Detecting Low-Variant Allele Frequency Mosaic Pathogenic Variants of NF1, TSC2, and AKT3 Genes from Blood in Patients with Neurodevelopmental Disorders

Growing evidence indicates that early and late postzygotic mosaicism can cause neurodevelopmental disorders (NDDs), but detection of low variant allele frequency (VAF) mosaic variants from blood remains a challenge. Data of 2162 patients with NDDs who underwent conventional genetic tests were reviewed and a deep sequencing was performed using a specifically designed mosaic next-generation sequencing (NGS) panel in the patients with negative genetic test results. Forty-four patents with neurocutaneous syndrome, malformation of cortical development, or nonlesional epileptic encephalopathies were included. In total, mosaic variants were detected from blood in 1.2% (25/2162) of the patients. Using conventional NGS panels, 22 mosaic variants (VAF, 8.8% to 29.8%) were identified in 18 different genes. Using a specifically designed mosaicism NGS panel, three mosaic variants of the NF1, TSC2, and AKT3 genes were identified (VAF, 2.0% to 11.2%). Mosaic variants were found frequently in the patients who had neurocutaneous syndrome (2/7, 28.6%), whereas only one or no mosaic variant was detected for patients who had malformations of cortical development (1/20, 5%) or nonlesional epileptic encephalopathies (0%, 0/17). In summary, mosaic variants that contribute to the spectrum of NDDs can be detected from blood via conventional NGS and specifically designed mosaicism NGS panels, and detection of mosaic variants using blood will increase diagnostic yield. Growing evidence indicates that early and late postzygotic mosaicism can cause neurodevelopmental disorders (NDDs), but detection of low variant allele frequency (VAF) mosaic variants from blood remains a challenge. Data of 2162 patients with NDDs who underwent conventional genetic tests were reviewed and a deep sequencing was performed using a specifically designed mosaic next-generation sequencing (NGS) panel in the patients with negative genetic test results. Forty-four patents with neurocutaneous syndrome, malformation of cortical development, or nonlesional epileptic encephalopathies were included. In total, mosaic variants were detected from blood in 1.2% (25/2162) of the patients. Using conventional NGS panels, 22 mosaic variants (VAF, 8.8% to 29.8%) were identified in 18 different genes. Using a specifically designed mosaicism NGS panel, three mosaic variants of the NF1, TSC2, and AKT3 genes were identified (VAF, 2.0% to 11.2%). Mosaic variants were found frequently in the patients who had neurocutaneous syndrome (2/7, 28.6%), whereas only one or no mosaic variant was detected for patients who had malformations of cortical development (1/20, 5%) or nonlesional epileptic encephalopathies (0%, 0/17). In summary, mosaic variants that contribute to the spectrum of NDDs can be detected from blood via conventional NGS and specifically designed mosaicism NGS panels, and detection of mosaic variants using blood will increase diagnostic yield. The term neurodevelopmental disorders (NDDs) applies to a group of disorders that affect cognitive and social communicative development. According to the International Classification of Diseases, 11th Revision, disorders of intellectual development, autism spectrum disorders, attention-deficit/hyperactivity disorder, developmental motor coordination disorder, speech or language disorders, learning disorders, and stereotyped movement disorders are classified as NDDs (https://icd.who.int/browse11/l-m/en#/http://id.who.int/icd/entity/1516623224, last accessed April 5, 2023). Neurologic conditions, including epilepsy and malformations of cortical development (MCDs), are common comorbidities. Recent studies have emphasized the role of de novo variants in NDDs. Haploinsufficiency of genes involved in conserved pathways can disrupt protein synthesis, synaptic signaling, and transcriptional or epigenetic regulation. Although de novo variants are the most commonly identified cause of NDDs, the detection rate for de novo variants associated with NDDs has been approximately 30% in large-size studies.1Lindy A.S. Stosser M.B. Butler E. Downtain-Pickersgill C. Shanmugham A. Retterer K. Brandt T. Richard G. McKnight D.A. Diagnostic outcomes for genetic testing of 70 genes in 8565 patients with epilepsy and neurodevelopmental disorders.Epilepsia. 2018; 59: 1062-1071Crossref PubMed Scopus (173) Google Scholar, 2Symonds J.D. Zuberi S.M. Stewart K. McLellan A. O'Regan M. MacLeod S. Jollands A. Joss S. Kirkpatrick M. Brunklaus A. Pilz D.T. Shetty J. Dorris L. Abu-Arafeh I. Andrew J. Brink P. Callaghan M. Cruden J. Diver L.A. Findlay C. Gardiner S. Grattan R. Lang B. MacDonnell J. McKnight J. Morrison C.A. Nairn L. Slean M.M. Stephen E. Webb A. Vincent A. Wilson M. Incidence and phenotypes of childhood-onset genetic epilepsies: a prospective population-based national cohort.Brain. 2019; 142: 2303-2318Crossref PubMed Scopus (202) Google Scholar, 3Symonds J.D. McTague A. Epilepsy and developmental disorders: next generation sequencing in the clinic.Eur J Paediatr Neurol. 2020; 24: 15-23Abstract Full Text Full Text PDF PubMed Scopus (75) Google Scholar Until recently, most identified variants were thought to be de novo heterozygous variants. However, growing evidence has indicated that mosaic variants from the brain can also cause NDDs.4Pirozzi F. Berkseth M. Shear R. Gonzalez L. Timms A.E. Sulc J. Pao E. Oyama N. Forzano F. Conti V. Guerrini R. Doherty E.S. Saitta S.C. Lockwood C.M. Pritchard C.C. Dobyns W.B. Novotny E. Wright J.N.N. Saneto R.P. Friedman S. Hauptman J. Ojemann J. Kapur R.P. Mirzaa G.M. Profiling PI3K-AKT-MTOR variants in focal brain malformations reveals new insights for diagnostic care.Brain. 2022; 145: 925-938Crossref PubMed Scopus (19) Google Scholar,5Tyburczy M.E. Dies K.A. Glass J. Camposano S. Chekaluk Y. Thorner A.R. Lin L. Krueger D. Franz D.N. Thiele E.A. Sahin M. Kwiatkowski D.J. Mosaic and intronic mutations in TSC1/TSC2 explain the majority of TSC patients with no mutation identified by conventional testing.PLoS Genet. 2015; 11e1005637Crossref PubMed Scopus (189) Google Scholar A further review of 893 patients with pathogenic variants in one of nine common epilepsy-causing genes has revealed that 3.5% of such patients have mosaic variants.6Stosser M.B. Lindy A.S. Butler E. Retterer K. Piccirillo-Stosser C.M. Richard G. McKnight D.A. High frequency of mosaic pathogenic variants in genes causing epilepsy-related neurodevelopmental disorders.Genet Med. 2018; 20: 403-410Abstract Full Text Full Text PDF PubMed Scopus (115) Google Scholar Somatic mosaicism has been reported in NDDs and related disorders, including MCDs, epileptic encephalopathies, and intellectual disability.7Beltrán-Corbellini Á. Aledo-Serrano Á. Møller R.S. Pérez-Palma E. García-Morales I. Toledano R. Gil-Nagel A. Epilepsy genetics and precision medicine in adults: a new landscape for developmental and epileptic encephalopathies.Front Neurol. 2022; 13777115Crossref PubMed Scopus (15) Google Scholar, 8Cao Y. Tokita M.J. Chen E.S. Ghosh R. Chen T. Feng Y. Gorman E. Gibellini F. Ward P.A. Braxton A. Wang X. Meng L. Xiao R. Bi W. Xia F. Eng C.M. Yang Y. Gambin T. Shaw C. Liu P. Stankiewicz P. A clinical survey of mosaic single nucleotide variants in disease-causing genes detected by exome sequencing.Genome Med. 2019; 11: 48Crossref PubMed Scopus (51) Google Scholar, 9D'Gama A.M. Walsh C.A. Somatic mosaicism and neurodevelopmental disease.Nat Neurosci. 2018; 21: 1504-1514Crossref PubMed Scopus (158) Google Scholar With the recent development of sensitive tools, including droplet digital PCR, and an increasing depth of coverage, additional mosaic variants associated with NDDs are expected to be identified,10`Mefford H.C. Mosaicism in clinical genetics.Cold Spring Harb Mol Case Stud. 2021; 7a006162PubMed Google Scholar but detection of early mosaic pathogenic variants from blood remains a challenge. Data regarding the frequencies of early mosaic variants in NDDs detected from blood remain limited. Here, early mosaicism related to NDDs was investigated in detail. Cases with a high variant allele frequency (VAF; 5% to 30%) were identified using a conventional next-generation sequencing (NGS) panel. Cases with a low VAF (<5%) were further investigated by deep sequencing using a customized mosaicism-specific panel and a bioinformatic pipeline for accurate detection of low-level mosaicisms. We hypothesized that mosaic variants with a low VAF in blood can be detected if a targeted NGS panel with high sensitivity is used. Overall, 2162 patients with NDDs, epilepsy, and MCDs underwent the authors’ conventional genetic tests (targeted NGS gene panel testing or Sanger sequencing) between 2016 and 2021. Among patients tested, epilepsy and delayed development were the most common diagnoses. A specific target gene panel was ordered by the primary physician based on the clinical diagnosis. The authors retrospectively reviewed the genetic test results of these patients and identified mosaic variants with an alternative allele frequency <30%. These mosaic variants had a frequency of 5% to 30% and were categorized into high-grade mosaicism. Then, a subset of patients with unidentified genetic etiology was selected for further mosaicism testing. Only patients who agreed to undergo further genetic testing by using already available DNA samples were included. The previous conventional genetic test results of all of the patients were negative. Patients were selected from the following three groups: infantile epileptic spasm syndrome (IESS), neurocutaneous syndrome, and MCD. For patients with MCD or neurocutaneous syndrome, those with multiple or diffuse lesions that infiltrate an extended area of the body were specifically selected. Patients who had undergone epilepsy surgery were also selected because samples of their brain tissue were available. For patients with IESS, those with severely delayed development but no identified etiology were selected. Only patients with negative etiologic conclusion from analyses, such as magnetic resonance imaging, and with metabolic test results were included. Informed consent was obtained. Genomic DNA (gDNA) was extracted from peripheral blood samples and a formalin-fixed, paraffin-embedded brain sample by using the QIAamp Blood Mini Kit (Qiagen, Hilden, Germany), according to the manufacturer's guidelines. For the conventional target panel testing, the library preparation and hybrid capture–based target enrichment using panels designed for epilepsy, MCD, and NDD were performed as previously described.11Rim J.H. Kim S.H. Hwang I.S. Kwon S.S. Kim J. Kim H.W. Cho M.J. Ko A. Youn S.E. Kim J. Lee Y.M. Chung H.J. Lee J.S. Kim H.D. Choi J.R. Lee S.T. Kang H.C. Efficient strategy for the molecular diagnosis of intractable early-onset epilepsy using targeted gene sequencing.BMC Med Genomics. 2018; 11: 6Crossref PubMed Scopus (42) Google Scholar,12Kim S.H. Kim B. Lee J.S. Kim H.D. Choi J.R. Lee S.T. Kang H.C. Proband-only clinical exome sequencing for neurodevelopmental disabilities.Pediatr Neurol. 2019; 99: 47-54Abstract Full Text Full Text PDF PubMed Scopus (17) Google Scholar Read depth varied depending on the used panel size, and all panels achieved a minimum mean depth of 100× with a minimum coverage of 30×. To reduce false variant calls, the threshold for VAF was set to 5%. For the mosaicism-specific target panel testing based on hybrid capture approach, to reduce the frequency of false-positive variants, gDNA was treated with uracil-DNA glycosylase (NEB, Ipswich, MA) by following the manufacturer's protocol. Briefly, 100 ng gDNA was incubated at 37°C with 1 unit uracil-DNA glycosylase for 30 minutes in a final volume of 50 μL and then purified using AmpureXP beads (Beckman Coulter Inc., Brea, CA). The purified gDNA was quantified using the Qubit BR dsDNA kit (Invitrogen, Carlsbad, CA). After uracil-DNA glycosylase treatment, approximately 100 ng gDNA was prepared using the Twist Library Preparation EF Kit (Twist Bioscience, San Francisco, CA). The enzymatic fragmentation of gDNA was performed at 32°C for 20 minutes, followed by enzyme inactivation at 65°C for 30 minutes. Hybrid capture–based target enrichment was performed using a custom-design enrichment panel following the manufacturer's instructions (Dxome, Seoul, Republic of Korea). The panel specially designed for mosaicism included 71 genes (AKT3, ARX, DCX, DEPDC5, MTOR, PAFAH1B1, PIK3CA, PIK3R2, TUBA1A, TUBB2B, TUBB3, CHD7, TUBA8, ALG13, CACNA1A, CDKL5, CHD2, DNM1, FOXG1, GABRA1, GABRB3, GNAO1, GRIN2A, GRIN2B, IQSEC2, HNRNPU, KCNT1, PCDH19, PTEN, SCN1A, SCN2A, SCN8A, SLC35A2, SPTAN1, STXBP1, ANKRD11, DNM1L, DYNC1H1, EEF1A2, KCNB1, KCNQ2, MECP2, SLC2A1, SLC9A6, SYNGAP1, GNAQ, NF1, TSC1, TSC2, KANSL1, KDM5C, PDHA1, ARID1B, CTNNB1, DYRK1A, POGZ, WDR45, ATP1A3, ARHGEF9, CACNA1E, DEAF1, GLRA1, MBD5, NSD1, PLP1, PRRT2, PURA, SHANK3, SMARCA2, UBE3A, and ZEB2). This gene panel consisted of commonly identified genes related to NDDs, with a focus on IESS, neurocutaneous syndrome, and MCD through the review of literature and selection of genes by pediatric neurologists and geneticists.8Cao Y. Tokita M.J. Chen E.S. Ghosh R. Chen T. Feng Y. Gorman E. Gibellini F. Ward P.A. Braxton A. Wang X. Meng L. Xiao R. Bi W. Xia F. Eng C.M. Yang Y. Gambin T. Shaw C. Liu P. Stankiewicz P. A clinical survey of mosaic single nucleotide variants in disease-causing genes detected by exome sequencing.Genome Med. 2019; 11: 48Crossref PubMed Scopus (51) Google Scholar,13Desikan R.S. Barkovich A.J. Malformations of cortical development.Ann Neurol. 2016; 80: 797-810Crossref PubMed Scopus (79) Google Scholar, 14Gürsoy S. Erçal D. Genetic evaluation of common neurocutaneous syndromes.Pediatr Neurol. 2018; 89: 3-10Abstract Full Text Full Text PDF PubMed Scopus (12) Google Scholar, 15He N. Lin Z.J. Wang J. Wei F. Meng H. Liu X.R. Chen Q. Su T. Shi Y.W. Yi Y.H. Liao W.P. Evaluating the pathogenic potential of genes with de novo variants in epileptic encephalopathies.Genet Med. 2019; 21: 17-27Abstract Full Text Full Text PDF PubMed Scopus (37) Google Scholar, 16Vissers L.E. Gilissen C. Veltman J.A. Genetic studies in intellectual disability and related disorders.Nat Rev Genet. 2016; 17: 9-18Crossref PubMed Scopus (501) Google Scholar, 17Parrini E. Conti V. Dobyns W.B. Guerrini R. Genetic basis of brain malformations.Mol Syndromol. 2016; 7: 220-233Crossref PubMed Scopus (131) Google Scholar, 18Hully M. Ropars J. Hubert L. Boddaert N. Rio M. Bernardelli M. Desguerre I. Cormier-Daire V. Munnich A. de Lonlay P. Reilly L. Besmond C. Bahi-Buisson N. Mosaicism in ATP1A3-related disorders: not just a theoretical risk.Neurogenetics. 2017; 18: 23-28Crossref PubMed Scopus (28) Google Scholar The target-enriched DNA libraries were sequenced using a NextSeq 550Dx System or NovaSeq 6000 instrument (Illumina, San Diego, CA), whereby the authors achieved approximately 150 million reads per sample and targeted at least >10,000× mean depth per sample. A 151-bp, dual-indexed, paired-end sequencing configuration was used. For further processing, the sequencing data were analyzed using the authors’ bioinformatic pipeline, as previously described.11Rim J.H. Kim S.H. Hwang I.S. Kwon S.S. Kim J. Kim H.W. Cho M.J. Ko A. Youn S.E. Kim J. Lee Y.M. Chung H.J. Lee J.S. Kim H.D. Choi J.R. Lee S.T. Kang H.C. Efficient strategy for the molecular diagnosis of intractable early-onset epilepsy using targeted gene sequencing.BMC Med Genomics. 2018; 11: 6Crossref PubMed Scopus (42) Google Scholar,12Kim S.H. Kim B. Lee J.S. Kim H.D. Choi J.R. Lee S.T. Kang H.C. Proband-only clinical exome sequencing for neurodevelopmental disabilities.Pediatr Neurol. 2019; 99: 47-54Abstract Full Text Full Text PDF PubMed Scopus (17) Google Scholar For samples tested by the mosaicism-specific panel, to accurately detect low-frequency variants, a positional indexing sequencing algorithm (Dxome) was used to call single-nucleotide variants and small insertions/deletions. The positional indexing is a modified molecular barcoding method without the use of unique molecular identifiers, which uses genomic position of the mapped reads as unique identifiers instead. Variants present in all reads of specific positional groups were regarded as true positive, and errors that occurred during amplification and sequencing could be successfully filtered, thereby improving the detection accuracy.19Lee K.S. Seo J. Lee C.K. Shin S. Choi Z. Min S. Yang J.H. Kwon W.S. Yun W. Park M.R. Choi J.R. Chung H.C. Lee S.T. Rha S.Y. Analytical and clinical validation of cell-free circulating tumor DNA assay for the estimation of tumor mutational burden.Clin Chem. 2022; 68: 1519-1528Crossref PubMed Scopus (6) Google Scholar,20Kim N. Kim Y.N. Lee K. Park E. Lee Y.J. Hwang S.Y. Park J. Choi Z. Kim S.W. Kim S. Choi J.R. Lee S.T. Lee J.Y. Feasibility and clinical applicability of genomic profiling based on cervical smear samples in patients with endometrial cancer.Front Oncol. 2022; 12942735Google Scholar Variants were classified into five categories (benign, likely benign, variant of uncertain significance, likely pathogenic, and pathogenic) based on the recommendation of the American College of Medical Genetics and Genomics.21Richards S. Aziz N. Bale S. Bick D. Das S. Gastier-Foster J. Grody W.W. Hegde M. Lyon E. Spector E. Voelkerding K. Rehm H.L. ACMG Laboratory Quality Assurance CommitteeStandards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology.Genet Med. 2015; 17: 405-424Abstract Full Text Full Text PDF PubMed Scopus (17560) Google Scholar Population frequency data from multiple databases [namely, 1000 Genomes (https://www.internationalgenome.org/data), the Genome Aggregation Database (https://gnomad.broadinstitute.org), the Exome Sequencing Project (https://evs.gs.washington.edu/EVS), and the Korean Reference Genome Database (http://coda.nih.go.kr/coda/KRGDB/index.jsp); all last accessed January 2, 2023] were used to interpret the detected variants. To annotate the variants, the authors searched and reviewed reports from ClinVar and the Human Gene Mutation Database. In silico analyses were conducted using SIFT, MutationTaster, FATHMM, and MetaSVM (http://database.liulab.science/dbNSFP, last accessed April 5, 2023). Finally, the clinical impact of the variants with genotype-phenotype correlations was reviewed by pediatric neurologists (Se Hee Kim, J.S.L., H.D.K., H.-C.K.) and geneticists (S.S.K., J.R.C., S.-T.L.). If needed, peripheral blood samples were obtained from parents to confirm the pathogenicity of the variant. Ideally, the alternative allele frequency of a heterozygous germline variant is expected to be 50%. Mosaicism was suspected when the alternative allele frequency of a variant was significantly lower than the expected allele frequency for a heterozygous germline variant. To exclude experimental and analytic effects on allele frequency, the authors defined a variant as mosaic when its alternative allele frequency was <30%. To confirm the mosaic variants, variants with a VAF ≥10% were assessed via Sanger sequencing, followed by 3′-modified oligonucleotides PCR and parental genetic tests. Primers targeting the locus containing each variant of interest were designed, and subsequent gene amplification was performed. The amplicons were subjected to Sanger sequencing by using a 3730 DNA Analyzer with a BigDye Terminator v3.1 Cycle Sequencing kit (Applied Biosystems, Foster City, CA). Variants with a VAF <10%, which can be missed by Sanger sequencing, were confirmed via NGS (coverage, >10,000×) and mutant enrichment with 3′-modified oligonucleotides PCR followed by Sanger sequencing. Specific primers and blockers per variant of interest were designed for the 3′-modified oligonucleotides PCR, which was then performed as previously described.22Lee S.T. Kim J.Y. Kown M.J. Kim S.W. Chung J.H. Ahn M.J. Oh Y.L. Kim J.W. Ki C.S. Mutant enrichment with 3'-modified oligonucleotides a practical PCR method for detecting trace mutant DNAs.J Mol Diagn. 2011; 13: 657-668Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar If available, variants were also detected from brain tissues. Brain specimens were obtained from previous epilepsy surgeries. Parent testing was also performed. Sequence variants were interpreted according to the recent recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology.21Richards S. Aziz N. Bale S. Bick D. Das S. Gastier-Foster J. Grody W.W. Hegde M. Lyon E. Spector E. Voelkerding K. Rehm H.L. ACMG Laboratory Quality Assurance CommitteeStandards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology.Genet Med. 2015; 17: 405-424Abstract Full Text Full Text PDF PubMed Scopus (17560) Google Scholar This study was reviewed and approved by the Institutional Review Board of Yonsei University Health System (4-2021-0740 and 2022-0744-001). The study was conducted in accordance with good clinical practices [national regulations and International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) E6] and the principles of the Declaration of Helsinki. Written informed consent was obtained from the parents or legal guardians of the patients before sample collection following a detailed explanation of the schedules and contents of the study. Overall, 2162 patients with NDDs underwent target gene NGS panel testing or Sanger sequencing, and causative pathogenic or likely pathogenic variants were identified in 686 cases (diagnostic yield, 31.7%). Germline heterozygous variants were identified in 664 (30.7%) cases, and 22 cases had mosaic variants (3.2% of the positive cases, and 1.0% of the total cases). In cases where mosaic variants were identified, the average of total sequencing depth was 511 (range, 51 to 1695), and the VAF ranged from 8.8% to 29.8%. The genetic test results are shown in Figure 1. In total, 22 mosaic variants were detected in 18 different genes (namely, TSC2, DCX, SLC2A1, PCDH19, DNM1, STXBP1, SCN2A, SCN1A, PURA, POGZ, PAFAH1B1, NF1, KIF21A, KCNQ2, GABRA1, EEF1A2, CDKL5, and ARID1B). Four distinct mosaic variants were identified in TSC2, whereas two distinct mosaic variants were detected in GABRA1. The detailed information is provided in Table 1.Table 1Mosaic Pathogenic Variants Detected in Blood via Routine NGS Panel TestingAgeSexClinical diagnosisGeneAccession no.NucleotideAmino acidDe novo variant?Population frequency, %VAF, %InterpretationTotal reads (alternative reads)Clinical impact1 YearMTuberous sclerosisTSC2NM_000548.5c.610_611delp.Leu204AlafsTer30Unknown09.8LP1235 (121)GDx9 MonthsFTuberous sclerosisTSC2NM_000548.5c.976–15G>AUnknown017.4P69 (12)GDx6 MonthsMTuberous sclerosisTSC2NM_000548.5c.1358_1361+14delDe novo021.0P309 (65)GDx1 Year 2 monthsMTuberous sclerosisTSC2NM_000548.5c.5238_5255delp.His1746_Arg1751delUnknown012.7LP1405 (179)GDx11 YearsFEpilepsyDCXNM_178152.3c.636_644delp.Asp213_Thr215delUnknown012.7LP55 (7)GDx2 Years 11 monthsMEpilepsySLC2A1NM_006516.4c.275_275+17delUnknown014.7LP68 (10)GDx, Tx4 Years 4 monthsFEpilepsyPCDH19NM_001184880.2c.1793_1809delinsCGAp.Gly598AlafsTer14Unknown018.3LP197 (36)GDx7 MonthsFEpilepsyDNM1NM_001005336.3c.415_423delp.Gly139_Thr141delDe novo022.4LP330 (74)GDx9 YearsFLennox-Gastaut syndromeSTXBP1NM_003165.6c.923_933delp.Lys308IlefsTer2Unknown014.4LP180 (26)GDx2 MonthsFEpilepsySCN2ANM_001040142.2c.4499C>Tp.Ala1500ValDe novo015.7LP1191 (187)GDx, Tx28 YearsFEpilepsySCN1ANM_001165963.4c.4934delp.Arg1645GlnfsTer5De novo021.5P279 (60)GDx, Tx1 YearMNeonatal seizurePURANM_005859.5c.72dupp.Gly25ArgfsTer176Unknown023.5LP51 (12)GDx8 MonthsMDelayed developmentPOGZNM_015100.4c.2517_2518delp.His840GlnfsTer23Unknown026.2LP360 (94)GDx8 YearsMEpilepsyPAFAH1B1NM_000430.4c.1019G>Ap.Trp340TerUnknown011.0LP1692 (186)GDx2 Years 9 monthsMCafé-au-lait maculesNF1NM_001042492.3c.334C>Tp.Gln112TerDe novo017.9P151 (27)GDx4 Years 6 monthsFDelayed developmentKIF21ANM_001173464.2c.387dupp.His130ThrfsTer5Unknown08.8LP203 (18)GDx7 MonthsFEpilepsyKCNQ2NM_172107.4c.1687G>Ap.Asp563AsnUnknown026.6LP741 (197)GDx, Px6 Years 7 monthsFDelayed developmentGABRA1NM_000806.5c.839C>Tp.Pro280LeuDe novo029.8LP131 (39)GDx, Px1 YearMEpilepsyGABRA1NM_000806.5c.134T>Cp.Ile45ThrDe novo027.9LP466 (130)GDx, Px1 Year 1 monthMLennox-Gastaut syndromeEEF1A2NM_001958.5c.46G>Tp.Val16LeuDe novo022.8LP600 (137)GDx11 Years 4 monthsFLennox-Gastaut syndromeCDKL5NM_003159.2c.2684C>Tp.Pro895LeuDe novo0.00112.7LP1405 (179)GDx5 Years 1 monthFCoffin-Siris syndromeARID1BNM_001374820.1c.1638_1647dupp.Gln550GlyfsTer71Unknown024.2LP120 (29)GDxVariant positions were based on reference sequence. All NM_ accession numbers from NCBI Nucleotide (https://www.ncbi.nlm.nih.gov/nuccore, last accessed April 5, 2023).F, female; M, male; GDx, genetic diagnosis; LP, likely pathogenic; NGS, next-generation sequencing; P, pathogenic; Px, prognosis prediction; Tx, treatment adjustment; VAF, variant allele frequency. Open table in a new tab Variant positions were based on reference sequence. All NM_ accession numbers from NCBI Nucleotide (https://www.ncbi.nlm.nih.gov/nuccore, last accessed April 5, 2023). F, female; M, male; GDx, genetic diagnosis; LP, likely pathogenic; NGS, next-generation sequencing; P, pathogenic; Px, prognosis prediction; Tx, treatment adjustment; VAF, variant allele frequency. All of the patients had NDDs, including intellectual disabilities. Common comorbidities included epilepsy (16/22, 72.7%), tuberous sclerosis (4/22, 18.1%), and MCDs (2/22, 9.1%). The most frequent variation type was frameshift (8/22, 36.4%), followed by missense (6/22, 27.3%) and in-frame deletion (3/22, 13.6%). One variant was a pathogenic intronic variant, which resulted in aberrant splicing.23Mayer K. Ballhausen W. Leistner W. Rott H. Three novel types of splicing aberrations in the tuberous sclerosis TSC2 gene caused by mutations apart from splice consensus sequences.Biochim Biophys Acta. 2000; 1502: 495-507Crossref PubMed Scopus (43) Google Scholar Parental genetic tests were performed in 9 of 22 patients and revealed that the variants were de novo. For samples without identified causative variants with conventional NGS tests, mosaicism-specific panel tests were conducted. Overall, 48 samples from 44 patients with NDDs were included in the analysis. From all of the 44 patients, peripheral blood samples were obtained for analysis. In addition, brain samples from four of the patients were analyzed. Among the 44 patients, 20 had MCDs and 7 had neurocutaneous syndromes, including 3 patients with tuberous sclerosis complex, 2 patients with café-au-lait spots, and 2 patients with Sturge-Weber syndrome. Most patients (40/44) had epilepsy. Specifically, 17 patients had IESS, 11 patients had Lennox-Gastaut syndrome, and 1 patient had Doose syndrome. Deep sequencing with the mosaicism-specific panel yielded an average depth of 21,314× (range, 12,295× to 69,773×). By using the panel designed to detect mosaicism, three mosaic variants were identified from the blood samples of the patients. These variants were in three different genes (namely, NF1, TSC2, and AKT3; diagnostic yield, 7.3%) (Table 2). The mean read depth was 50,170× (range, 3615× to 130,792×), and the VAF of the detected variants ranged from 2.0% to 11.2%, indicating that variants with a low VAF could be detected via the mosaicism-specific target panel analysis. Overall, one missense, one intronic deletion, and one frameshift were detected.Table 2Mosaic Pathogenic Variants Detected in Blood via Mosaicism Panel TestingAgeSexComorbiditySampleGeneAccession no.NucleotideAmino acidVAF, %InterpretationCoverage2 YearsFNCSBloodNF1NM_001042492.3c.1527+4_1527+7del11.2P36239 Years 3 monthsMNCSBloodTSC2NM_000548.5c.4297_4298dupp.Gly1434ArgfsTer432.9LP134,6708 Years 2 monthsMMCDBloodAKT3NM_005465.7c.863C>Tp.Thr288Ile2.0LP20,696Variant positions were based on reference sequence. All NM_ accession numbers from NCBI Nucleotide (https://www.ncbi.nlm.nih.gov/nuccore, last accessed April 5, 2023).F, female; M, male; LP, likely pathogenic; MCD, malformation of cortical development; NCS, neurocutaneous syndrome; P, pathogenic; VAF, variant allele frequency. Open table in a new tab Variant positions were based on reference sequence. All NM_ accession numbers from NCBI Nucleotide (https://www.ncbi.nlm.nih.gov/nuccore, last accessed April 5, 2023). F, female; M, male; LP, likely pathogenic; MCD, malformation of cortical development; NCS, neurocutaneous syndrome; P, pathogenic; VAF, variant allele frequency. Regarding etiology, in neurocutaneous syndrome, the diagnostic yield was high (2/7, 28.6%). In MCD, the diagnostic yield was low (5%, 1/20). No mosaic variant was detected among the 17 patients who had IESS (0%, 0/17). For neurocutaneous syndrome, first, a pathogenic NF1 variant with a VAF of 11.2% was detected in a patient with multiple café-au-lait macules. Second, a frameshift variant of TSC2 with a VAF of 2.9% was detected in a patient diagnosed with tuberous sclerosis. The latter patient had seizures since the age of 6 years. The patient was diagnosed with tuberous sclerosis at the age of 7 years via imaging studies, including brain magnetic resonance imaging and kidney sonography, which revealed multifocal tubers in addition to calcified subependymal nodules, and angiomyolipoma, respectively. Previous evaluations of the TSC1 and TSC2 genes via Sanger sequencing yielded negative results. Regarding MCD, a mosaic variant of AKT3 (c.863C>T, p.Thr288Ile, with a VAF of 2.0%) was found in a patient. This patient had dr