but we detected significantly less mutations in SHH: 3.5 vs 10.7% (P=0.043) and significantly more in SIX3: 10.5 vs 4.3% (P=0.018). For TGIF1 and ZIC2 mutation the rate was in conformity with earlier reports. About half of the mutations were de novo, one was a germ line mosaic. The familial mutations displayed extensive heterogeneity in clinical manifestation. Of seven familial index patients only two parental carriers showed minor HPE signs, five were completely asymptomatic. Therefore, each novel mutation should be considered as a risk factor for clinically manifest HPE, with the caveat of reduced clinical penetrance.