Updated recommendations for CFTR carrier screening: A position statement of the American College of Medical Genetics and Genomics (ACMG)

Abstract

Disclaimer: This statement is designed primarily as an educational resource for medical geneticists and other clinicians to help them provide quality medical services. Adherence to this statement is completely voluntary and does not necessarily assure a successful medical outcome. This statement should not be considered inclusive of all proper procedures and tests or exclusive of other procedures and tests that are reasonably directed to obtaining the same results. In determining the propriety of any specific procedure or test, clinicians should apply their own professional judgment to the specific clinical circumstances presented by the individual patient or specimen.Clinicians are encouraged to document the reasons for the use of a particular procedure or test, whether or not it is in conformance with this statement. Clinicians also are advised to take notice of the date this statement was adopted, and to consider other medical and scientific information that becomes available after that date. It also would be prudent to consider whether intellectual property interests may restrict the performance of certain tests and other procedures. Where individual authors are listed, the views expressed may not reflect those of authors’ employers or affiliated institutions.Requests for permissions must be directed to the American College of Medical Genetics and Genomics, as rights holder.Pathogenic variants in the CFTR gene are causative of cystic fibrosis (CF) as well as CF-related disorders, such as isolated congenital bilateral absence of the vas deferens (CBAVD). In 2001, several professional organizations joined in acknowledging the importance and technologic advances that would make CF amenable to population-based carrier screening.1Grody W.W. Cutting G.R. Klinger K.W. et al.Laboratory standards and guidelines for population-based cystic fibrosis carrier screening.Genet Med. 2001; 3: 149-154https://doi.org/10.1097/00125817-200103000-00010Abstract Full Text Full Text PDF PubMed Scopus (395) Google Scholar However, the technology and knowledge had not advanced far enough to allow for an equitable application. Variant databases were far less advanced when compared with those that are easily and widely accessible today. Sequencing technology was also early in development. This limited screening to small sets of variants that were most commonly characterized in Ashkenazi Jewish and Northern European populations using targeted, allele-specific testing approaches rather than DNA sequencing. For this reason, recommendations at that time were that screening should be “offered” to those of Ashkenazi Jewish and Northern European descent and “made available” to other groups,1Grody W.W. Cutting G.R. Klinger K.W. et al.Laboratory standards and guidelines for population-based cystic fibrosis carrier screening.Genet Med. 2001; 3: 149-154https://doi.org/10.1097/00125817-200103000-00010Abstract Full Text Full Text PDF PubMed Scopus (395) Google Scholar partly in recognition that the carrier frequencies are highest in those 2 ethnicities, but with the additional implication being that the identification of heterozygotes was not the same across all racial and/or ethnic groups (and indeed quite suboptimal in some). The American College of Medical Genetics and Genomics (ACMG) ultimately recommended a set of 25 pathogenic variants, later reduced to 23 pathogenic variants2Watson M.S. Cutting G.R. Desnick R.J. et al.Cystic fibrosis population carrier screening: 2004 revision of American College of Medical Genetics mutation panel.Genet Med. 2004; 6: 387-391https://doi.org/10.1097/01.gim.0000139506.11694.7cAbstract Full Text Full Text PDF PubMed Scopus (0) Google Scholar with an allele frequency of ≥0.1% in patients with CF in the US population to represent a minimum variant set for pan-ethnic carrier screening of individuals with no family history of CF. This minimum variant set (often referred to as the “ACMG-23”) has remained unchanged since then, even as molecular diagnostic technologies and genetic knowledge have dramatically advanced. Disclaimer: This statement is designed primarily as an educational resource for medical geneticists and other clinicians to help them provide quality medical services. Adherence to this statement is completely voluntary and does not necessarily assure a successful medical outcome. This statement should not be considered inclusive of all proper procedures and tests or exclusive of other procedures and tests that are reasonably directed to obtaining the same results. In determining the propriety of any specific procedure or test, clinicians should apply their own professional judgment to the specific clinical circumstances presented by the individual patient or specimen. Clinicians are encouraged to document the reasons for the use of a particular procedure or test, whether or not it is in conformance with this statement. Clinicians also are advised to take notice of the date this statement was adopted, and to consider other medical and scientific information that becomes available after that date. It also would be prudent to consider whether intellectual property interests may restrict the performance of certain tests and other procedures. Where individual authors are listed, the views expressed may not reflect those of authors’ employers or affiliated institutions. Requests for permissions must be directed to the American College of Medical Genetics and Genomics, as rights holder. The original recommendation left open the option for laboratories to offer an expanded CFTR variant set beyond the recommended set, and at the time, expanded variant sets were met with some controversy on the basis of the available technologies and limited phenotypic knowledge of rare variants.3Grody W.W. Cutting G.R. Watson M.S. The cystic fibrosis mutation “arms race”: when less is more.Genet Med. 2007; 9: 739-744https://doi.org/10.1097/gim.0b013e318159a331Abstract Full Text Full Text PDF PubMed Scopus (0) Google Scholar However, several aspects have now evolved, including the widespread availability of cost-effective, high-throughput DNA sequencing, as well as more standardized variant classification and interpretation at both the general (eg, Richards et al4Richards S. Aziz N. Bale S. et al.Standards 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: 405-424https://doi.org/10.1038/gim.2015.30Abstract Full Text Full Text PDF PubMed Scopus (15792) Google Scholar; ClinVar [https://www.ncbi.nlm.nih.gov/clinvar/]) and gene-specific (eg, CFTR2 [http://cftr2.org]) level. In 2020, the ACMG published an updated set of technical standards for CFTR variant testing, which recommended that laboratories could now use either targeted or comprehensive (ie, next-generation sequencing [NGS]) methods for testing and reaffirmed the original set of 23 variants as the minimum set for CF carrier screening;5Deignan J.L. Astbury C. Cutting G.R. et al.CFTR variant testing: a technical standard of the American College of Medical Genetics and Genomics (ACMG).Genet Med. 2020; 22: 1288-1295https://doi.org/10.1038/s41436-020-0822-5Abstract Full Text Full Text PDF PubMed Scopus (25) Google Scholar an overlapping workgroup subsequently convened to evaluate whether an update to the minimum CFTR variant set was necessary. In addition, in 2021, the ACMG published a new carrier screening clinical practice resource, which continued to recommended offering testing of CFTR (now along with many additional genes) to all pregnant patients, as well as those planning a pregnancy.6Gregg A.R. Aarabi M. Klugman S. et al.Screening for autosomal recessive and X-linked conditions during pregnancy and preconception: a practice resource of the American College of Medical Genetics and Genomics (ACMG).Genet Med. 2021; 23: 1793-1806https://doi.org/10.1038/s41436-021-01203-zAbstract Full Text Full Text PDF PubMed Scopus (86) Google Scholar The original ACMG-23 CF variant set was derived primarily from databases comprising individuals with well-characterized CF who were Non-Hispanic White or Ashkenazi Jewish, thereby allowing individuals from those ancestries to be more easily identified during carrier screening. However, given that CF has been reported across all races, ethnicities, and ancestries, improved equity in variant detection is both necessary and desirable. Self-reported ethnicity also has its own flaws,7Shraga R. Yarnall S. Elango S. et al.Evaluating genetic ancestry and self-reported ethnicity in the context of carrier screening.BMC Genet. 2017; 18: 99https://doi.org/10.1186/s12863-017-0570-yCrossref PubMed Scopus (29) Google Scholar,8Kaseniit K.E. Haque I.S. Goldberg J.D. Shulman L.P. Muzzey D. Genetic ancestry analysis on >93,000 individuals undergoing expanded carrier screening reveals limitations of ethnicity-based medical guidelines.Genet Med. 2020; 22: 1694-1702https://doi.org/10.1038/s41436-020-0869-3Abstract Full Text Full Text PDF PubMed Scopus (34) Google Scholar and this workgroup wanted to maintain a uniform recommended screening approach that would not be subject to this bias. ClinVar (https://www.ncbi.nlm.nih.gov/clinvar/) is a public archive of genetic variants and associated phenotypes with occasional submitted supporting evidence. ClinVar accepts submissions from multiple sources including clinical testing laboratories, research laboratories, genetics clinics, patient registries, locus-specific databases, and expert panels. CFTR variants that are submitted to ClinVar may have been identified in patients with CF or a CFTR–related disorder (including CBAVD and pancreatitis). Importantly, they may have also been identified in unaffected individuals undergoing carrier screening or genetic testing for other indications. ClinVar aggregates the records submitted for each variant and reports the level of review supporting the assertion of clinical significance for each variant ranging from 0 to 4 stars, with 0 stars reflecting a variant assertion that was submitted by an individual without any supportive criteria provided and 4 stars reflecting a variant assertion that is present in a practice guideline. It is important to highlight that often the details regarding the phenotype of the associated individual may be limited or unavailable at the time a genetic variant is submitted. ClinVar does not curate information (ie, determine its validity) or modify classifications once they are submitted; it instead applies the star rating described above based on the data submitted and the source of the submission. The CFTR-France database (https://cftr.iurc.montp.inserm.fr/cftr/) is a national database focused on sharing genetic and phenotypic information of rare CFTR variants that have been identified by genetic testing laboratories in partnership with the French Clinical Registry of patients with CF.9Claustres M. Thèze C. des Georges M. et al.CFTR-France, a national relational patient database for sharing genetic and phenotypic data associated with rare CFTR variants.Hum Mutat. 2017; 38: 1297-1315https://doi.org/10.1002/humu.23276Crossref PubMed Scopus (51) Google Scholar The database was established in 2012 and includes approximately 900 different CFTR variants, which are retrospectively reported from 10 French diagnostic laboratories specializing in CFTR molecular testing. These variants were all identified in individuals with CF, CFTR-RDs, fetuses with echogenic bowel, newborns with pending or inconclusive diagnoses, and asymptomatic individuals with 2 CFTR variants in trans. CFTR-France categorizes variants as CF-causing, CFTR-RD–causing, non–disease-causing, variants of unknown clinical significance, and as variants of varying clinical consequence (VVCCs). VVCCs are either associated with CF or a CFTR-RD (when in trans with a known CF-causing variant), and this can vary within families and across populations. The Clinical and Functional TRanslation of CFTR (CFTR2) database (http://cftr2.org)10Sosnay P.R. Siklosi K.R. Van Goor F. et al.Defining the disease liability of variants in the cystic fibrosis transmembrane conductance regulator gene.Nat Genet. 2013; 45: 1160-1167https://doi.org/10.1038/ng.2745Crossref PubMed Scopus (447) Google Scholar currently includes a total of 485 CFTR variants with various annotations. This database includes phenotype and genotype information collected from approximately 89,000 individuals from national CF patient registries and large clinics from 43 different countries, and CFTR2 is actively trying to further increase the diversity of their collection. The CFTR2 website was established in 2012 and is generally updated on an annual basis. The disease liability of CFTR variants was evaluated using clinical, functional, and epidemiological data, using aggregate information on sweat chloride levels, lung function, pancreatic status, and Pseudomonas infection rates in patients harboring specific combinations of variants. CFTR2 classifies variants as either CF-causing, VVCCs, non-CF causing, or variants of unknown significance. CFTR2 defines VVCCs as those variants that are associated with CF in some individuals but not in others when the variant is present in trans with a CF-causing variant (note that this is a slightly different definition than the one used for VVCCs by CFTR-France). CFTR2 also does not have a classification of “CFTR-RD–causing” as exists in CFTR-France. CFTR variants classified by CFTR2 are given a 3-star assertion status in ClinVar because CFTR2 has been designated as an expert panel by ClinGen. The current workgroup ultimately decided to only use CFTR2 as a source for the pathogenicity of CF-causing variants. Although many pathogenic or likely pathogenic CFTR variant assertions exist in ClinVar, there is often insufficient phenotypic information provided for the individuals who are tested to conclusively associate the variant with a specific CF-related phenotype based solely on that database. As many of the submitted variants were detected during carrier screening and not diagnostic testing, the true phenotypic impact and penetrance of the variant may not yet be well established. Separately, though the information contained in CFTR-France is based on diagnostic testing in affected individuals, the evaluations are performed on a more biogeographically limited population than those used for CFTR2. Therefore, CFTR-France was only used for potentially excluding variants (see Methods) but not as a standalone pathogenicity source for including variants in the minimum set. Other smaller or commercial databases were likewise not pursued, given the barriers to access, varying degrees of curation quality, and/or the potential for bias toward specific populations. The initial list of variants under consideration for inclusion (Figure 1) was a compilation of 2 data sets: (1) variants that were previously included in the set recommended by the ACMG for inclusion in carrier screening (n = 23)2Watson M.S. Cutting G.R. Desnick R.J. et al.Cystic fibrosis population carrier screening: 2004 revision of American College of Medical Genetics mutation panel.Genet Med. 2004; 6: 387-391https://doi.org/10.1097/01.gim.0000139506.11694.7cAbstract Full Text Full Text PDF PubMed Scopus (0) Google Scholar and (2) variants that were interpreted as CF-causing by the CFTR2 project (n = 401, as of the April 29, 2022, data release). The combination of these 2 data sets resulted in an initial list of 416 CFTR variants (Supplemental Table 1). No CFTR2 VVCCs (other than R117H) were included in the initial list because they may not cause CF in some individuals (also see Future revisions of the minimum variant set). Variants interpreted as CF-causing by CFTR2 are also all classified as pathogenic or likely pathogenic variants in ClinVar, which is consistent with the previous 2020 CFTR technical standards.5Deignan J.L. Astbury C. Cutting G.R. et al.CFTR variant testing: a technical standard of the American College of Medical Genetics and Genomics (ACMG).Genet Med. 2020; 22: 1288-1295https://doi.org/10.1038/s41436-020-0822-5Abstract Full Text Full Text PDF PubMed Scopus (25) Google Scholar The initial list of 416 variants underwent further revision, with variants being automatically excluded if they were part of complex alleles (2 variants known to occur in cis and interpreted as a single allele), if they were structural variants involving a deletion or duplication of ≥1 exons, or if they were absent from the gnomAD data set (v2.1.1 or v3.1.2) within the 6 ancestral populations specified below (see Frequency and coverage evaluation). In all excluded complex alleles, 1 of the 2 variants involved was also independently classified and included separately in the initial list. Large structural variants within CFTR were excluded because they are generally rare, may be technically more difficult to detect (and therefore potentially inaccurate or not reported in reference population sequencing), and often have unknown/ambiguous breakpoints. Presence in the gnomAD data set was considered a requirement for inclusion in the final recommended set of variants so that the population frequency and coverage within the United States could be evaluated. A total of 219 variants were excluded using these additional criteria. A manual review was conducted on any included variants that were also deemed CFTR-RD–causing or non–disease-causing variants by CFTR-France because these variants are defined as leading to CFTR-RDs (such as CBAVD) instead of CF or have insufficient clinical and/or functional evidence to be disease causing, respectively. Based on these criteria, only 2 variants required manual review: c.350A>G (p.Arg117His; legacy: R117H) and c.1013C>T (p.Thr338Ile; legacy: T338I), both of which were considered to be CFTR-RD–causing variants by CFTR-France. It was determined that both variants should remain on the list for consideration because of the clinical and functional evidence in CFTR2, which was consistent with the variants causing CF.11Raraigh K.S. Han S.T. Davis E. et al.Functional assays are essential for interpretation of missense variants associated with variable expressivity.Am J Hum Genet. 2018; 102: 1062-1077https://doi.org/10.1016/j.ajhg.2018.04.003Abstract Full Text Full Text PDF PubMed Scopus (49) Google Scholar,12Kiesewetter S. Macek Jr., M. Davis C. et al.A mutation in CFTR produces different phenotypes depending on chromosomal background.Nat Genet. 1993; 5: 274-278https://doi.org/10.1038/ng1193-274Crossref PubMed Scopus (364) Google Scholar However, the interrupting TG variants in the R117H-associated intronic polyT tract were not also included (see Considerations for laboratories). R117H was also present on the originally recommended set of 23 CFTR variants. Variants remaining after automatic and manual exclusion (n = 197) underwent further evaluation to estimate their frequencies in the US population.13Guo M.H. Gregg A.R. Estimating yields of prenatal carrier screening and implications for design of expanded carrier screening panels.Genet Med. 2019; 21: 1940-1947https://doi.org/10.1038/s41436-019-0472-7Abstract Full Text Full Text PDF PubMed Scopus (45) Google Scholar The gnomAD data set, an aggregation of exome and genome sequencing samples, was selected as a reference database from which to obtain carrier frequency estimates for CF alleles because it is large, uniformly processed, and has samples from multiple ancestral populations. We started with the data from gnomAD v2.1.1 (n = 125,748 exomes and n = 15,708 genomes) and the non-v2 subset of gnomAD v3.1.2 genome data (n = 56,456) for a total of 197,912 nonoverlapping samples. We then extracted the allele frequencies for the 197 variants under consideration across each of 6 ancestral populations from gnomAD v2.1.1 plus gnomAD v3.1.2: African/African American (n = 26,863 total individuals), Latino/Admixed American (n = 24,598), Ashkenazi Jewish (n = 6723), East Asian (n = 11,515), non-Finnish European (n = 90,591), and South Asian (n = 17,254), for a total of 177,544 individuals. We note that there are several other ancestral populations (eg, Finnish individuals) in gnomAD that were not included in the analyses because they represent a small proportion of the US population. Carrier frequency for each variant was approximated as 2 times the allele frequency. This approximation is appropriate when allele frequencies are low and because there are no individuals who are homozygous for any of these alleles. Next, for each ancestral population, we ranked the CFTR variants present in the population in order of decreasing frequency. We then tabulated, for each ancestral population, the minimum number of variants needed such that 95% of the total CFTR carrier frequency for the population is achieved; this was based on previous carrier screening practice guidelines14Gross S.J. Pletcher B.A. Monaghan K.G. Carrier screening in individuals of Ashkenazi Jewish descent.Genet Med. 2008; 10: 54-56https://doi.org/10.1097/GIM.0b013e31815f247cAbstract Full Text Full Text PDF PubMed Scopus (164) Google Scholar (Supplemental Figure 1). To derive the final set of CFTR variants, we merged the 95% variant lists from each component ancestry to achieve a nonoverlapping set of 100 variants (Supplemental Figure 2). This approach ensured that at least 95% of the total carrier frequency in each population is represented in the final variant set. The updated minimum CFTR variant set is presented in Table 1. Additional information about the variants is also listed in Supplemental Table 1.Table 1CFTR Carrier Screening Variant Set (n = 100)DNA VariantProtein VariantLegacy Namec.4C>Tp.Gln2TerQ2Xc.178G>Tp.Glu60TerE60Xc.200C>Tp.Pro67LeuP67Lc.223C>Tp.Arg75TerR75Xc.254G>Ap.Gly85GluG85EaVariants that were part of the previously recommended minimum 23-variant set.c.262_263delp.Leu88IlefsTer22394delTTc.271G>Ap.Gly91ArgG91Rc.274-1G>Ap.?406-1G->Ac.292C>Tp.Gln98TerQ98Xc.293A>Gp.Gln98ArgQ98Rc.313delp.Ile105SerfsTer2444delAc.328G>Cp.Asp110HisD110Hc.349C>Tp.Arg117CysR117Cc.350G>Ap.Arg117HisR117HaVariants that were part of the previously recommended minimum 23-variant set.c.489+1G>Tp.?621+1G->TaVariants that were part of the previously recommended minimum 23-variant set.c.571T>Gp.Phe191ValF191Vc.579+1G>Tp.?711+1G->TaVariants that were part of the previously recommended minimum 23-variant set.c.579+3A>Gp.?711+3A->Gc.617T>Gp.Leu206TrpL206Wc.653T>Ap.Leu218TerL218Xc.695T>Ap.Val232AspV232Dc.803delp.Asn268IlefsTer17935delAc.868C>Tp.Gln290TerQ290Xc.988G>Tp.Gly330TerG330Xc.1000C>Tp.Arg334TrpR334WaVariants that were part of the previously recommended minimum 23-variant set.c.1013C>Tp.Thr338IleT338Ic.1021_1022dupp.Phe342HisfsTer281154insTCc.1029delp.Cys343Ter1161delCc.1040G>Ap.Arg347HisR347Hc.1040G>Cp.Arg347ProR347PaVariants that were part of the previously recommended minimum 23-variant set.c.1055G>Ap.Arg352GlnR352Qc.1155_1156dupp.Asn386IlefsTer31288insTAc.1327_1330dupp.Ile444ArgfsTer31461ins4c.1364C>Ap.Ala455GluA455EaVariants that were part of the previously recommended minimum 23-variant set.c.1367T>Cp.Val456AlaV456Ac.1373delp.Gly458AspfsTer111504delGc.1393-1G>Ap.?1525-1G->Ac.1397C>Gp.Ser466TerS466Xc.1400T>Cp.Leu467ProL467Pc.1519_1521delp.Ile507delI507delaVariants that were part of the previously recommended minimum 23-variant set.c.1521_1523delp.Phe508delF508delaVariants that were part of the previously recommended minimum 23-variant set.c.1572C>Ap.Cys524TerC524Xc.1584+1G>Ap.?1716+1G->Ac.1585-1G>Ap.?1717-1G->AaVariants that were part of the previously recommended minimum 23-variant set.c.1624G>Tp.Gly542TerG542XaVariants that were part of the previously recommended minimum 23-variant set.c.1646G>Ap.Ser549AsnS549Nc.1647T>Gp.Ser549ArgS549Rc.1651G>Ap.Gly551SerG551Sc.1652G>Ap.Gly551AspG551DaVariants that were part of the previously recommended minimum 23-variant set.c.1657C>Tp.Arg553TerR553XaVariants that were part of the previously recommended minimum 23-variant set.c.1673T>Cp.Leu558SerL558Sc.1675G>Ap.Ala559ThrA559Tc.1679G>Cp.Arg560ThrR560TaVariants that were part of the previously recommended minimum 23-variant set.c.1679+1G>Ap.?1811+1G->Ac.1680-886A>Gp.?1811+1634A->Gc.1680A>Cp.Arg560SerR560Sc.1682C>Ap.Ala561GluA561Ec.1692delp.Asp565MetfsTer71824delAc.1705T>Gp.Tyr569AspY569Dc.1753G>Tp.Glu585TerE585Xc.1766+1G>Ap.?1898+1G->AaVariants that were part of the previously recommended minimum 23-variant set.c.1766+5G>Tp.?1898+5G->Tc.1837G>Ap.Ala613ThrA613Tc.1882G>Ap.Gly628ArgG628Rc.2052dupp.Gln685ThrfsTer42184insAc.2052delp.Lys684AsnfsTer382184delAaVariants that were part of the previously recommended minimum 23-variant set.c.2125C>Tp.Arg709TerR709Xc.2175dupp.Glu726ArgfsTer42307insAc.2290C>Tp.Arg764TerR764Xc.2353C>Tp.Arg785TerR785Xc.2374C>Tp.Arg792TerR792Xc.2490+1G>Ap.?2622+1G->Ac.2657+5G>Ap.?2789+5G->AaVariants that were part of the previously recommended minimum 23-variant set.c.2668C>Tp.Gln890TerQ890Xc.2739T>Ap.Tyr913TerY913Xc.2834C>Tp.Ser945LeuS945Lc.2909G>Ap.Gly970AspG970Dc.2988G>Ap.Gln996=3120G->Ac.2988+1G>Ap.?3120+1G->AaVariants that were part of the previously recommended minimum 23-variant set.c.3067_3072delp.Ile1023_Val1024del3199del6c.3107C>Ap.Thr1036AsnT1036Nc.3140-26A>Gp.?3272-26A->Gc.3196C>Tp.Arg1066CysR1066Cc.3197G>Ap.Arg1066HisR1066Hc.3266G>Ap.Trp1089TerW1089Xc.3294G>Cp.Trp1098CysW1098Cc.3353C>Tp.Ser1118PheS1118Fc.3472C>Tp.Arg1158TerR1158Xc.3484C>Tp.Arg1162TerR1162XaVariants that were part of the previously recommended minimum 23-variant set.c.3528delp.Lys1177SerfsTer153659delCaVariants that were part of the previously recommended minimum 23-variant set.c.3612G>Ap.Trp1204TerW1204Xc.3659delp.Thr1220LysfsTer83791delCc.3717+5G>Ap.?3849+5G->Ac.3718-2477C>Tp.?3849+10kbC->TaVariants that were part of the previously recommended minimum 23-variant set.c.3744delp.Lys1250ArgfsTer93876delAc.3764C>Ap.Ser1255TerS1255Xc.3808delp.Asp1270MetfsTer83940delGc.3846G>Ap.Trp1282TerW1282XaVariants that were part of the previously recommended minimum 23-variant set.c.3889dupp.Ser1297PhefsTer54016insTc.3909C>Gp.Asn1303LysN1303KaVariants that were part of the previously recommended minimum 23-variant set.a Variants that were part of the previously recommended minimum 23-variant set. Open table in a new tab Current limitations of specific methodologies/platforms were not factored in when determining the updated minimum variant set because the capabilities and availabilities of specific methodologies/platforms are expected to change over time. The workgroup is also aware that there are not likely any existing targeted CF tests available that contain all of the newly recommended variants. However, some laboratories may have previously chosen to offer CF carrier screening using either Sanger or NGS of CFTR, and these methods should encompass all of the genomic regions containing the recommended variants.15Beauchamp K.A. Johansen Taber K.A. Grauman P.V. et al.Sequencing as a first-line methodology for cystic fibrosis carrier screening.Genet Med. 2019; 21: 2569-2576https://doi.org/10.1038/s41436-019-0525-yAbstract Full Text Full Text PDF PubMed Scopus (20) Google Scholar Because the knowledge base in both gnomAD and CFTR2 expands, it may be easier for laboratories to continue testing for the updated minimum variant set if comprehensive testing methods (ie, NGS) are used instead of targeted testing methods.5Deignan J.L. Astbury C. Cutting G.R. et al.CFTR variant testing: a technical standard of the American College of Medical Genetics and Genomics (ACMG).Genet Med. 2020; 22: 1288-1295https://doi.org/10.1038/s41436-020-0822-5Abstract Full Text Full Text PDF PubMed Scopus (25) Google Scholar Laboratories may also need to assess the financial impacts of developing, validating/verifying, and offering a new CF carrier screening method. A CPT code (81223) already exists for CFTR full gene sequencing, though the coverage and reimbursement for this code may need to be reassessed considering its potential recommended use for carrier screening. In addition, if targeted methods are ultimately used, it may not be possible to validate all of the variants in the updated minimum variant set, and laboratories may need to use other strategies to ensure that they can accurately detect all of these variants; however, this should be less of a concern if comprehensive methods are used for testing. Regardless of the method/platform ultimately used by a particular laboratory, the updated minimum variant set is a significant improvement compared with the previously recommended alternative. The new set of 100 variants represents an updated minimum CFTR carrier screening variant set, but it does not represent a limit on the total number of variants that a laboratory can choose to assess, and it is likely that laboratories may already have many (but likely not all) of these variants included as a part of their tests. An informal analysis of the additional CF-causing variants that are present as a part of some current clinical laboratory CFTR tests revealed that most of them were either (1) not in gnomAD and therefore they would not have been considered for inclusion by our workgroup, or (2) they would have been included in our updated minimum variant set if we expanded our desired coverage from 95% to 99%. It is reassuring to note that none of the representative clinical laboratory CFTR tests that were evaluated tested for any known non-CF causing variants. However, a small number of commonly included CFTR2 VVCCs were noted to be present as a part of multiple clinical laboratory CFTR tests, mainly D1152H (c.3454G>C, p.Asp1152His) and F312del (c.935_937delTCT, p.Phe312del; also known as [delta]F311). Laboratories are encouraged to further review the clinical implications of these variants when deciding whether they should remain as a part of their tests, especially when used in the setting of prenatal carrier screening. All of the clinical laboratory CFTR tests that were evaluated were missing a number of established CF-causing variants from CFTR2 that are part of the updated minimum set of 100 variants. Although the group recommended keeping R117H as a part of the updated minimum variant set, the group did not automatically include the associated interrupting TG variants within the intronic polyT tract even though they are typically also detectable using established methods because some of the combinations are not classified as a CF-causing variant in CFTR2 (5T;TG11 [c.1210-7_1210-6del], 5T;TG12 [c.1210-11T>G]) or are not currently present in gnomAD (5T;TG13; c.1210-11delinsGTG). The group reaffirmed the prior recommendation to reflex to polyT analysis and/or reporting for carrier screening only when R117H is also present. However, the group now recommends assessing and reporting the polyTG results whenever 5T is detected because increased numbers of polyTG repeats in cis with 5T are typically associated with increased severity and penetrance of CF-related symptoms compared with individuals with 5T in cis with fewer polyTG repeats.16Nykamp K. Truty R. Riethmaier D. et al.Elucidating clinical phenotypic variability associated with the polyT tract and TG repeats in CFTR.Hum Mutat. 2021; 42: 1165-1172https://doi.org/10.1002/humu.24250Crossref PubMed Scopus (2) Google Scholar In addition, when R117H occurs in cis with 5T, it is typically in association with a TG12 allele, and when it is in cis with 7T, it is typically in association with a TG10 allele; therefore, inferences regarding the phase can be made to further inform reproductive counseling even if familial samples are unavailable for confirmatory follow-up testing.17Raraigh K.S. Aksit M.A. Hetrick K. et al.Complete CFTR gene sequencing in 5,058 individuals with cystic fibrosis informs variant-specific treatment.J Cyst Fibros. 2022; 21: 463-470https://doi.org/10.1016/j.jcf.2021.10.011Abstract Full Text Full Text PDF PubMed Scopus (4) Google Scholar The updated minimum variant set for CF carrier screening is based on (1) evidence that the variant has been established as CF causing and (2) presence of the variant in the gnomAD database. For this version, we took a conservative approach and established a framework that only incorporates well-established pathogenic and likely pathogenic variants to minimize concerns that patients would make reproductive decisions based on limited information. This version of the variant set included CF-causing variants that were annotated as of April 2022 in the CFTR2 database, and additional variants should be reassessed when new classifications are available. Other databases such as ClinVar and CFTR-France could be considered during future revisions if additional strong evidence for variant pathogenicity in affected individuals becomes available. Because CFTR is unique and public databases similar to CFTR2 may not exist for other conditions and their associated genes, the workgroup recognizes that an inability to use the publicly available ClinVar classifications for the selection of variants for other conditions may hamper future carrier screening standardization efforts. However, the use of 3-star review level variants in ClinVar, which have undergone review by ClinGen expert panels, will provide a growing high-quality resource over time. Multiple factors play a role in the identification of variants associated with CF, particularly in the context of biogeographically diverse populations. For this statement, the minimum set of variants was evaluated for population frequency and coverage within 6 global ancestral populations using the gnomAD data set. Notably, though self-reported race and/or ethnicity has been used as a proxy for genetic ancestry in clinical testing, these may inaccurately capture individuals’ ancestral backgrounds.18Braun L. Fausto-Sterling A. Fullwiley D. et al.Racial categories in medical practice: how useful are they?.PLoS Med. 2007; 4e271https://doi.org/10.1371/journal.pmed.0040271Crossref Scopus (150) Google Scholar,19Roth W.D. The multiple dimensions of race.Ethn Racial Stud. 2016; 39: 1310-1338https://doi.org/10.1080/01419870.2016.1140793Crossref Scopus (207) Google Scholar However, these limitations are expected to be mitigated as more ancestral diversity is represented in population databases. Future versions of this minimum variant set should reassess the feasibility and utility of incorporating additional information from other population databases, such as All of Us, TOPMed (gnomAD only contains a subset of these data), and the UK Biobank.20Denny J.C. Rutter J.L. Goldstein D.B. et al.All of Us Research Program InvestigatorsThe “All of Us” research program.N Engl J Med. 2019; 381: 668-676https://doi.org/10.1056/NEJMsr1809937Crossref PubMed Scopus (517) Google Scholar, 21Taliun D. Harris D.N. Kessler M.D. et al.Sequencing of 53,831 diverse genomes from the NHLBI TOPMed Program.Nature. 2021; 590: 290-299https://doi.org/10.1038/s41586-021-03205-yCrossref PubMed Scopus (470) Google Scholar, 22Bycroft C. Freeman C. Petkova D. et al.The UK biobank resource with deep phenotyping and genomic data.Nature. 2018; 562: 203-209https://doi.org/10.1038/s41586-018-0579-zCrossref PubMed Scopus (2638) Google Scholar Although these databases were generally not used to inform this updated minimum variant set because of challenges in accessing the data and the lack of complete phenotypic/biogeographic information, future revisions should re-evaluate their inclusion to be as biogeographically diverse as possible. gnomAD was selected as part of the inclusion criteria because previous population databases were not as comprehensive for assessing the frequency of CFTR variants (eg, NHLBI Exome Sequencing Project); gnomAD is also expected to expand and become more biogeographically diverse over time, thereby ensuring that any CF-causing variants that are common to a specific biogeographic ancestry are appropriately represented in future revisions of the variant set. The updated minimum variant set also does not include any structural variants because of insufficient gnomAD data to support their inclusion at the time of data analysis. However, we expect that future revisions of the minimum variant set will incorporate structural variants as the detection and annotation of these variants improves. Finally, the workgroup set a minimum 95% carrier detection rate for all biogeographic ancestries listed in gnomAD v2.1.1 plus gnomAD v3.1.2 based on precedent from previous carrier screening recommendations.14Gross S.J. Pletcher B.A. Monaghan K.G. Carrier screening in individuals of Ashkenazi Jewish descent.Genet Med. 2008; 10: 54-56https://doi.org/10.1097/GIM.0b013e31815f247cAbstract Full Text Full Text PDF PubMed Scopus (164) Google Scholar The group realizes that some laboratories may choose to incorporate more variants as a part of their CFTR tests to reach carrier detection rates approaching 99%. Future versions of this variant set may reassess the minimum detection rate and the potential harms of not including all well-established CF-causing variants annotated in population databases. In addition, because many of these variants are relatively rare, their carrier frequencies and inclusion in the current 95% threshold variant set could be susceptible to changes in the reference samples from which the carrier frequencies are estimated. However, because their pathogenicity will have already been established, any variants that are included in previous versions of the minimum recommended variant set should be included in any future versions of the variant set (eg, the previously recommended ACMG-23 variants are all included in the updated variant set). The new CFTR variant set represents an updated minimum recommended variant set for CF carrier screening, and this new set now supersedes the previous set of 23 CFTR variants recommended by the ACMG. These revised recommendations apply only to carrier screening. They do not apply to CFTR variant testing for diagnosis or newborn screening. All other aspects of the updated 2020 ACMG CFTR technical standards still apply.5Deignan J.L. Astbury C. Cutting G.R. et al.CFTR variant testing: a technical standard of the American College of Medical Genetics and Genomics (ACMG).Genet Med. 2020; 22: 1288-1295https://doi.org/10.1038/s41436-020-0822-5Abstract Full Text Full Text PDF PubMed Scopus (25) Google Scholar