Use of next-generation sequencing to detect LDLR gene copy number variation in familial hypercholesterolemia

Familial hypercholesterolemia (FH) is a heritable condition of severely elevated LDL cholesterol, caused predominantly by autosomal codominant mutations in the LDL receptor gene (LDLR). In providing a molecular diagnosis for FH, the current procedure often includes targeted next-generation sequencing (NGS) panels for the detection of small-scale DNA variants, followed by multiplex ligation-dependent probe amplification (MLPA) in LDLR for the detection of whole-exon copy number variants (CNVs). The latter is essential because ∼10% of FH cases are attributed to CNVs in LDLR; accounting for them decreases false negative findings. Here, we determined the potential of replacing MLPA with bioinformatic analysis applied to NGS data, which uses depth-of-coverage analysis as its principal method to identify whole-exon CNV events. In analysis of 388 FH patient samples, there was 100% concordance in LDLR CNV detection between these two methods: 38 reported CNVs identified by MLPA were also successfully detected by our NGS method, while 350 samples negative for CNVs by MLPA were also negative by NGS. This result suggests that MLPA can be removed from the routine diagnostic screening for FH, significantly reducing associated costs, resources, and analysis time, while promoting more widespread assessment of this important class of mutations across diagnostic laboratories. Familial hypercholesterolemia (FH) is a heritable condition of severely elevated LDL cholesterol, caused predominantly by autosomal codominant mutations in the LDL receptor gene (LDLR). In providing a molecular diagnosis for FH, the current procedure often includes targeted next-generation sequencing (NGS) panels for the detection of small-scale DNA variants, followed by multiplex ligation-dependent probe amplification (MLPA) in LDLR for the detection of whole-exon copy number variants (CNVs). The latter is essential because ∼10% of FH cases are attributed to CNVs in LDLR; accounting for them decreases false negative findings. Here, we determined the potential of replacing MLPA with bioinformatic analysis applied to NGS data, which uses depth-of-coverage analysis as its principal method to identify whole-exon CNV events. In analysis of 388 FH patient samples, there was 100% concordance in LDLR CNV detection between these two methods: 38 reported CNVs identified by MLPA were also successfully detected by our NGS method, while 350 samples negative for CNVs by MLPA were also negative by NGS. This result suggests that MLPA can be removed from the routine diagnostic screening for FH, significantly reducing associated costs, resources, and analysis time, while promoting more widespread assessment of this important class of mutations across diagnostic laboratories. Familial hypercholesterolemia (FH) is an inherited dyslipidemia characterized by a lifelong exposure to elevated LDL cholesterol levels with increased risk of premature atherosclerosis causing CVD, particularly coronary heart disease (CHD) (1Nordestgaard B.G. Chapman M.J. Humphries S.E. Ginsberg H.N. Masana L. Descamps O.S. Wiklund O. Hegele R.A. Raal F.J. Defesche J.C. et al.Familial hypercholesterolaemia is underdiagnosed and undertreated in the general population: guidance for clinicians to prevent coronary heart disease: consensus statement of the European atherosclerosis society.Eur. Heart J. 2013; 34: 3478-3490Crossref PubMed Scopus (1787) Google Scholar). Although FH has multiple genetic etiologies, ∼90% of molecularly defined cases result from autosomal codominant mutations in the LDL receptor gene (LDLR) (1Nordestgaard B.G. Chapman M.J. Humphries S.E. Ginsberg H.N. Masana L. Descamps O.S. Wiklund O. Hegele R.A. Raal F.J. Defesche J.C. et al.Familial hypercholesterolaemia is underdiagnosed and undertreated in the general population: guidance for clinicians to prevent coronary heart disease: consensus statement of the European atherosclerosis society.Eur. Heart J. 2013; 34: 3478-3490Crossref PubMed Scopus (1787) Google Scholar, 2Usifo E. Leigh S.E.A. Whittall R.A. Lench N. Taylor A. Yeats C. Orengo C.A. Martin A.C.R. Celli J. Humphries S.E. Low-density lipoprotein receptor gene familial hypercholesterolemia variant database: update and pathological assessment.Ann. Hum. Genet. 2012; 76: 387-401Crossref PubMed Scopus (163) Google Scholar). Pathogenic loss-of-function variants in LDLR affect every functional domain of the encoded protein and include a spectrum of both point mutations and large-scale deletions or duplications spanning whole exons—known as copy number variants (CNVs) (3Pollex R.L. Hegele R.A. Copy number variation in the human genome and its implications for cardiovascular disease.Circulation. 2007; 115: 3130-3138Crossref PubMed Scopus (61) Google Scholar). Autosomal codominant FH may also occasionally be caused by specific protein-altering mutations in the apo B gene (APOB) or by gain-of-function mutations in the proprotein convertase subtilisin/kexin type 9 gene (PCSK9). Heterozygous FH (HeFH) has been shown to be more common than previously thought: recent population-based studies in the United Kingdom (4Wald D.S. Bestwick J.P. Morris J.K. Whyte K. Jenkins L. Wald N.J. Child-parent familial hypercholesterolemia screening in primary care.N. Engl. J. Med. 2016; 375: 1628-1637Crossref PubMed Scopus (194) Google Scholar), the Netherlands (5Sjouke B. Kusters D.M. Kindt I. Besseling J. Defesche J.C. Sijbrands E.J.G. Roeters van Lennep J.E. Stalenhoef A.F.H. Wiegman A. de Graaf J. et al.Homozygous autosomal dominant hypercholesterolaemia in the Netherlands: prevalence, genotype-phenotype relationship, and clinical outcome.Eur. Heart J. 2015; 36: 560-565Crossref PubMed Scopus (320) Google Scholar), Northern Europe (6Benn M. Watts G.F. Tybjærg-Hansen A. Nordestgaard B.G. Mutations causative of familial hypercholesterolaemia: screening of 98 098 individuals from the Copenhagen General Population Study estimated a prevalence of 1 in 217.Eur. Heart J. 2016; 37: 1384-1394Crossref PubMed Scopus (280) Google Scholar), Poland (7Pajak A. Szafraniec K. Polak M. Drygas W. Piotrowski W. Zdrojewski T. Jankowski P. Prevalence of familial hypercholesterolemia: a meta-analysis of six large, observational, population-based studies in Poland.Arch. Med. Sci. 2016; 12: 687-696Crossref PubMed Scopus (38) Google Scholar), and the United States (8Abul-Husn N.S. Manickam K. Jones L.K. Wright E.A. Hartzel D.N. Gonzaga-Jauregui C. O'Dushlaine C. Leader J.B. Lester Kirchner H. Lindbuchler D.M. et al.Genetic identification of familial hypercholesterolemia within a single US health care system.Science. 2016; 354: aaf7000Crossref PubMed Scopus (250) Google Scholar) suggest that HeFH affects ∼1 in 250 individuals in the general population. Furthermore, the prevalence in certain founder populations is even higher, such as ∼1 in 200 in French Canadians, ∼1 in 165 in Tunisians, ∼1 in 85 in Christian Lebanese, and ∼1 in 72 in South African Afrikaners (9Austin M.A. Hutter C.M. Zimmern R.L. Humphries S.E. Genetic causes of monogenic heterozygous familial hypercholesterolemia: a HuGE prevalence review.Am. J. Epidemiol. 2004; 160: 407-420Crossref PubMed Scopus (478) Google Scholar). Homozygous FH (HoFH) is rare, with an incidence of ∼1 in 160,000 to 1 in 300,000 (5Sjouke B. Kusters D.M. Kindt I. Besseling J. Defesche J.C. Sijbrands E.J.G. Roeters van Lennep J.E. Stalenhoef A.F.H. Wiegman A. de Graaf J. et al.Homozygous autosomal dominant hypercholesterolaemia in the Netherlands: prevalence, genotype-phenotype relationship, and clinical outcome.Eur. Heart J. 2015; 36: 560-565Crossref PubMed Scopus (320) Google Scholar). By using these prevalence figures, there are ∼34 million individuals globally with FH; however, <1% have been diagnosed, although detection rates vary widely by country (10Nordestgaard B.G. Benn M. Genetic testing for familial hypercholesterolaemia is essential in individuals with high LDL cholesterol: who does it in the world?.Eur. Heart J. 2017; 38: 1580-1583Crossref PubMed Scopus (49) Google Scholar). Early identification and treatment of FH patients is essential, as lipid-lowering therapies have been proven to reduce LDL cholesterol levels to those of the general population (1Nordestgaard B.G. Chapman M.J. Humphries S.E. Ginsberg H.N. Masana L. Descamps O.S. Wiklund O. Hegele R.A. Raal F.J. Defesche J.C. et al.Familial hypercholesterolaemia is underdiagnosed and undertreated in the general population: guidance for clinicians to prevent coronary heart disease: consensus statement of the European atherosclerosis society.Eur. Heart J. 2013; 34: 3478-3490Crossref PubMed Scopus (1787) Google Scholar). Recently, FH has progressed toward the forefront of precision medicine as patients worldwide are more commonly offered genetic testing in diagnosis (1Nordestgaard B.G. Chapman M.J. Humphries S.E. Ginsberg H.N. Masana L. Descamps O.S. Wiklund O. Hegele R.A. Raal F.J. Defesche J.C. et al.Familial hypercholesterolaemia is underdiagnosed and undertreated in the general population: guidance for clinicians to prevent coronary heart disease: consensus statement of the European atherosclerosis society.Eur. Heart J. 2013; 34: 3478-3490Crossref PubMed Scopus (1787) Google Scholar, 11Iacocca M.A. Hegele R.A. Recent advances in genetic testing for familial hypercholesterolemia.Expert Rev. Mol. Diagn. 2017; 17: 641-651Crossref PubMed Scopus (42) Google Scholar). Advantages of genetic testing for FH are manifold. They include: 1) achieving certainty in the context of incomplete clinical criteria, such as family history or typical physical findings; 2) motivating cascade screening of family members; 3) initiating genotype-directed treatment strategies (12Raal F.J. Honarpour N. Blom D.J. Hovingh G.K. Xu F. Scott R. Wasserman S.M. Stein E.A. TESLA Investigators Inhibition of PCSK9 with evolocumab in homozygous familial hypercholesterolaemia (TESLA Part B): a randomised, double-blind, placebo-controlled trial.Lancet. 2015; 385: 341-350Abstract Full Text Full Text PDF PubMed Scopus (547) Google Scholar); and 4) supporting insurance coverage of certain medications. Successful molecular diagnosis depends on the ability of the designated method to assess both locus and allele heterogeneity associated with FH (11Iacocca M.A. Hegele R.A. Recent advances in genetic testing for familial hypercholesterolemia.Expert Rev. Mol. Diagn. 2017; 17: 641-651Crossref PubMed Scopus (42) Google Scholar). The cost-effectiveness of such methods may limit their widespread implementation and routine use. Traditionally, the genetic screening strategy for FH has been Sanger sequencing for assessment of all coding regions in LDLR plus one or two specific exons in APOB, followed by multiplex ligation-dependent probe amplification (MLPA) for detection of CNVs in LDLR (13Wang J. Ban M.R. Hegele R.A. Multiplex ligation-dependent probe amplification of LDLR enhances molecular diagnosis of familial hypercholesterolemia.J. Lipid Res. 2005; 46: 366-372Abstract Full Text Full Text PDF PubMed Scopus (63) Google Scholar). The latter method is essential as ∼10% of FH cases have been attributed to CNVs in LDLR (13Wang J. Ban M.R. Hegele R.A. Multiplex ligation-dependent probe amplification of LDLR enhances molecular diagnosis of familial hypercholesterolemia.J. Lipid Res. 2005; 46: 366-372Abstract Full Text Full Text PDF PubMed Scopus (63) Google Scholar); identifying CNVs increases diagnostic yield, avoiding false-negative diagnoses. Next-generation sequencing (NGS) techniques offer superior analysis with the potential to assess a wider range of genetic abnormalities. However, the ability to detect CNVs in LDLR using NGS data outputs is currently unevaluated. Accurate identification of CNV mutations from NGS data is important because this class of variation comprises a significant proportion of FH cases and not all sequencing facilities have the resources, time, or interest to establish a parallel MLPA system for detecting them. We have developed a dedicated high-coverage targeted NGS panel to detect rare variants in several dyslipidemias, of which FH is the most important clinically (14Johansen C.T. Dube J.B. Loyzer M.N. MacDonald A. Carter D.E. McIntyre A.D. Cao H. Wang J. Robinson J.F. Hegele R.A. LipidSeq: a next-generation clinical resequencing panel for monogenic dyslipidemias.J. Lipid Res. 2014; 55: 765-772Abstract Full Text Full Text PDF PubMed Scopus (98) Google Scholar). Here, we tested whether NGS data could be bioinformatically probed to detect CNVs in the LDLR gene and compared the results with MLPA, which is currently considered to be the “reference standard” or “gold standard” method to detect such variants. We studied 388 Canadian individuals aged ≥18 years who were referred to a tertiary lipid clinic for treatment of severe hypercholesterolemia. Diagnosis of at least possible FH was made by using the Dutch Lipid Clinic Network criteria; all patients had untreated LDL cholesterol ≥5 mmol/l (194 mg/dl), plus family history of hypercholesterolemia, plus some with either personal or family history of premature CHD. Our protocol was approved by the Western University Research Ethics Board, and all participants provided informed consent for genetic analyses. A total of 313 patients studied here were part of our recent report on polygenic FH (15Wang J. Dron J.S. Ban M.R. Robinson J.F. McIntyre A.D. Alazzam M. Zhao P.J. Dilliott A.A. Cao H. Huff M.W. et al.Polygenic versus monogenic causes of hypercholesterolemia ascertained clinically.Arterioscler. Thromb. Vasc. Biol. 2016; 36: 2439-2445Crossref PubMed Scopus (137) Google Scholar). Genomic DNA was isolated from whole blood by using the Puregene DNA Blood Kit (Gentra Systems, Qiagen, Mississauga, Canada) and was subject to targeted NGS using our LipidSeq panel (14Johansen C.T. Dube J.B. Loyzer M.N. MacDonald A. Carter D.E. McIntyre A.D. Cao H. Wang J. Robinson J.F. Hegele R.A. LipidSeq: a next-generation clinical resequencing panel for monogenic dyslipidemias.J. Lipid Res. 2014; 55: 765-772Abstract Full Text Full Text PDF PubMed Scopus (98) Google Scholar). With LipidSeq, each sample is sequenced for 73 key genes in lipid metabolism, including all coding regions, ∼150 bp at intron-exon boundaries, and ∼1,000 bp of the 5′ untranslated region (UTR) of all FH major and minor phenocopy genes, namely, LDLR, APOB, PCSK9, LDLRAP1, APOE, STAP1, LIPA, ABCG5, and ABCG8. Library preparation was performed by using the Nextera Rapid Capture Custom Enrichment kit (Illumina, and samples were sequenced on a personal by using and in with were and by using a in for to human genome variant of and region coverage of was by using with a Our LipidSeq method has an of coverage of for each of for detected variants was performed as previously (15Wang J. Dron J.S. Ban M.R. Robinson J.F. McIntyre A.D. Alazzam M. Zhao P.J. Dilliott A.A. Cao H. Huff M.W. et al.Polygenic versus monogenic causes of hypercholesterolemia ascertained clinically.Arterioscler. Thromb. Vasc. Biol. 2016; 36: 2439-2445Crossref PubMed Scopus (137) Google Scholar, R.A. Ban M.R. Cao H. McIntyre A.D. Robinson J.F. Wang J. next-generation sequencing in monogenic 2015; PubMed Scopus Google Scholar). Sanger sequencing was to the of small-scale variants detected by The MLPA kit The was for the detection of large-scale whole-exon and in The kit for LDLR for each of the and all with the of two for plus one probe for of LDLR and for gene on autosomal The probe also that to that the DNA and are for The and of probe are as previously J.P. R. F. of by multiplex ligation-dependent probe Res. PubMed Scopus Google Scholar), and protocol followed the amplification was in a and were by using a DNA MLPA analysis was performed by using of are compared with within the to the copy number for each We one sample copy number and copy number were were for all the of The CNV an within the variant was for analysis of our LipidSeq data for CNV CNV and by as for each plus a that the region and probe for the specific NGS panel the uses analysis as its principal method, an in sample across a compared with suggests a in and a in sample suggests a the coverage data, the uses a of We provided the a population of from which it with the in coverage data compared with the sample of samples were the was were to for and regions that were to A and were for each The was as the sample coverage by the sample The the number of that a coverage was from the sample A CNV on the that for each region these two 1) 2) heterozygous 3) homozygous or 4) Furthermore, the also across a region as an supporting for CNV as variant allele a of or 1 provided a such as or provided for analysis multiple regions to CNV the limit of CNV detection was the whole-exon limit the limit was the LDLR gene CNV CNVs were on and A of and of were to identify heterozygous a of and of were for from region were also as the of this CNVs detected in LDLR were of were performed in were compared by using while were compared by using analyses. The of was at clinical and of the individuals studied here are shown in for Canadian FH cholesterol, cholesterol, cholesterol, history of on data from history of on data from or FH on data from are as years in and years in Dutch Lipid Clinic on data from in a are as years in and years in Dutch Lipid Clinic of 388 FH patients were for whole-exon CNVs in LDLR detected by MLPA The of of these patients had heterozygous of which multiple were detected The most common CNV a heterozygous of the and in of 38 was in of 38 the exons in LDLR, exons and were by CNV the samples had MLPA outputs from MLPA for two of CNVs are shown in 1 and whole-exon CNVs identified in 388 patients with CNVs the reported and are across each region in a methods of detection of a CNV in the LDLR gene in a patient with FH with a in LDLR 1 for of the MLPA method in LDLR CNV method in LDLR for each LipidSeq region in LDLR; of coverage to copy number and a number of the is from the is the to a are as in CNVs the reported and are across each region of 388 FH patients were for CNVs in LDLR detected by NGS. CNVs and their associated were in 100% concordance with those detected by MLPA Furthermore, the 350 samples negative for CNVs by MLPA were also negative by NGS. MLPA as the “gold there were false and false using our procedure applied to NGS data, which to a diagnostic and of 100% each outputs from CNV for two of CNVs to MLPA referred to are shown in analysis of copy number variants by MLPA and NGS data in a The principal here is that NGS data for the LDLR gene that is bioinformatically from patient samples referred for assessment of FH has a 100% concordance for of large-scale CNVs using MLPA as the “gold standard” The ability to detect the spectrum of mutations in LDLR is in a molecular diagnosis for FH, to or more of such mutations are large-scale CNVs than small-scale DNA variants, on the and M. Sijbrands Low-density lipoprotein receptor analysis in diagnosis of familial 2017; PubMed Scopus Google Scholar). The current procedure for diagnostic often includes targeted NGS followed by Our suggest that the potential CNVs also within NGS data and that MLPA is particularly for the LDLR NGS with has the ability to identify both and large-scale variant detection within a single and single in our analysis of 388 samples referred for FH 38 reported CNVs detected by MLPA were also successfully detected by sample that was for a CNV by MLPA was by our bioinformatic with a and of there were or false-negative from NGS data compared with Furthermore, this targeted NGS method identified a range of CNV including those all both and and both deletions and duplications The prevalence of whole-exon CNVs in FH patients is in our compared with those previously studied R. L. T. P. O. J. L. Genomic of of the LDLR gene in patients with familial Med. Genet. PubMed Scopus Google Scholar, A. Martin B. Wang K. Humphries Multiplex ligation-dependent probe amplification analysis to for deletions and duplications of the LDLR gene in patients with familial Genet. 76: PubMed Scopus Google Scholar). The LDLR locus is to have an high of it to CNV by R. L. T. P. O. J. L. Genomic of of the LDLR gene in patients with familial Med. Genet. PubMed Scopus Google Scholar, M.A. T. and of in the human lipoprotein receptor Sci. PubMed Scopus Google Scholar). The of CNV detected across LDLR with the of these analysis in LDLR has that the of CNV are within and which is are most A. J. J. A. P. M. B. et mutations of of the LDL receptor gene are a of familial Genet. PubMed Scopus Google Scholar). This exons and were by CNVs in our The high of 1 heterozygous deletions can be attributed to the of French in our This is a in to be in of French with FH, the of the French who have had with J. in the gene for the receptor in a of French with familial Engl. J. Med. PubMed Scopus Google Scholar). of the high prevalence of this specific CNV analysis has been an important of FH screening in the or MLPA has been as the “gold standard” for CNV detection in of CNVs from NGS data has been however, it a and CNV include N. A. M. Copy number variation detection and from Res. 2012; PubMed Scopus Google Scholar), J. M. E. et for data in sequencing and implications for copy number variant 2012; PubMed Scopus Google Scholar), A. R. M. for CNV detection in sequencing Genet. Mol. Biol. PubMed Scopus Google Scholar), M. K. E. S.E. D.M. M.J. et and of variation from sequencing J. Hum. Genet. 2012; Full Text Full Text PDF PubMed Scopus Google Scholar), and C. a method to detect copy number variation using PubMed Scopus Google Scholar); however, of these designated methods have shown high rates of which a major on potential clinical use. of the CNV have been and for or NGS which is from targeted NGS because the latter on a genes, with copy number and a coverage in for methods to be a The for each LDLR using targeted NGS versus or NGS increases the to detect our of our of LDLR MLPA and negative samples as to the of this to CNV detection in the clinical diagnostic context for to the of analysis is of for and sequenced with the and NGS panel as the sample of and for and Although proven in our has some in In the of a the CNV does not the of By this is a result of the in as copy is the to whole-exon CNV as these in the regions, which are on our LipidSeq however, although such may be for it does not affect the of a CNV for the of to CNV detection from targeted NGS data has Our for MLPA analysis in and patient which for this of 388 FH be a method to NGS data as such data are for small-scale variant analysis that CNV We have the for CNV detection an for a of Furthermore, all targeted genes on the designated NGS panel are for CNVs CNV analysis can be to all genes the of namely, APOB, PCSK9, LDLRAP1, APOE, STAP1, LIPA, ABCG5, and at Although causative CNVs in these genes are to be rare, have because MLPA methods are either not or not applied for genes the CNV analysis to all such genes our ability to for all genetic of FH in decreases false-negative findings. with this we have identified one patient with a in and two patients with duplications in PCSK9 who had mutations to their not In we report 100% concordance for the detection of whole-exon CNVs in LDLR between a applied to NGS data and the “gold standard” method of This result suggests that the latter method can be removed from the routine molecular diagnostic for FH, costs, resources, and analysis and an even more assessment of this important class of mutations across diagnostic in the apo B gene coronary heart disease copy number variant of coverage familial hypercholesterolemia heterozygous FH homozygous FH LDL receptor gene multiplex ligation-dependent probe amplification next-generation sequencing proprotein convertase subtilisin/kexin type 9 gene variant allele

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.