ACMG SF v3.2 list for reporting of secondary findings in clinical exome and genome sequencing: A policy statement of the American College of Medical Genetics and Genomics (ACMG)

Abstract

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.

Introduction

The American College of Medical Genetics and Genomics (ACMG) previously published guidance for reporting secondary findings (SF) in the context of clinical exome and genome sequencing.^1–5 The ACMG Secondary Findings Working Group (SFWG) and Board of Directors (BOD) have agreed that the list of recommended genes should now be updated annually, but with an ongoing goal of maintaining this as a minimum list. Reporting of SFs should be considered neither a replacement for indication-based diagnostic clinical genetic testing nor a form of population screening.

Per nomenclature guidance put forth by the ACMG SFWG and approved by the BOD,² versioning of the SF list was designed to differentiate major vs minor revisions. Major revisions include conceptual changes to the categories or genes/variants in the SF list or the removal/addition of a large number of genes in a single update; these changes are denoted by updating the version number to the next integer (eg, v4.0, v5.0, etc). Minor revisions reflect the addition or removal of one or a few genes or variants without any policy change, and are denoted by an incremental change to the number after the decimal point (eg, v3.1, v3.2, etc).

The current SFWG includes clinical geneticists, molecular and/or cytogenetics clinical laboratory directors, genetic counselors, cardiologists, a bioinformatician and a bioethicist. The SFWG has met at least monthly via web conferencing to review nomination forms and vote on inclusion or exclusion of gene-phenotype pairs for the ACMG SF v3.2 list. Details on the nomination and review process have been published.³

Internal nominations from SFWG committee members and external nominations were considered for the SF v3.2 list. Internal nominations from committee members included the CALM1, CALM2, and CALM3 genes as gene-phenotype pairs with long QT syndrome (LQTS) and catecholaminergic polymorphic ventricular tachycardia (CPVT). An external nomination was reviewed for the ATP7A gene associated with Menkes disease. No nominations were requested by other professional organizations. The final proposed ACMG SF v3.2 list from the SFWG was sent to the ACMG Board of Directors for review and approval in October, 2022. Member comments were received in January, 2023 and the working group submitted a revision to the Board in February, 2023,

Recommendations for the ACMG SF v3.2 List

The overall charge of the SFWG is to provide recommendations for a minimum list of gene-phenotype pairs for opportunistic screening to facilitate the identification and/or management of risks for selected genetic disorders through established interventions aimed at preventing or significantly reducing morbidity and mortality.² The complete ACMG SF v3.2 list is presented in Table 1 (and is also presented as a spreadsheet in Supplemental Table 1). As shown in Table 2, three new genes, CALM1, CALM2, and CALM3, were added to the v3.2 list, with a brief description of the factors considered in adding each of these genes. Only one gene, ATP7A, was considered for inclusion but ultimately excluded from the v3.2 list (Table 3); ATP7A could be reviewed again in the future if new data emerges related to either Menkes disease or other phenotypes associated with this gene.

Table 1.

ACMG SF v3.2 gene and associated phenotypes recommended for return as secondary findings from clinical exome and genome sequencing

Phenotype	ACMG SF List Version	MIM disorder	Gene	Inheritance	Variants to reporta
Genes Related to Cancer Phenotypes
Familial adenomatous polyposis (FAP)	1.0	175100	APC	AD	All P and LP
Familial medullary thyroid cancer/Multiple endocrine neoplasia 2	1.0	155240 171400 162300	RET	AD	All P and LP
Hereditary breast and/or ovarian cancer	1.0 1.0 3.0	604370 612555 114480	BRCA1 BRCA2 PALB2	AD	All P and LP
Hereditary paraganglioma-pheochromocytoma syndrome	1.0 1.0 1.0 1.0 3.0 3.0	168000 601650 605373 115310 171300 171300	SDHD SDHAF2 SDHC SDHB MAX TMEM127	AD	All P and LP
Juvenile polyposis syndrome (JPS)	2.0	174900	BMPR1A	AD	All P and LP
Juvenile polyposis syndrome (JPS)/Hereditary hemorrhagic telangiectasia syndrome	2.0	175050	SMAD4	AD	All P and LP
Li-Fraumeni syndrome	1.0	151623	TP53	AD	All P and LP
Lynch syndrome (Hereditary Nonpolyposis Colorectal Cancer; HNPCC)	1.0	609310 120435 614350 614337	MLH1 MSH2 MSH6 PMS2	AD	All P and LP
Multiple endocrine neoplasia type 1	1.0	131100	MEN1	AD	All P and LP
MUTYH-associated polyposis (MAP)	1.0	608456	MUTYH	AR	P and LP (2 variants)
NF2-related schwannomatosis	1.0	101000	NF2	AD	All P and LP
Peutz-Jeghers syndrome (PJS)	1.0	175200	STK11	AD	All P and LP
PTEN hamartoma tumor syndrome	1.0	158350	PTEN	AD	All P and LP
Retinoblastoma	1.0	180200	RB1	AD	All P and LP
Tuberous sclerosis complex	1.0 1.0	191100 613254	TSC1 TSC2	AD	All P and LP
von Hippel-Lindau syndrome	1.0	193300	VHL	AD	All P and LP
WT1-related Wilms tumor	1.0	194070	WT1	AD	All P and LP
Genes Related to Cardiovascular Phenotypes
Aortopathies	1.0 1.0 1.0 1.0 1.0 1.0	154700 609192 610168 613795 611788 132900	FBN1 TGFBR1 TGFBR2 SMAD3 ACTA2 MYH11	AD	All P and LP
Arrhythmogenic right ventricular cardiomyopathy (a subcategory of Arrhythmogenic Cardiomyopathy or ACM)	1.0 1.0 1.0 1.0 1.0	609040 607450 610476 604400 610193	PKP2 DSPb DSC2 TMEM43 DSG2	AD	All P and LP
Catecholaminergic polymorphic ventricular tachycardia	1.0 3.0 3.0	604772 611938 615441	RYR2 CASQ2 TRDN c	AD AR AR	All P and LP P and LP (2 variants)
Dilated cardiomyopathy	1.0 1.0 3.0 3.0 3.1 3.1 3.1 3.1	601494 115200 617047 604145 613881 604765 613172 611879	TNNT2 d LMNA e FLNC d, TTN f BAG3 DES RBM20 TNNC1	AD	All P and LP (See text)
Ehlers-Danlos syndrome, vascular type	1.0	130050	COL3A1	AD	All P and LP
Familial hypercholesterolemia	1.0 1.0 1.0	143890 144010 603776	LDLR APOB PCSK9	SD AD AD	All P and LP
Hypertrophic cardiomyopathyg	1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0	192600 115197 613690 115196 608751 612098 600858 608758	MYH7 b MYBPC3 TNNI3 TPM1 MYL3 ACTC1 PRKAG2 MYL2	AD	All P and LP
Long QT syndrome types 1 and 2	1.0 1.0	192500 613688	KCNQ1 KCNH2	AD	All P and LP
Long QT syndrome 3; Brugada syndrome	1.0	603830, 601144	SCN5A b	AD	All P and LP
Long QT syndrome types 14–16	3.2	616247 616249 618782	CALM1 g CALM2 g CALM3 g	AD AD AD	All P and LP
Genes Related to Inborn Errors of Metabolism Phenotypes
Biotinidase deficiency	3.0	253260	BTD	AR	P and LP (2 variants)
Fabry disease	1.0	301500	GLA h	XL	All hemi, het, homozygous P and LP
Ornithine transcarbamylase deficiency	2.0	311250	OTC	XL	All hemi, het, homozygous P and LP
Pompe disease	3.0	232300	GAA	AR	P and LP (2 variants)
Genes Related to Miscellaneous Phenotypes
Hereditary hemochromatosis	3.0	235200	HFE	AR	HFE p.C282Yi homozygotes only
Hereditary hemorrhagic telangiectasia	3.0 3.0	600376 187300	ACVRL1 ENG	AD	All P and LP
Malignant hyperthermia	1.0 1.0	145600 601887	RYR1 j CACNA1S	AD	All P and LP
Maturity-Onset of Diabetes of the Young	3.0	600496	HNF1A	AD	All P and LP
RPE65-related retinopathy	3.0	204100, 613794	RPE65	AR	P and LP (2 variants)
Wilson disease	2.0	277900	ATP7B	AR	P and LP (2 variants)
Hereditary TTR (transthyretin) amyloidosis	3.1	105210	TTR	AD	All P and LP

MIM Mendelian Inheritance of Man, AD autosomal dominant, AR autosomal recessive, SD semidominant, XL X-linked, LP likely pathogenic, P pathogenic.

^aVariants within genes associated with autosomal dominant phenotypes should be classified as pathogenic or likely pathogenic to be reportable. Genes associated with phenotypes inherited in an autosomal recessive fashion would need two likely pathogenic and/or pathogenic variants to meet the threshold for reporting even when phase is undetermined, as follow-up family variant testing can often resolve phase. Finally, pathogenic and likely pathogenic variants within genes associated with X-linked phenotypes that are apparently hemizygous, heterozygous or homozygous should be reported, as often heterozygous females can have adverse medical events at a reasonable frequency and treatment or amelioration of disease is available. Variants of uncertain significance should not be reported in any gene.

^bAlso associated with dilated cardiomyopathy (DCM) as a primary disease.

^cAlso associated with long QT syndrome.

^dAlso associated with hypertrophic cardiomyopathy (HCM).

^ePathogenic and Likely Pathogenic (P/LP) LMNA variants that have any case level phenotype evidence of association with cardiac disease (eg, DCM, ARVC, ACM, and/or arrhythmia) should be reported, whereas previously reported P/LP missense variants never associated with cardiac disease should not be reported. Also, for novel pLOF variants that reach LP without case observations, these variants should be reported given the general association of pLOF LMNA variants with cardiac disease and the evidence summary should include mention of the spectrum of phenotypes that may be observed with LMNA pLOF variation.

^fWe currently recommend that only frameshift and nonsense variants, and variants known to impact the splicing of TTN exons with high PSI (see references 9-11), be evaluated for pathogenicity and returned as secondary findings if classified as P/LP.

^gAlso associated with catecholaminergic polymorphic ventricular tachycardia

^hGene also applies to the cardiovascular category.

ⁱTranscript for the HFE gene is NM_000410.3.

^jRYR1 also causes a neuromuscular phenotype. Only P/LP variants associated with malignant hyperthermia should be reported as a secondary finding.

Table 2.

New Gene/Phenotype Pairs for SF v3.2 List

Gene/Phenotype	Additional Comments
Genes Related to Cardiovascular Phenotypes
CALM1/Long QT syndrome	Similar prevalence/penetrance rates to other sudden cardiac death (SCD) genes previously on ACMG SF list
CALM2/Long QT syndrome	Similar prevalence/penetrance rates to other sudden cardiac death (SCD) genes previously on ACMG SF list
CALM3/Long QT syndrome	Similar prevalence/penetrance rates to other sudden cardiac death (SCD) genes previously on ACMG SF list

Table 3.

Genes Not Selected for SF v3.2 List

Gene/Phenotype	Category	Additional Comments
ATP7A/Menkes disease	Inborn Errors of Metabolism	Lack of demonstrated effectiveness, and possible toxicity, of the available treatment

Considerations for Specific Phenotypic Categories

Genes Related to Cancer Phenotypes

Recommended for addition to, or removal from, the SF v3.2 list: None

Genes Related to Cardiovascular Phenotypes

Recommended for addition to the SF v3.2 list: CALM1, CALM2, and CALM3

Cardiovascular genes have been represented on the SF list since its inception, due to the morbidity and mortality of heart failure and sudden cardiac death (SCD), which can both be treated or prevented with well-established interventions.^6,7

For version 3.2, three additional genes (CALM1, CALM2, and CALM3) were reviewed. These genes cause predisposition to LQTS, and the available evidence supports a similar or greater risk of morbidity and mortality compared to other SCD genes already included in previous versions of the SF list. The three calmodulin genes (CALM1, CALM2, and CALM3) are located on different chromosomes but encode identical 149 amino acid proteins. All three were previously classified by ClinGen as having Definitive evidence for LQTS with atypical features such as presentation in infancy or early childhood and with functional heart block and severe QT prolongation.⁸

A member comment suggested updating the nomenclature used for reportable variants in the TTN gene, as outlined in the Table 1 footnote. Since the exact disease mechanisms are still being elucidated, it was suggested to refer to TTN truncating variants as TTNtv, instead of loss-of-function variants. This update has been included as part of the ACMG SF v3.2 list. A member comment also requested additional guidance regarding which truncating variants in the Titin gene (TTNtv) should be reported as secondary findings. Specifically, a suggestion was made to “add specific details to include consideration of the cardiac isoforms/transcripts, highly expressed exons, and established regions with enrichment for TTNtv and dilated cardiomyopathy (DCM).”

We currently recommend that only frameshift and nonsense variants, and variants known to impact the splicing of TTN exons with high proportion spliced-in (PSI), be evaluated for pathogenicity and returned as secondary findings if classified as P/LP.^9–11 This update has been included as part of the ACMG SF v3.2 list and provided as a footnote in Table 1. We anticipate that additional guidance may be provided from experts in the field over time and defer to further guidance that may be published in the future. (Note “variants known to impact splicing” refers to variants affecting the invariable +/− 1, 2 positions and other coding or noncoding variants with demonstrated impact.)

Genes Related to Inborn Errors of Metabolism Phenotypes

Nominated for addition to the SF list: ATP7A

The working group carefully considered the nomination of ATP7A as a gene-disease pair for Menkes disease. Menkes disease is infantile onset, has a high morbidity, the causative gene (ATP7A) can be assessed by standard exome sequencing, and there is a potential treatment. In order to further evaluate this gene-phenotype pair, we consulted an ad hoc expert for feedback about available treatment options. After careful consideration, we determined that there was insufficient evidence that the only available treatment, subcutaneous injections of copper histidinate, is efficacious. In addition, there was concern that this treatment is potentially toxic.¹² We also noted that P/LP variants would likely be identified as a primary (diagnostic) result as opposed to a secondary finding.

Pathogenic variants in ATP7A can also result in occipital horn syndrome (OHS), and ATP7A-related distal motor neuropathy (DMN). OHS and ATP7A-related DMN are childhood or adult onset, and hence could be considered secondary findings, but this gene was only reviewed by the working group in relation to Menkes disease. Although the other conditions were not specifically reviewed, the concern about insufficient evidence for efficacy of copper histidinate would also apply to OHS and ATP7A-related DMN.

Conclusions

With the 2021 publication of the SF policy statements for reporting of secondary findings and the SF v3.0 gene list,^3,4 the SFWG created a mechanism for separating updates to the policy and principles for SF reporting from updates to the SF gene list. This dual publication approach facilitates more frequent updates to the actual SF gene list. Going forward, we foresee updates to the general policy only as needed, likely every few years. In contrast, updates to the gene list will be targeted to occur on an annual basis, and to be published at approximately the same time each year so that all stakeholders can expect an update and be prepared to revise laboratory and reporting processes. We recognize that clinical laboratories must integrate updates into their workflow, and clinicians must familiarize themselves with the genes on the list for the purposes of genetic counseling and informed consent. Our intention is to publish an updated list each year in January.

The SFWG will continue to review this list of actionable genes, and new nominations, throughout the course of the year. We also wish to remind the community that ACMG members may nominate genes or variants to be added to, or removed from, the list based on an evolving evidence base and/or evolving standards in the practice of medicine. We will also consider nominations submitted through representatives of other professional organizations. Nomination forms can be found on the ACMG website (https://form.jotform.com/93256282335156). We hope that the detailed descriptions of our decision process during the preparation of this update will help the community better understand the types of genes and variants that we consider appropriate for this list to guide nominations going forward.