Opportunistic genomic screening has clinical utility: An interventional cohort study

Abstract

Purpose:

Practice is shifting toward genome-first approaches, such as opportunistic screening for secondary findings (SFs). Analysis of SFs could be extended beyond medically actionable results to include non-medically actionable monogenic disease risks, carrier status, pharmacogenomic variants, and risk variants for common complex disease. However, evidence on the clinical utility of returning these results is lacking. We assessed the outcomes of opportunistic screening for a broad spectrum of SFs by evaluating the yield, impact on clinical management, and consistency between SFs and participants’ clinical features and family history.

Methods:

Adult cancer patients had exome sequencing with the option to learn multiple categories of SFs. Outcomes data were collected through chart review and participant-reported measures up to one year after return of results.

Results:

All participants (n = 139, 85.6% female, average 54.6 years old) who elected to learn SFs had ≥1 variant reported (100% [139/139]). The yield of reportable findings was highest for pharmacogenomic variants (97.8% [135/138] of participants), followed by common disease risk variants (89.4% [118/132]), carrier status (89.3% [117/131]), and variants related to Mendelian (27.2% [34/125]), medically actionable (15.2% [21/138]), and early-onset neurodegenerative (2.6% [3/117]) disease risks. SFs from the American College of Medical Genetics and Genomics list (v3.2, noncancer genes) were reported in 1.4% (2/138) of participants. SFs across all categories demonstrated clinical utility by prompting management changes in 28.1% (39/139) of participants. Moreover, a considerable proportion of participants had suggestive clinical features (49.0% (24/49)]) or family history (21.8% (27/124)) potentially related to their SFs.

Conclusion:

Our findings indicate there are potential benefits from opportunistic screening for a broad range of SFs.

Introduction

Practice is shifting toward genome-first approaches, such as opportunistic screening for secondary findings (SFs). Although guidelines prioritize medically actionable SFs,^1–5 analysis could be extended to include nonmedically actionable monogenic disease risks, carrier status, pharmacogenomic variants, and risk variants for common complex disease.⁶ However, there are concerns about overwhelming patients with a large volume of information and the burden of analyzing and returning results⁷ in the absence of evidence of clinical benefits.

Clinical utility is a central outcome in genomic medicine,^8,9 although it is inconsistently defined and measured.^8,9 Clinical utility refers to the usefulness of a clinical intervention, or its effectiveness in changing clinical management or health outcomes at the system level.^8,9 Studies suggest that SFs could have clinical utility by informing patient management,^10–14 but the body of evidence on the outcomes of SF disclosure has been characterized as weak.¹⁰ In particular, the prevalence of phenotypes consistent with patients’ SFs and medical outcomes of SF receipt have been identified as key research priorities.¹⁰ Moreover, most studies have returned limited types of SFs (eg, only medically actionable results).^10–13,15 Evidence on the clinical utility of other types of SFs is needed to inform clinical practice and policy.

We aimed to evaluate the clinical utility of returning all types of medically relevant SFs. First, we characterized the yield of reported SFs. We then evaluated whether the reported SFs had clinical utility, through impacts on participants’ medical management, and whether participants had clinical features or a family history consistent with their SF-related condition. This is critical evidence to establish as practice continues to shift toward genome-first approaches.

Materials and Methods

This study was conducted within the intervention arm of the Incidental Genomics randomized clinial trial (RCT, ClinicalTrials.gov Identifier: NCT03597165). The parent RCT aimed to evaluate the health outcomes and costs of returning SFs.¹⁶ The primary outcome of the parent RCT (distress) will be published as a separate report. The current report includes only the intervention arm, because SFs were only offered to participants within the intervention arm of the trial.

We use “SFs” to refer to all reported variants which were deliberately analyzed but unrelated to the primary indication (cancer). The trial protocol is published.¹⁶ Research ethics board approval was obtained from all participating sites through Clinical Trials Ontario (#0819), and from the University of Toronto (#00044247).

Patient population

Participants were recruited from familial cancer clinics in Toronto, Canada between September 2018 and January 2021. Inclusion criteria for the parent RCT were personal and family history of cancer or polyposis suggestive of a hereditary cancer syndrome, uninformative results from standard-of-care¹⁷ genetic testing for hereditary cancer syndromes, ≥18 years of age, and able to speak and read English. Exclusion criteria were a positive genetic test result in a cancer-associated gene from previous testing, prior genome or exome sequencing, patient or partner pregnant or actively planning a pregnancy (to avoid any stress related to carrier results; if a participant or their partner were to become pregnant over the study period, they would not be excluded), recurrent or metastatic cancer (stage 4), participation in the decision aid (DA) usability study,¹⁸ or DA RCT¹⁹ that preceded this RCT.¹⁶

Trial participants were randomized after consent and baseline measures. The control arm received primary cancer findings. The intervention arm received primary cancer findings and a choice of SFs (detailed below).

Intervention arm participants who consented to learn SFs and had SFs analyzed were included in this cohort study.

Study procedures

Pretest counseling

Intervention arm participants used the Genomics ADvISER DA (https://genomicsda.com/), which provided education and decision support related to learning SFs,^18–20 followed by one-on-one pretest genetic counseling with a certified genetic counselor (GC, R.K., and S.S.).

SFs categories

Participants could choose to learn SFs from the following categories: medically actionable disease risks (121 genes, 42 which are on the American College of Medical Genetics and Genomics [ACMG] list [v3.2, noncancer genes only]¹), Mendelian disease risks (3837 genes curated from OMIM), early-onset neurodegenerative disease risks (60 genes), carrier status for autosomal recessive (AR) and X-linked conditions (684 genes curated from the Clinical Sequencing Exploratory Research consortium²¹), pharmacogenomic variants (24 Class A variants from the Clinical Pharmacogenetics Implementation Consortium [CPIC]²²), and risk variants for common, multifactorial diseases (26 variants from the National Human Genome Research Institute-European Bioinformatics Institute Catalog [v1.0.2.] reaching genome-wide significance, with an associated odds ratio of ≥2.0).⁶ Methods for developing our gene lists have been published,⁶ and details are provided in the Supplemental Methods (Supplemental Appendix 1). Gene lists can be found in Supplemental Appendix 2. Conditions associated with genes in the first 4 categories all exhibit Mendelian inheritance, but only 1 is called “Mendelian” because of a lack of more specific descriptors applicable to all genes in that category. The term “carrier status” is used to refer to participants who were heterozygous for a pathogenic or likely pathogenic (P/LP) variant in a gene related to a recessive condition.

The medically actionable and Mendelian categories included genes related to conditions with autosomal dominant (AD) inheritance (eg, FBN1), genes related to AR conditions with cases of AD occurrence reported (eg, SLC7A9 and VWF), and genes related to AR conditions. Heterozygous variants associated with AD or AD/AR disease were reported as medically actionable or Mendelian disease risks. Variants related to only AR disease were reported as medically actionable or Mendelian disease risks if a homozygous variant or 2 heterozygous variants were identified and otherwise reported as carrier status.

Sequencing and variant classification

DNA was extracted from peripheral blood or saliva samples. Library preparation, exome sequencing, and bioinformatics were performed at The Centre for Applied Genomics at the Hospital for Sick Children.^6,23–26 Bioinformatics, filtration, variant analysis, and reporting procedures are published.^6,27 Details are provided in Supplemental Methods (Supplemental Appendix 1). Analysis was performed using GRCh37.

SFs were analyzed in participants’ chosen categories; others were masked. Filtration, variant classification, and reporting were performed by study team members (E.R., J.S., R.K., C.M., and S.S.) under the direction of board-certified clinical laboratory geneticists (J.L.-E., J.-M.C.-C., E.G., and A.N.). Sequence variants in genes associated with monogenic conditions were classified following ACMG/Association for Molecular Pathology guidelines,²⁸ with ClinGen specifications as applicable.²⁹ Copy-number variants, structural variants, repeat expansions, and mitochondrial variants were not analyzed, nor were polygenic risk scores constructed. Variant phasing was not performed (further detail is provided in the Supplemental Methods [Supplemental Appendix 1]).

P/LP variants in genes related to monogenic disorders were reported as SFs in the categories medically actionable disease risks, Mendelian disease risks, early-onset neurodegenerative disease risks, and carrier status. Pharmacogenomic variants and risk variants for common multifactorial disease are not compatible with the ACMG/Association for Molecular Pathology variant classification framework. Pharmacogenomic diplotypes were reported with the associated phenotype (using CPIC-recommended terminology),³⁰ and risk variants were reported individually along with the associated disease and odds ratio from the genome-wide association study from which the variant was identified. Supplemental Appendix 2: Final Gene and Variant Lists contains the lists of pharmacogenomic genotypes that were investigated with the associated CPIC recommendations and PharmGKB phenotype summaries (“Pharmacogenomic Variants” tab) and the list of risk variants (“Common disease risk variants tab”) with the odds ratios from the genome-wide association study from which they were identified. All reports were reviewed and signed out by a board-certified clinical laboratory geneticist (J.L.-E., J.-M.C.-C., E.G., and A.N.). Variant nomenclature was validated using VariantValidator (https://variantvalidator.org/)³¹ for this manuscript.

Return of results

The study medical geneticist (R.H.K.) and GCs (R.K., S.S.) determined whether medical actions were warranted for each SF. Recruiting clinics coordinated any necessary referrals, eg, to a medical geneticist (C.F.M.) at an adult medical genetics clinic for monogenic disease risks. Our referral structure is published,³² and details are provided in the Supplemental Methods (Supplemental Appendix 1).

Study GCs returned exome sequencing results in person, over the phone, or by videoconference. This included a discussion of the results, their implications, inheritance patterns, penetrance, expressivity, recommendations, and opportunities for participants’ questions. The study GC asked the participant questions to determine the presence of signs or symptoms or a family history suggestive of the condition(s) related to their SF(s).³²

Letters summarizing the session and recommendations were prepared by the study GCs and study medical geneticist. These were issued to the patient and their family physician^27,33 with the exome sequencing report.

Outcomes

The outcomes of this study were (1) yield of reported SFs, defined as the proportion of participants with SFs reported, and clinical utility outcomes, (2) changes in clinical management, defined as any medical actions prompted by participants’ SFs, and (3) concordance with phenotype or family history, defined as the presence of signs or symptoms suggestive of an SF-associated condition in the participant or biological family member.

The yield and clinical utility of primary cancer results has been assessed separately,^34,35 this study reports only on SFs.

Data collection

Chart review

Research charts from the RCT (exome sequencing reports, GC consult letters, and referral log) and clinical consult notes from specialist evaluations triggered by SFs were reviewed to obtain data on reported SFs, medical follow-up, phenotypic features, and family history. Chart review was conducted in accordance with methodological recommendations (by C.M., R.K., S.S., S.G., E.R., J.S., and D.H.), following a pilot study.^36,37 Details are provided in the Supplemental Methods (Supplemental Appendix 1). Microsoft Access (v17) was used for data collection and management.

Participant self-report

An adapted version of the Behavioral Risk Factor Surveillance System (BRFSS)³⁸ was administered 2 weeks, 6 weeks, 6 months, and 12 months after return of results. The survey interrogated completed and scheduled medical actions and the SF to which each action was attributed. Data were collected using REDCap electronic data capture tools^39,40 hosted at Unity Health Toronto.

Data analysis

Analysis was descriptive, rather than inferential.

Yield

The yield of SFs was calculated as the proportion of participants who received SFs, with the denominator being the number of participants for whom SFs were analyzed. This was calculated overall and for each category of SFs and for the subset of ACMG-recommended SFs (v3.2, cancer genes not included; single heterozygous variants in AR genes not counted). SFs were categorized as they were described on participants’ exome sequencing reports. Throughout the course of the study, some genes were reported in different categories based on refinements to the reporting protocol, the nature of the specific variant identified and its role in disease (eg, SLC3A1 NM_000341.4:c.163del has been reported in affected heterozygotes and was therefore reported as “Mendelian,” whereas SLC3A1 NM_000341.4:c.1400T>C has been reported in affected homozygotes and compound heterozygotes so was reported as “carrier status”), and at the discretion of the study clinicians.

Changes in management

Any completed medical action prompted by SFs was considered a “change in management,” including health care practitioner consultations, investigations, and treatment changes. Recommended and scheduled actions were also characterized. The proportion of patients with a completed change in medical management prompted by their SFs was calculated overall and for each SF category.

Concordance with phenotype and family history

For monogenic SFs (medically actionable, Mendelian, early-onset neurodegenerative, and carrier status), 2 reviewers (C.M. and R.K., a certified GC), assessed whether participants had clinical features or a family history suggestive of the condition(s) associated with their SF(s) using data obtained from the chart review (Supplemental Appendix 1: Supplemental Methods). If the participant was evaluated by a medical geneticist and data were available, the geneticist’s assessment of the case was used (eg, clinical features in keeping with SF-associated condition).

To determine phenotypic relevance for pharmacogenomic variants, participants were asked at return of results whether they had ever taken a medication associated with their variant(s), and whether they had required an alternative medication or a dose adjustment.

We did not assess phenotype or family history concordance for common disease risk variants given the relatively high prevalence of the conditions in the general population and the relatively small genetic contribution to risk for the associated conditions.

Results

Participant characteristics

Following consent, 145 participants were randomized to the intervention arm. One participant was subsequently found to be ineligible for the RCT, 4 dropped out before exome sequencing, and 1 chose not to learn any SFs, and were excluded from this analysis. Therefore, 139 participants had SFs analyzed and were included in this study.

Participants were on average 54.6 years old (SD 11.2, range 25-80), and 85.6% were female (Table 1). All included participants elected to learn at least 1 SF category: 99.3% (138/139) chose medically actionable disease risks and pharmacogenomic variants, 90.0% (125/139) chose Mendelian disease risks, 84.2% (117/139) chose early-onset neurodegenerative disease risks, 94.2% (131/139) chose carrier status, 95.0% (132/139) chose common disease risk variants, and 78.4% (109/139) chose all categories.

Table 1.

Participants’ demographic and clinical characteristics

Characteristics	% (n)
Mean age (years) at enrollment (SD, range)	54.6 (11.2, 25-80)
Sex
Female	85.6 (119)
Male	14.3 (20)
Participant-reported race or ethnicity
Ashkenazi Jewish	10.8 (15)
Asian-East	4.3 (6)
Asian-South	3.6 (5)
Black-Caribbean Region	2.2 (3)
Black-North American	0.7 (1)
French Canadian	2.2 (3)
Latin American	2.2 (3)
Middle Eastern and North African	3.6 (5)
Other or mixed heritage	10.1 (14)
White/European	60.4 (84)
Personal history of cancera
Breast cancer	74.1 (103)
Other	14.4 (20)
Prostate cancer	8.6 (12)
Ovarian cancer	7.2 (10)
Skin cancer	7.9 (11)
Colon cancer	5 (7)
Kidney cancer	3.6 (5)
Polyps (>10)	2.9 (4)
Uterine cancer	1.4 (2)
Brain cancer	1.4 (2)
Stomach cancer	0.7 (1)
Lymphoma	0.7 (1)

^aTotals add to >100% due to some participants having multiple cancers.

Yield of reportable SFs

Overall, 100% (139/139) of participants had at least one SF reported (Table 2) and 98.6% (137/139) had SFs reported in multiple categories. Two hundred and nineteen P/LP variants in genes related to monogenic conditions (medically actionable, Mendelian, early-onset neurodegenerative, and carrier status) were reported in 89.2% (124/139) of participants. P/LP variants in genes related to dominant monogenic conditions (medically actionable, Mendelian, early-onset neurodegenerative) were reported in 35.3% (49/139) of participants. One hundred and seventy-nine P/LP variants were observed only once within our cohort, whereas 40 P/LP variants were observed in multiple participants (Supplemental Table 1). One hundred and nineteen variants were classified as P, one of which was hypomorphic (Supplemental Table 1). One hundred variants were classified as LP, with 4 considered hypomorphic (Supplemental Table 1). All reported variants, variant classifications with supporting evidence, and associated conditions can be found in Supplemental Table 1.

Table 2.

Yield of reported SFs and clinical utility outcomes

Variables	%	n
Yield of SFs
Participants with SFs reported	100	139/139
Medically actionable	15.2	21/138
ACMG-recommended subset	1.4	2/138
Mendelian	27.2	34/125
Early-onset neurodegenerative	2.6	3/117
Carrier status	89.3	117/131
Pharmacogenomic variants	97.8	135/138
Common disease risk variants	89.4	118/132
Changes in management
Participants with changes in management attributed to SFsa	28.1	39/139
Attributed to medically actionable result	71.4	15/21
Attributed to Mendelian result	50	17/34
Attributed to early-onset neurodegenerative result	66.7	2/3
Attributed to carrier status	7.7	9/117
Attributed to pharmacogenomic variants	4.4	6/135
Attributed to common disease risk variants	5.1	6/118
Suggestive clinical features and family history
Presence of features of SF-related condition	49	24/49b
Medically actionable	33.3	7/21
Mendelian	47.1	16/34
Early-onset neurodegenerative	0	0/3
Carrier status	4.3	5/117
Family history of features of SF-related condition	21.8	27/124c
Medically actionable	19.0	4/21
Mendelian	32.4	11/34
Early-onset neurodegenerative	0	0/3
Carrier status	12.8	15/117

For yield, the denominator reflects the number of participants who selected to learn and had SFs analyzed in each category. For changes in management and suggestive features, the denominators reflect the number of participants with SFs reported in each category.

SF, secondary finding.

^aIf a participant attributed a change in management to “All secondary findings,” each category in which they had a variant reported is counted.

^bDenominator reflects number of participants with results related to monogenic disease risk (medically actionable, Mendelian, and early-onset neurodegenerative)

^cDenominator reflects number of participants with monogenic SFs (medically actionable, Mendelian, and early-onset neurodegenerative, carrier status).

Medically actionable disease risks

Medically actionable SFs were reported in 15.2% (21/138) of participants (Figure 1). SFs were most commonly heterozygous variants in F5 (HGNC:3542) associated with factor V Leiden thrombophilia (5.1% [7/138] of participants, AD/AR inheritance), MEFV (HGNC:6998) associated with familial Mediterranean fever (4.3% [6/138], AD/AR), and F11 (HGNC:3529) associated with factor XI deficiency (2.2% [3/138], AD/AR). Homozygous variants were reported in HFE (HGNC:4886) (1 participant) and CYP21A2 (HGNC:2600) (1 participant). Considering the subset of ACMG-recommended SFs (v3.2, noncancer genes only), 1.4% (2/138) of participants had ACMG-recommended SFs: 1 with a heterozygous P variant in LDLR (HGNC:6547) associated with familial hypercholesterolemia (AD), and 1 with a heterozygous P variant in MYBPC3 (HGNC:7551) associated with hypertrophic cardiomyopathy (AD).

Figure 1. Reported pathogenic and likely pathogenic variants.

Distribution of reported pathogenic and likely pathogenic variants across genes for monogenic conditions, in the category in which they were reported. A. Medically actionable disease risks. B. Mendelian disease risks. C. Carrier status. Because the large number of genes in which carrier status variants were reported, only genes in which 3 or more participants had variants reported are included here; all carrier status variants are provided in Supplemental Table 1. *One participant had 2 variants reported. **Homozygous variant.

Mendelian disease risks

Mendelian disease risks were reported in 27.2% (34/125) of participants (Figure 1). Variants were most commonly reported in FLG (HGNC:3748) associated with ichthyosis vulgaris (10.4% [13/125] of participants, AD/AR inheritance), SLC7A9 (HGNC:11067) associated with cystinuria (2.4% [3/125]; AD/AR), and F2 (HGNC:3535) associated with thrombophilia 1 due to thrombin defect (2.4% [3/125], AD/AR).

Early-onset neurodegenerative disease risks

Early-onset neurodegenerative disease risks were reported in 2.6% (3/117) of participants. Two participants (1.7%, 2/117) had variants in SQSTM1 (HGNC:11280), associated with Frontotemporal dementia and/or amyotrophic lateral sclerosis 3 (AD) and Paget’s disease of bone (AD). This gene was also counted under “Mendelian disease risks” because the association with Paget’s disease of bone. One participant (0.9%, 1/117) had a LP variant in LRRK2 (HGNC:18618) associated with Parkinson disease.

Carrier status

Carrier status results were reported in 89.3% (117/131) of participants. The most common results were heterozygous variants in HFE associated with carrier status for hemochromatosis (26.7% [35/131], AR), BTD (HGNC:1122) associated with carrier status for biotinidase deficiency (12.2% [16/131], AR), and SERPINA1 (HGNC:8941) associated with carrier status for alpha-1-antitrypsin deficiency (10.7% [14/131], AR).

Pharmacogenomic variants

Pharmacogenomic results were reported in 97.8% (135/138) of participants, and 86.2% (119/138) of participants harbored variants associated with nonstandard dosing recommendations. Variants were most commonly reported in VKORC1 (HGNC:23663), UGT1A1 (HGNC:12530) and CYP2C19 (HGNC:2621). The most common metabolizer phenotypes were VKORC1 intermediate metabolizer (37% [52/138] of participants]), UGT1A1 intermediate metabolizer (34.1% [47/138] of participants), and CYP4F2 intermediate metabolizer (31.9% [44/138] of participants) (Supplemental Table 2). Supplemental Table 1B lists all reported diplotypes.

Risk variants for common multifactorial disease

Risk variants associated with common disease were reported in 89.4% (118/132) of participants (Supplemental Table 3). Participants most frequently had risk variants associated with age-related macular degeneration (88.6% [117/132]]), Crohn’s disease (81.8% [108/132]) and type I diabetes (70.5% [93/132]). Supplemental Table 1C includes the frequency of all common disease risk variants in our study.

Changes in management

The study clinicians recommended referrals for 30.9% (43/139) of participants (Table 3). Of these participants, 27.9% (12/43) declined the referral. Additionally, the study clinicians recommended that 2.9% (4/139) participants discuss their SFs with a health care practitioner whom they were already seeing. Of the 43 participants with recommended referrals, 65.1% (28/43) completed the referrals. In addition, 12.2% (17/139) participants had self-initiated changes in management.

Table 3.

SFs that prompted changes in management, or for which patients had suggestive features or a family history of suggestive features of the associated condition

Category	Gene (HGNC)	Variant	HGVS Genomic GRCh37	Classification	Criteria (Year)	Condition (Inheritance), as Reported on Exome Sequencing Report	Change in Management Prompted by Result	Patient Features	Family History
Medically actionable	CYP21A2 (HGNC:2600)	NM_000500.9:c.1360C>T, p.(P454S), query homozygote (technical validation required)	NC_000006.11:g.32008783C>T	Pathogenic	PS3, PM3_S (2020)	Adrenal hyperplasia, congenital, due to 21-hydroxylase deficiency (Autosomal recessive); Hyperandrogenism, nonclassic type, due to 21-hydroxylase deficiency (Autosomal recessive)	Medical genetics evaluation	Symptoms of mild, nonclassic CAH	-
Medically actionable	F11 (HGNC:3529)	NM_000128.4:c.403G>T, p.(E135*), Heterozygous	NC_000004.11:g.187195347G>T	Pathogenic	PVS1, PS4, PS3, PP1 (2020)	Factor XI Deficiency (Autosomal Dominant and Autosomal Recessive)	Referral for hematology evaluation recommended, declined by patient.	-	First-degree relative with frequent nosebleeds resulting in fainting. Multiple first-degree relatives with heavy menstrual bleeding.
Medically actionable	F11 (HGNC:3529)	NM_000128.4:c.901T>C, p.(F301L), Heterozygous	NC_000004.11:g.187201412T>C	Pathogenic	PS3, PS4, PM2 (2019)	Factor XI Deficiency (Autosomal Dominant and Autosomal Recessive)	Referral to hematology recommended, declined by patient.	-	-
Medically actionable	F11 (HGNC:3529)	NM_000128.4:c.901T>C, p.(F301L), Heterozygous	NC_000004.11:g.187201412T>C	Pathogenic	PS3, PS4, PM2 (2019)	Factor XI Deficiency (Autosomal Dominant and Autosomal Recessive)	Hematology evaluation; family doctor appointment (patient initiated)	-	-
Medically actionable	F5 (HGNC:3542)	NM_000130.5:c.1601G>A, p.(R534Q), Heterozygous	NC_000001.10:g.169519049=	Pathogenic	PS4, PS3, PM1 (2020)	Factor V Leiden Thrombophilia (Autosomal Dominant and Autosomal Recessive)	Hematology evaluation	-	Family history of varicose veins; 2 second-degree relatives on the same side of the family with a history of DVT.
Medically actionable	F5 (HGNC:3542)	NM_000130.5:c.1601G>A, p.(R534Q), Heterozygous	NC_000001.10:g.169519049=	Pathogenic	PS4, PS3, PM1 (2020)	Factor V Leiden Thrombophilia (Autosomal Dominant and Autosomal Recessive)	Hematology evaluation	-	-
Medically actionable	F5 (HGNC:3542)	NM_000130.5:c.1601G>A, p.(R534Q), Heterozygous	NC_000001.10:g.169519049=	Pathogenic	PS4, PS3, PM1 (2020)	Factor V Leiden Thrombophilia (Autosomal Dominant and Autosomal Recessive)	Hematology evaluation	-	-
Medically actionable	F5 (HGNC:3542)	NM_000130.5:c.1601G>A, p.(R534Q), Heterozygous	NC_000001.10:g.169519049=	Pathogenic	PS4, PS3, PM1 (2020)	Factor V Leiden Thrombophilia (Autosomal Dominant and Autosomal Recessive)	Referral to hematology recommended, declined by patient.	-	First-degree relative homozygous for Factor V Leiden variant (identified before the current study)
Medically actionable	F5 (HGNC:3542)	NM_000130.5:c.1601G>A, p.(R534Q), Heterozygous	NC_000001.10:g.169519049=	Pathogenic	PS4, PS3, PM1 (2020)	Factor V Leiden Thrombophilia (Autosomal Dominant and Autosomal Recessive)	Medical genetics evaluation; referral to hematology (recommended by medical geneticist)	-	-
Medically actionable	F5 (HGNC:3542)	NM_000130.5:c.1601G>A, p.(R534Q), Heterozygous	NC_000001.10:g.169519049=	Pathogenic	PS4, PS3, PM1 (2020)	Factor V Leiden Thrombophilia (Autosomal Dominant and Autosomal Recessive)	Referral for medical genetics evaluation recommended, declined by patient.	Pulmonary embolism following previous surgery.	-
Medically actionable	F5 (HGNC:3542)	NM_000130.5:c.1601G>A, p.(R534Q), Heterozygous	NC_000001.10:g.169519049=	Pathogenic	PS4, PS3, PM1 (2020)	Factor V Leiden Thrombophilia (Autosomal Dominant and Autosomal Recessive)	-	Blood clot during cancer treatment. Followed by hematologist before study.	-
Medically actionable	HFE (HGNC:4886)	NM_000410.4:c.187C>G, p.(H63D), homozygous	NC_000006.11:g.26091179C>G	Pathogenic	PS3, PM3_S (2018)	Hemochromatosis (Autosomal recessive)	Medical genetics evaluation; screening (patient initiated)	-	-
Medically actionable (ACMG list)	LDLR (HGNC:6547)	NM_000527.5:c.1702C>G, p.(L568V), Heterozygous	NC_000019.9:g.11226885C>G	Pathogenic	PS3_M, PS4_M, PM2_P, PM3_S, PM5, PP1, PP3 (2021)	Familial Hypercholesterolemia (Autosomal Dominant)	Medical genetics evaluation; statin initiation (recommended by medical geneticist).	High cholesterol since teenage years; ocular xanthoma	First-degree relative with high cholesterol, coronary artery disease.
Medically actionable	MEFV (HGNC:6998)	NM_000243.3:c.2080A>G, p.(M694V), Heterozygous	NC_000016.9:g.3293407T>C	Pathogenic	PS3, PS4, BP4 (2019)	Familial Mediterranean Fever (Autosomal Dominant and Autosomal Recessive)	Medical genetics evaluation (recommended).	Episode of joint inflammation accompanied by fever.	-
Medically actionable	MEFV (HGNC:6998)	NM_000243.3:c.2040G>A, p.(M680I), heterozygous	NC_000016.9:g.3293447C>T	Pathogenic	PS1, PS4, PS3 (2019)	Familial Mediterranean Fever (Autosomal Dominant and Autosomal Recessive)	Medical genetics evaluation; rheumatologist evaluation; family doctor appointment (patient initiated)	History of periodic fevers with sweats, headaches, joint pain.	-
Medically actionable	MEFV (HGNC:6998)	NM_000243.3:c.2040G>A, p.(M680I), Heterozygous	NC_000016.9:g.3293447C>T	Pathogenic	PS1, PS4, PS3 (2019)	Familial Mediterranean Fever (Autosomal dominant and autosomal recessive)	Medical genetics evaluation	-	-
Medically actionable	MEFV (HGNC:6998)	NM_000243.3:c.2084A>G, p.(K695R), Heterozygous	NC_000016.9:g.3293403T>C	Likely pathogenic	PS4, PM1 (2020)	Familial Mediterranean Fever (Autosomal Dominant and Autosomal Recessive)	Medical genetics evaluation	-	-
Medically actionable	MEFV (HGNC:6998)	NM_000243.3:c.2230G>T, p.(A744S), Heterozygous	NC_000016.9:g.3293257C>A	Likely pathogenic	PS4_M, PM3_S, BP4 (2020)	Familial Mediterranean Fever (Autosomal Dominant and Autosomal Recessive)	Medical genetics evaluation; family doctor appointment (patient initiated)	-	First-degree relative with clinical diagnosis of familial Mediterranean fever (clinical diagnosis obtained before study).
	MEFV (HGNC:6998)	NM_000243.3:c.2177T>C, p.(V726A), Heterozygous	NC_000016.9:g.3293310A>G	Pathogenic	PS3, PS4, PM3_S, BP4 (2020)
Medically actionable (ACMG list)	MYBPC3 (HGNC:7551)	NM_000256.3:c.26-2A>G, Heterozygous	NC_000011.9:g.47373058T>C	Pathogenic	PVS1, PS4_m, PM2_P (2021)	Hypertrophic cardiomyopathy (Autosomal Dominant)	Cardiology evaluation; medical genetics evaluation; Electrocardiogram; Referral to hypertrophic cardiomyopathy clinic for assessment and management (recommended by medical geneticist)	Significant history of cardiac disease in keeping with a late-stage condition.	Sudden death in 50s/60s among multiple second-degree relatives on same side of the family. One first degree relative on the same side of the family with abnormal heart rhythm.
Medically actionable	PKD1 (HGNC:9008)	NM_001009944.3:c.8293C>T, p.(R2765C), Heterozygous	NC_000016.9:g.2153765G>A	Likely pathogenic (hypomorphic)	PS4, PP4, PP3 (2021)	Polycystic kidney disease (Autosomal Dominant)	Medical genetics evaluation	-	-
Medically actionable	VWF (HGNC:12726)	NM_000552.5:c.8069_8070insA, p.(T2691Hfs*7), heterozygous	NC_000012.11:g.6061602_6061603insT	Likely pathogenic	PVS1, PM2 (2019)	Von Willebrand Disease (Autosomal Dominant and Autosomal Recessive)	Family doctor appointment (patient initiated); medical genetics evaluation (recommended)	Easy bruising, heavy menstrual bleeding	-
Medically actionable	VWF (HGNC:12726)	NM_000552.5:c.2561G>A, p.(R854Q), Heterozygous	NC_000012.11:g.6143978C>T	Pathogenic	PS3, PS4, PP1, PP3 (2020)	Von Willebrand Disease (Autosomal Dominant and Autosomal Recessive)	Referral for medical genetics evaluation recommended, declined by patient.	Easy bruising, heavy menstrual bleeding, excessive blood loss during previous surgery.	-
Mendelian	ALG8 (HGNC:23161)	NM_024079.5:c.1090C>T, p.(R364*), heterozygous	NC_000011.9:g.77817941G>A	Pathogenic	PVS1, PM2, PM3 (2019)	Polycystic liver disease 3 with or without kidney cysts (Autosomal Dominant); Congenital disorder of glycosylation, type Ih (Autosomal Recessive)	Medical genetics evaluation; family doctor appointment (patient initiated)	-	-
Mendelian	DCHS1 (HGNC:13681)	NM_003737.4:c.6988C>T, p.(R2330C), Heterozygous	NC_000011.9:g.6646587G>A	Likely pathogenic	PP1_S, PS3 (2020)	Mitral valve prolapse 2 (Autosomal Dominant); Van Maldergem syndrome 1 (Autosomal Recessive)	Medical genetics evaluation; Echocardiogram (recommended)	-	-
Mendelian	EVC2 (HGNC:19747)	NM_147127.5:c.229-1G>T, Heterozygous	NC_000004.11:g.5699375C>A	Likely pathogenic	PVS1, PM2 (2020)	Ellis-van Creveld syndrome (Autosomal recessive); Weyers acrofacial dysostosis (Autosomal dominant)	Referral for medical genetics evaluation recommended, declined by patient.	-	-
Mendelian	F2 (HGNC:3535)	NM_000506.5:c.*97G>A, Heterozygous	NC_000011.9:g.46761055G>A	Pathogenic	PS3, PS4 (2020)	Prothrombin-Related Thrombophilia (Autosomal Dominant)	Hematology evaluation	-	-
Mendelian	F2 (HGNC:3535)	NM_000506.5:c.*97G>A, Heterozygous	NC_000011.9:g.46761055G>A	Pathogenic	PS3, PS4 (2020)	Prothrombin-Related Thrombophilia (Autosomal Dominant)	-	Blood clots during cancer treatment, followed by hematologist before current study	-
Mendelian	F2 (HGNC:3535)	NM_000506.5:c.*97G>A, Heterozygous	NC_000011.9:g.46761055G>A	Pathogenic	PS3, PS4 (2020)	Prothrombin-Related Thrombophilia (Autosomal Dominant)	Hematology evaluation; family doctor appointment (patient initiated)	-	Second-degree relative died of a pulmonary embolism following surgery.
Mendelian	FAM83H (HGNC:24797)	NM_198488.5:c.2377del, p.(Leu793Cysfs*181), heterozygous	NC_000008.10:g.144809255del	Likely pathogenic	PVS1, PM2 (2018)	Amelogenesis imperfecta, type IIIA (Autosomal dominant)	Medical genetics evaluation; recommended to share findings with dentist and continue regular dental care (recommended by medical geneticist)	Braces, receding gums, cracking teeth, multiple extractions, multiple root canals and cavities.	One first-degree relative with multiple dental extractions for cracking teeth.
Mendelian	FECH (HGNC:3647)	NM_000140.5:c.315-48T>C, Homozygous	NC_000018.9:g.55238820A>G	Pathogenic	PS3, PM3_S (2021)	Erythropoietic Protoporphyria (Autosomal Recessive)	Medical genetics evaluation	-	-
Mendelian	FLG (HGNC:3748)	NM_002016.2:c.1501C>T, p.(R501*), heterozygous	NC_000001.10:g.152285861G>A	Pathogenic	PVS1, PS4_M, PM2 (2019)	Ichthyosis vulgaris (Autosomal Dominant); Atopic dermatitis, susceptibility to, 2	Medical genetics evaluation	Dry skin	Eczema, asthma, and allergies in 2 first-degree and one second-degree relative on same side of the family.
Mendelian	FLG (HGNC:3748)	NM_002016.2:c.1501C>T, p.(R501*), Heterozygous	NC_000001.10:g.152285861G>A	Pathogenic	PVS1, PS4_M, PM2 (2019)	Ichthyosis vulgaris (Autosomal Dominant); Atopic dermatitis, susceptibility to, 2	Medical genetics evaluation; family doctor appointment (patient initiated)	Dermatologic urticaria	First-degree relative with eczema
Mendelian	FLG (HGNC:3748)	NM_002016.2:c.7487del, p.(T2496Nfs*104), Heterozygous	NC_000001.10:g.152279875del	Likely pathogenic	PVS1, PS4_M, PM2_P (2021)	Ichthyosis vulgaris (Autosomal Dominant); Atopic dermatitis, susceptibility to, 2	-	Dry, scaly skin; eczema	First-degree relative with a history of dry skin and followed by dermatologist for skin challenges.
Mendelian	FLG (HGNC:3748)	NM_002016.2:c.7339C>T, p.(R2447*), Heterozygous	NC_000001.10:g.152280023G>A	Pathogenic	PVS1, PS4 (2020)	Ichthyosis vulgaris (Autosomal Dominant); Atopic dermatitis, susceptibility to, 2	Medical genetics evaluation	Dry, scaly skin; keratosis pilaris; multiple infections in childhood; recurrent pneumonia since childhood	First-degree relative with keratosis pilaris and IgE-related health issues; first-degree relative with eczema.
Mendelian	FLG (HGNC:3748)	NM_002016.2:c.2282_2285del, p.(S761Cfs*36), Heterozygous	NC_000001.10:g.152285081_152285084del	Pathogenic	PVS1, PS4 (2020)	Ichthyosis vulgaris (Autosomal Dominant); Atopic dermatitis, susceptibility to, 2	-	-	Multiple second-degree relatives on same side of the family with dry skin.
Mendelian	FLG (HGNC:3748)	NM_002016.2:c.6950_6957del, p.(S2317*), Heterozygous	NC_000001.10:g.152280408_152280415del	Likely pathogenic	PVS1_S, PS4_M, PM2_P (2020)	Ichthyosis vulgaris (Autosomal Dominant); Atopic dermatitis, susceptibility to, 2	-	Dry skin.	First-degree relative with dry skin, asthma. First-degree relative with eczema, dry skin.
Mendelian	FLG (HGNC:3748)	NM_002016.2:c.7339C>T, p.(R2447*), Heterozygous	NC_000001.10:g.152280023G>A	Pathogenic	PVS1, PS4 (2020)	Ichthyosis vulgaris (Autosomal Dominant); Atopic dermatitis, susceptibility to, 2	-	Eczema, asthma, allergies, hyperlinear palms and soles.	-
Mendelian	FLG (HGNC:3748)	NM_002016.2:c.1501C>T, p.(R501*), Heterozygous	NC_000001.10:g.152285861G>A	Pathogenic	PVS1, PS4_M, PM2 (2019)	Ichthyosis vulgaris (Autosomal Dominant); Atopic dermatitis, susceptibility to, 2	Medical genetics evaluation.	Dry, scaly skin.	Multiple family members on same side of the family with dry, scaly skin.
Mendelian	FLG (HGNC:3748)	NM_002016.2:c.9740C>A, p.(S3247*), Heterozygous	NC_000001.10:g.152277622G>T	Pathogenic	PVS1, PS4_M, PM2_P (2021)	Ichthyosis vulgaris (Autosomal Dominant); Atopic dermatitis, susceptibility to, 2	-	History of environmental allergies.	Multiple first-degree relatives with environmental allergies.
Mendelian	FLG (HGNC:3748)	NM_002016.2:c.1501C>T, p.(R501*), Heterozygous	NC_000001.10:g.152285861G>A	Pathogenic	PVS1, PS4_M, PM2 (2019)	Ichthyosis vulgaris (Autosomal Dominant); Atopic dermatitis, susceptibility to, 2	-	Dry skin, crevices on soles of feet.	Two first-degree relatives with eczema, one with asthma and environmental allergies.
Mendelian	FLG (HGNC:3748)	NM_002016.2:c.1501C>T, p.(R501*), Heterozygous	NC_000001.10:g.152285861G>A	Pathogenic	PVS1, PS4_M, PM2 (2019)	Ichthyosis vulgaris (Autosomal Dominant); Atopic dermatitis, susceptibility to, 2	Medical genetics evaluation	Dry skin; hay fever; asthma	Multiple second-degree relatives on same side of the family with dry skin concerns.
Mendelian	FLG (HGNC:3748)	NM_002016.2:c.2282_2285del, p.(S761Cfs*36), Heterozygous	NC_000001.10:g.152285081_152285084del	Pathogenic	PVS1, PS4 (2020)	Ichthyosis vulgaris (Autosomal Dominant); Atopic dermatitis, susceptibility to, 2	-	Hay fever and eczema.	-
Mendelian	KCNK18 (HGNC:19439)	NM_181840.1:c.414_415del, p.(F139Wfs*25), Heterozygous	NC_000010.10:g. 118969069_118969070del	Likely pathogenic	PS3, PP1_M (2020)	Susceptibility to migraines, with or without aura (Autosomal Dominant)	Medical genetics evaluation recommended, declined by patient.	-	-
Mendelian	KCNK18 (HGNC:19439)	NM_181840.1:c.414_415del, p.(F139Wfs*25), Heterozygous	NC_000010.10:g. 118969069_118969070del	Likely pathogenic	PS3, PP1_M (2020)	Susceptibility to migraines, with or without aura (Autosomal Dominant)	Medical genetics evaluation recommended, declined by patient.	-	-
Mendelian	MYH2 (HGNC:7572)	NM_017534.6:c.923_941dup, p.(F315Efs*4), heterozygous	NC_000017.10:g. 10443978_10443996dup	Likely pathogenic	PVS1, PM2 (2019)	Inclusion Body Myopathy (Autosomal Dominant and Autosomal Recessive)	Family doctor appointment (patient initiated); medical genetics evaluation (recommended)	-	-
Mendelian	NRL (HGNC:8002)	NM_001354768.3:c.22del, p.(L8Wfs*11), Heterozygous	NC_000014.8:g. 24552040del	Likely pathogenic	PVS1, PM2_P (2021)	Retinitis Pigmentosa (Autosomal Dominant and Recessive)	Medical genetics evaluation; referral to ophthalmology for ongoing monitoring (recommended by medical geneticist); Referral to prenatal genetics (recommended by medical geneticist)	-	-
Mendelian	REN (HGNC:9958)	NM_000537.4:c.249+1G>A, Heterozygous	NC_000001.10:g. 204131140C>T	Pathogenic	PVS1, PM2, PM3_P (2020)	Autosomal Dominant Tubulointerstitial Kidney Disease, REN-Related (Autosomal Dominant); Renal tubular dysgenesis (Autosomal Recessive)	Medical genetics evaluation; recommendation to initiate annual renal monitoring through family physician, including renal ultrasound, urinalysis and renal function (recommended by medical geneticist)	Anemia, elevated creatinine, slightly decreased glomerular filtration rate.	First-degree relative with gout.
Mendelian	RP1 (HGNC:10263)	NM_006269.2:c.2025dup, p.(S676Ifs*22), Heterozygous	NC_000008.10:g. 55538467dup	Pathogenic	PVS1, PM3, PM2 (2020)	Retinitis Pigmentosa (Autosomal dominant and autosomal recessive)	Referral for medical genetics evaluation recommended, declined by patient.	-	-
Mendelian	SLC3A1 (HGNC:11025)	NM_000341.4:c.163del, p.(Q55Rfs*51), Heterozygous	NC_000002.11:g. 44502837del	Pathogenic	PVS1, PM3 (2020)	Cystinuria (Autosomal Dominant and Autosomal Recessive)	Referral for medical genetics evaluation recommended, declined by patient.	History of multiple (>10) kidney stones.	First-degree relative with bladder stones, second-degree relative on same side of the family with kidney stones.
Mendelian	SLC7A9 (HGNC:11067)	NM_014270.5:c.544G>A, p.(A182T), heterozygous	NC_000019.9:g. 33353427C>T	Likely pathogenic	PS4_M, PM2, PS3 (2020)	Cystinuria (Autosomal Dominant and Autosomal Recessive)	Medical genetics evaluation; urine amino acids profile; cystine measurement	-	-
Mendelian	SLC7A9 (HGNC:11067)	NM_014270.5:c.544G>A, p.(A182T), Heterozygous	NC_000019.9:g. 33353427C>T	Likely pathogenic	PS4, PM2, PS3 (2020)	Cystinuria (Autosomal dominant and autosomal recessive)	Referral for medical genetics evaluation recommended, declined by patient.	-	-
Mendelian	SLC7A9 (HGNC:11067)	NM_014270.5:c.544G>A, p.(A182T), Heterozygous	NC_000019.9:g. 33353427C>T	Likely pathogenic	PS4, PM2, PS3 (2020)	Cystinuria (Autosomal Dominant and Autosomal Recessive)	Medical genetics evaluation; abdominal ultrasound; urine amino acid investigations; recommended to maintain daily fluid intake of 3L (by medical geneticist)	-	-
Mendelian	TNFRSF13B (HGNC:18153)	NM_012452.3:c.204dup, p.(L69Tfs*12), Heterozygous	NC_000017.10:g. 16852293dup	Pathogenic	PVS1, PS4, PP1 (2020)	Common variable immunodeficiency-2 (Autosomal Dominant and Autosomal Recessive); Immunoglobulin A deficiency-2	Medical genetics evaluation; IgG, IgA, IgM quantification (recommended by medical geneticist)	History of immune system issues, recurrent infections since childhood, including ear nose, and throat infections, bacterial superinfections, bladder and urinary infections.	Multiple family members on same side of the family with recurrent bacterial infections, and pneumonia; second-degree relative with lupus and first-degree relative with asthma.
Mendelian	TNFRSF13B (HGNC:18153)	NM_012452.3:c.310T>C, p.(C104R), Heterozygous	NC_000017.10:g. 16852187A>G	Pathogenic	PS3, PM3_VS, PP1 (2021)	Common variable immunodeficiency-2 (Autosomal dominant and autosomal recessive); Immunoglobulin A deficiency-2	Medical genetics evaluation	-	-
Mendelian/Early-onset neurodegenerative	SQSTM1 (HGNC:11280)	NM_003900.5:c.1175C>T, p.(P392L), Heterozygous	NC_000005.9:g. 179263445C>T	Likely pathogenic	PS4, PS3, PP1_S (2019)	Paget disease of bone (Autosomal Dominant); Frontotemporal dementia and/or amyotrophic lateral sclerosis 3 (Autosomal Dominant); Distal myopathy with rimmed vacuoles (Autosomal Dominant); Childhood-onset neurodegeneration with ataxia, dystonia and gaze palsy (Autosomal Recessive)	Family doctor appointment (patient initiated); medical genetics evaluation; evaluation by osteoporosis clinic; laboratory investigations; bone mineral density test; initiation of Zoledronic acid infusions	Longstanding history of bone pain; calcium build up in heels and posterior knees requiring surgical removal; hairline fracture of collarbone in response to fall.	-
Mendelian/Early-onset neurodegenerative	SQSTM1 (HGNC:11280)	NM_003900.5:c.1175C>T, p.(P392L), Heterozygous	NC_000005.9:g. 179263445C>T	Likely pathogenic	PS4, PS3, PP1_S (2019)	Paget disease of bone (Autosomal Dominant); Frontotemporal dementia and/or amyotrophic lateral sclerosis 3 (Autosomal Dominant); Distal myopathy with rimmed vacuoles (Autosomal Dominant); Childhood-onset neurodegeneration with ataxia, dystonia and gaze palsy (Autosomal Recessive)	Referral for medical genetics evaluation recommended, declined by patient.	-	-
Early-onset neuro-degenerative	LRRK2 (HGNC:18618)	NM_198578.4:c.4321C>T, p.(R1441C), Heterozygous	NC_000012.11:g. 40704236C>T	Pathogenic	PS3, PS4, PM2, PP1 (2020)	Parkinson disease (Autosomal Dominant)	Medical genetics evaluation; recommended that family doctor refer to movement disorders clinic should Parkinson symptoms develop in future.	-	-
Carrier status	ABCA4 (HGNC:34)	NM_000350.3:c.5882G>A, p.(G1961E), Heterozygous	NC_000001.10:g. 94473807C>T	Pathogenic	PM3_S, PP1, PM2, PS3 (2020)	Stargardt Disease (Autosomal Recessive); Retinitis Pigmentosa (Autosomal Recessive); Age-related Macular Degeneration (Autosomal Dominant); Cone-Rod Dystrophy (Autosomal Recessive)	None		First-degree relative with blindness in 30s-40s.
Carrier status	ABCA4 (HGNC:34)	NM_000350.3:c.6079C>T, p.(L2027F), Heterozygous	NC_000001.10:g. 94471065G>A	Pathogenic	PS3, PM3_S, PM2 (2019)	Stargardt Disease (Autosomal Recessive); Retinitis Pigmentosa (Autosomal Recessive); Age-related Macular Degeneration (Autosomal Dominant); Cone-Rod Dystrophy (Autosomal Recessive)	-	Eye findings consistent with macular degeneration on previous evaluation. Followed by ophthalmologist before study. Variant known before study.	Vision loss in second-degree and third-degree relatives on same side of the family.
Carrier status	BEST1 (HGNC:12703)	NM_004183.4:c.602T>C, p.(I201T), Heterozygous	NC_000011.9:g. 61724436T>C	Pathogenic	PM3, PP1_M, PS3, PM2_P, PP3 (2021)	Bestrophinopathy (Autosomal Dominant and Recessive)	Referral for medical genetics evaluation recommended, declined by patient.	-	-
Carrier status	CFTR (HGNC:1884)	NM_000492.4:c.2657+2_2657+3insA, Heterozygous	NC_000007.13:g. 117242919_117242920insA	Likely pathogenic	PVS1, PM2 (2020)	Cystic fibrosis (Autosomal Recessive); Congenital bilateral absence of vas deferens (Autosomal Recessive); CFTR-related disorders	Share results with respirologist (Recommended)	Chronic obstructive pulmonary disease and bronchiectasis, followed by respirologist before study.	First-degree relative with asthma.
Carrier status	CFTR (HGNC:1884)	NM_000492.4:c.3205G>A, p.(G1069R), Heterozygous	NC_000007.13:g. 117251700G>A	Likely pathogenic	PS3, PP1 (2021)	Cystic fibrosis (Autosomal Recessive); Congenital bilateral absence of vas defense (Autosomal Recessive); CFTR-related disorders	Evaluation by respirologist	-	First-degree relative with history of recurrent bronchitis and pneumonia.
Carrier status	COL18A1 (HGNC:2195)	NM_030582.4:c.2698-2_2698-1del, Heterozygous	NC_000021.8:g. 46912447_46912448del	Likely pathogenic	PVS1, PM2_P (2021)	Knobloch syndrome-1 (Autosomal Recessive); Primary closed-angle glaucoma (Autosomal Dominant)	Medical genetics evaluation; Recommended to be regularly followed by optometry (recommended by medical geneticist)	-	-
Carrier status	CRB1 (HGNC:2343)	NM_201253.3:c.2506C>A, p.(P836T), Heterozygous	NC_000001.10:g. 197396961C>A	Pathogenic	PM3_S, PM2_P, PP1, PP3 (2021)	Leber congenital amaurosis (Autosomal Recessive); Pigmented paravenous chorioretinal atrophy (Autosomal Dominant); Retinitis pigmentosa (Autosomal Recessive)	Medical genetics evaluation	-	First-degree relative with clinical diagnosis of retinitis pigmentosa (diagnosed before study). Second-degree relative on same side of family with vision loss in 20s.
Carrier status	CYP27A1 (HGNC:2605)	NM_000784.4:c.1435C>G, p.(R479G), Heterozygous	NC_000002.11:g. 219679439C>G	Likely pathogenic	PM3_S, PM2, PP3 (2019)	Cerebrotendinous xanthomatosis (Autosomal recessive)	-	Myocardial infarction (there are reports of cardiovascular disease among carriers75)	-
Carrier status	ETFDH (HGNC:3483)	NM_001281738.1:c.631G>A, p.(G211R), Heterozygous	NC_000004.11:g. 159616778G>A	Likely pathogenic	PM3_VS, PM2 (2020)	Multiple acyl-CoA dehydrogenase deficiency (Autosomal Recessive)	-	-	Multiple third-degree relatives on same side of the family died in infancy from suspected genetic condition causing liver issues.
Carrier status	FECH (HGNC:3647)	NM_001012515.4:c.1154del, p.(K385Rfs*21), Heterozygous	NC_000018.9:g. 55218549del	Likely pathogenic	PVS1, PM2 (2020)	Erythropoietic protoporphyria (Autosomal Recessive)	Investigation by family doctor (Recommended)	Arm pain following sun exposure	-
Carrier status	FGA (HGNC:3661)	NM_021871.4:c.532C>T, p.(R178*), Heterozygous	NC_000004.11:g. 155508049G>A	Likely pathogenic	PVS1, PM3_P, PM2_P (2021)	Congenital afibrinogenemia (Autosomal Recessive); Congenital dysfibrinogenemia (Autosomal Dominant and Recessive); Congenital hypodysfibrinogenemia (Autosomal Dominant and Recessive); Familial visceral amyloidosis (Autosomal Dominant)	-	-	First-degree relative with frequent nosebleeds, heavy menstrual cycles.
Carrier status	HBA1 (HGNC:4823)	NM_000558.5:c.95+2_95+6del, heterozygous	NC_000016.9:g. 226812_226816del	Likely pathogenic	PVS1, PM2 (2019)	Alpha-Thalassemia (Autosomal Recessive)	Medical genetics evaluation; Hematologist evaluation	Low mean corpuscular volume. Receiving regular iron infusions before study. Followed by hematologist before study.	-
Carrier status	HFE (HGNC:4886)	NM_000410.4:c.187C>G, p.(H63D), Heterozygous	NC_000006.11:g. 26091179C>G	Pathogenic	PS3, PM3_S (2018)	Hemochromatosis (Autosomal recessive)	-	-	Second-degree male relative frequently donated blood for health reasons.
Carrier status	PKHD1 (HGNC:9016)	NM_138694.4:c.8935C>T, p.(R2979*), Heterozygous	NC_000006.11:g. 51618014G>A	Pathogenic	PVS1, PM2_P, PM3 (2021)	Polycystic kidney disease (Autosomal Recessive)	-	-	Second-degree relative diagnosed with polycystic kidney disease (before study), required kidney transplant. Second-degree relative on same side of the family and third-degree relatives have renal cysts.
Carrier status	SERPINA1 (HGNC:8941)	NM_000295.5:c.863A>T, p.(E288V), Heterozygous	NC_000014.8:g. 94847262T>A	Pathogenic	PM3, PS3, PS4 (2019)	Alpha-1-antitrypsin deficiency	None	-	Family history of chronic obstructive pulmonary disease.
Carrier status	SERPINA1 (HGNC:8941)	NM_000295.5:c.863A>T, p.(E288V), Heterozygous	NC_000014.8:g. 94847262T>A	Pathogenic	PM3, PS3, PS4 (2019)	Alpha-1-antitrypsin deficiency (Autosomal recessive)	-	-	First-degree relative (smoker) with chronic obstructive pulmonary disease and liver disease.
Carrier status	SERPINA1 (HGNC:8941)	NM_000295.5:c.739C>T, p.(R247C), Heterozygous	NC_000014.8:g. 94847386G>A	Likely pathogenic	PS3, PM3 (2020)	Alpha-1-antitrypsin deficiency (Autosomal recessive)	-	-	Multiple third-degree relatives on same side of the family died in infancy from a suspected genetic condition causing liver issues.
Carrier status	SERPINA1 (HGNC:8941)	NM_000295.5:c.1096G>A, p.(E366K), Heterozygous	NC_000014.8:g. 94844947C>T	Pathogenic	PM3_VS, PS3 (2019)	Alpha-1-antitrypsin Deficiency (Autosomal Recessive)	-	-	First-degree relative with emphysema, chronic obstructive pulmonary disease. First-degree relative died of liver disease.
Carrier status	SYNE1 (HGNC:17089)	NM_182961.4:c.5943dup, p.(A1982Sfs*14), Heterozygous	NC_000006.11:g. 152737634dup	Likely pathogenic	PVS1, PM2_P (2021)	SYNE1 Deficiency (Autosomal Recessive); Emery-Dreifuss muscular dystrophy 4 (Autosomal Dominant)	Medical genetics evaluation; referral to cardiology (recommended by medical geneticist); referral to neurology if muscle pain/cramping develop in future (recommended by medical geneticist). Referral to prenatal genetics (recommended by medical geneticist)	-	-
Carrier status	TYR (HGNC:12442)	NM_000372.5:c.1392dup, p.(K465*), heterozygous	NC_000011.9:g. 89028336dup	Likely pathogenic	PVS1_S, PM2_P, PS4_m (2021)	Oculocutaneous albinism type IA/B (Autosomal Recessive); Susceptibility to cutaneous malignant melanoma (Autosomal Dominant)	Medical genetics evaluation; Recommended to be regularly followed by dermatology	-	First-degree relative with white hair, fair skin, near-sightedness, sun sensitivity
Carrier status	WDR19 (HGNC:18340)	NM_025132.4:c.2129T>C, p.(L710S), Heterozygous	NC_000004.11:g. 39233563T>C	Likely pathogenic	PM3, PP3, PP1, PM2 (2019)	Nephronophthisis 13 (Autosomal Recessive); Senior-Loken syndrome 8 (Autosomal Recessive); Cranioectodermal dysplasia 4 (Autosomal Recessive); Short-rib thoracic dysplasia 5 with or without polydactyly (Autosomal Recessive); Senior-Loken syndrome 8 (Autosomal Recessive);	-	-	Third-degree relative required a kidney transplant, died from kidney disease in 30s.
Carrier status	WNT10A (HGNC:13829)	NM_025216.3:c.682T>A, p.(F228I), Heterozygous	NC_000002.11:g. 219755011T>A	Likely pathogenic	PM3, PP1_S, PM2 (2020)	Hypohidrotic Ectodermal Dysplasia (Autosomal Recessive); Odontoonychodermal dysplasia (Autosomal Recessive); Schopf-Schulz-Passarge syndrome (Autosomal Recessive); Selective tooth agenesis (Autosomal Dominant and Recessive)	-	-	First-degree relative with selective tooth agenesis.
PGx	VKORC1 (HGNC:23663)	*1/*2	NC_000016.9:g. 31107689C>T	N/A	N/A	VKORC1 Intermediate metabolizer. Associated medications include warfarin	Family doctor appointment (Scheduled, patient initiated)	-	-
	CYP2C19 (HGNC:2621)	*1/*17	NC_000010.10:g. 96521657C>T	N/A	N/A	CYP2C19 Intermediate metabolizer. Associated medications include amitriptyline, clopidogrel, citalopram, escitalopram, voriconazole
	UGT1A1 (HGNC:12530)	*28/*28	NC_000002.11:g. 234668893_234668894dup	N/A	N/A	UGT1A1 Intermediate metabolizer. Associated medications include atazanavir
PGx	CYP2C19 (HGNC:2621)	*1/*2	NC_000010.10:g. 96541616G>A	N/A	N/A	CYP2C19 Intermediate Metabolizer. Associated medications include amitriptyline, clopidogrel, citalopram, escitalopram, voriconazole, lansoprazole	Family doctor appointment (patient initiated)	-	-
PGx	CYP2C19 (HGNC:2621)	*1/*2	NC_000010.10:g. 96541616G>A	N/A	N/A	CYP2C19 Intermediate Metabolizer. Associated medications include amitriptyline, clopidogrel, citalopram, escitalopram, voriconazole, lansoprazole.	Changed medication (changed to a different type, patient initiated)	-	-
PGx	CYP2C19 (HGNC:2621)	*1/*17	NC_000010.10:g. 96521657C>T	N/A	N/A	CYP2C19 Rapid Metabolizer. Associated medications include amitriptyline, clopidogrel, citalopram, escitalopram, voriconazole, lansoprazole	Changed medication (increased dose, patient initiated)	Took amitriptyline in the past (before study), which was ineffective. Taking lansoprazole at the time of return of results.	-
PGx	CYP2C19 (HGNC:2621)	*1/*2	NC_000010.10:g. 96541616G>A	N/A	N/A	CYP2C19 Intermediate Metabolizer. Associated medications include amitriptyline, clopidogrel, citalopram, escitalopram, voriconazole, lansoprazole	-	Took citalopram or escitalopram in the past and required dose adjustment (before study).
PGx	CYP2C19 (HGNC:2621)	*2/*2	NC_000010.10:g. 96541616G>A	N/A	N/A	CYP2C19 Poor Metabolizer. Associated medications include amitriptyline, clopidogrel, citalopram, escitalopram, voriconazole, lansoprazole	-	Took amitriptyline in the past, discontinued because of severe side effects (before study).
PGx	CYP2C19 (HGNC:2621)	*1/*17	NC_000010.10:g. 96521657C>T	N/A	N/A	CYP2C19 Rapid Metabolizer. Associated medications include amitriptyline, clopidogrel, citalopram, escitalopram, voriconazole, lansoprazole	-	Taking escitalopram at return of results. Had required a dose increase, as initial dose was ineffective (before study).
PGx	CYP2C19 (HGNC:2621)	*1/*17	NC_000010.10:g. 96521657C>T	N/A	N/A	CYP2C19 Rapid Metabolizer. Associated medications include amitriptyline, clopidogrel, citalopram, escitalopram, voriconazole, lansoprazole	-	Taking amitriptyline at return of results, had required a dose increase (before study). Taking pantoprazole at return of results, had required a temporary dose increase (before study).
Common disease risk variants	CFH (HGNC:4883)	rs1061170-C (heterozygous for risk allele)	NC_000001.10:g. 196659237C=	N/A	N/A	Susceptibility to age-related macular degeneration	Ophthalmology evaluation (patient initiated)	-	-
	CFH (HGNC:4883)	rs800292-G (homozygous for risk allele)	NC_000001.10:g. 196642233G=	N/A	N/A
	SKIV2L (HGNC:10898)	rs429608-G (heterozygous for risk allele)	NC_000006.11:g. 31930462G=	N/A	N/A
	CFB (HGNC:1037)	rs641153-G (heterozygous for risk allele)	NC_000006.11:g. 31914180G=	N/A	N/A
	C2 (HGNC:1248)	rs9332739-G (homozygous for risk allele)	NC_000006.11:g. 31903804G=	N/A	N/A
	NELFE (HGNC:13974)	rs522162-T (heterozygous for risk allele)	NC_000006.11:g. 31919917T=	N/A	N/A
Common disease risk variants	ARMS2 (HGNC:32685)	rs10490924-T (heterozygous for risk allele)	NC_000010.10:g. 124214448G>T	N/A	N/A	Susceptibility to age-related macular degeneration	Ophthalmology evaluation (scheduled, patient initiated)	-	-
	CFB (HGNC:1037)	rs641153-G (homozygous for risk allele)	NC_000006.11:g. 31914180G=	N/A	N/A
Common disease risk variants	INS (HGNC:6081)	rs689-T (heterozygous for risk allele)	NC_000011.9:g. 2182224A>T	N/A	N/A	Susceptibility to type I diabetes	Family doctor appointment (patient initiated); Alternative medicine practitioner appointment (Scheduled, patient initiated)	-	-
	HLA-DQA1 (HGNC:4942)	rs9272346-A (heterozygous for risk allele)	NC_000006.11:g. 32604372G>A	N/A	N/A
All SFs	Carrier status, pharmacogenomic variants, risk variants						Family doctor appointment (patient initiated)	-	-
All SFs	Carrier status, risk variants						Family doctor appointment (patient initiated)	-	-
All SFs	Risk variants and pharmacogenomic variants						Family doctor appointment (Scheduled, patient initiated)	-	-
All SFs	Carrier status, risk variants						Mental health practitioner appointment (patient initiated)	-	-
All SFs	Carrier status, risk variants, pharmacogenomic variants						Family doctor appointment (patient initiated)	-	-
All SFs	Medically actionable, Mendelian, pharmacogenomic variants, risk variants						Family doctor appointment (patient initiated)	-	-

Medical actions that were not confirmed to have been completed are indicated as “recommended,” or “scheduled.” Those that were recommended and declined are indicated as “declined by patient.” All other management changes were confirmed to have been completed. Actions that were initiated by patients are marked as “patient initiated”; all other actions had been recommended by the study clinicians or specialists to whom participants were referred. Variant classifications reflect evidence available at the time of classification; the year is indicated in brackets. Supplemental Table 1 contains written evidence summaries for all pathogenic/likely pathogenic variants.

CAH, congenital adrenal hyperplasia; DVT, deep vein thrombosis; N/A, not applicable; PGx, pharmacogenomic; SF, secondary finding.

Overall, 28.1 (39/139) of participants had a change in management prompted by their SFs (Figure 2, Tables 2 and 3). Changes in management were predominately specialist evaluations, (71.8% [28/39] of participants with a change in management, 20.1% [28/139] of participants overall) and patient-initiated family doctor appointments (30.8% [12/39] of participants with a change in management, 8.6% [12/139] of participants overall, Figure 2). Specialist evaluations were mainly by medical genetics (53.8% [21/39] with a change in management, 15.1% [21/139] overall), prompted by SFs related to monogenic disease risks (Figure 2). Early in the study, the adult medical genetics clinic accepted referrals for FLG variants but later stopped accepting referrals for this type of result. Patients also had evaluations by hematology (eg, for results associated with coagulation disorders such as F5), cardiology (MYBPC3 variant), rheumatology (MEFV variant), respirology (CFTR (HGNC:2621) variant), and an osteoporosis specialty clinic (SQSTM1 variant). One participant sought an ophthalmologist consultation because of their age-related macular degeneration risk variants. Six participants had consultations with multiple types of health care practitioners. Other changes in management included imaging, laboratory investigations, treatment initiation, and medication changes (Figure 2, Table 3). Importantly, changes in management were triggered by results across all SF categories, although medically actionable SFs resulted in the highest proportion of participants with SF-attributed changes in management (71.4 % [15/21] of participants with medically actionable results, Table 2).

Figure 2. Changes in management attributed to SFs.

Additional details can be found in Table 3. A. Completed changes in management. B. Recommended or scheduled changes in management that were not confirmed to have been completed. Note, SQSTM1 is shaded pink/orange when the action was attributed to both Paget’s disease of bone and frontotemporal dementia and pink when only Paget’s disease of bone was the attribution. HCM, hypertrophic cardiomyopathy.

There were additional recommended or scheduled medical actions that were not confirmed to have been completed. These included referrals, monitoring, investigations, and treatment initiation (Table 3, Figure 2).

Suggestive features or family history

Monogenic results

Among participants with SFs related to monogenic disease risk (medically actionable, Mendelian, and early-onset neurodegenerative), 49.0% (24/49) had a history of suggestive features of the related condition (Tables 2 and 3). Most frequently, these were participants with FLG variants and a history of dry and scaly skin, eczema, asthma, and/or allergies (12/24 participants) or participants with MEFV variants and a history of periodic fevers, inflammation, and joint pain (3/24 participants, Table 3). Other examples included 1 participant with a P LDLR variant and a history of elevated cholesterol and ocular xanthoma, a participant with a LP SQSTM1 variant and a history of bone pain, and a participant with a P TNFRSF13B (HGNC:18153) variant with a history of recurrent infections and immune system issues (Table 3). Among participants with carrier status SFs, 4.3% (5/117) exhibited some attenuated features which may be related to their variant. Table 3 provides details for all patients with features or a family history potentially related to their SFs.

With regard to family history, 21.8% (27/124) of participants with monogenic SFs had a family history of suggestive features of the related condition (Tables 2 and 3). Again, these were frequently participants (10/27) with FLG variants whose relatives had dry and scaly skin, eczema, asthma, or allergies. Other recurrent findings included participants with SERPINA1 variants and a family history of liver and/or lung disease (3/27 participants), and participants with ABCA4 (HGNC:34) variants and a family history of vision loss (2/27 participants, Table 3). Other examples included a participant with a P MYBPC3 variant and a family history of multiple sudden deaths and a participant with a P SLC3A1 variant and a family history of kidney and bladder stones. Three participants reported a first- or second-degree relative with an established clinical diagnosis of a condition associated with one of their SFs (1 with retinitis pigmentosa, 1 with familial Mediterranean fever, and 1 with polycystic kidney disease) who did not have a molecular diagnosis (Table 3). It should be noted that relatives did not undergo genetic testing as part of this study; therefore, we cannot be certain of variant transmission in each family or the relationship between the variant and relatives’ phenotypes.

Pharmacogenomic results

Among the 119 participants with pharmacogenomic results associated with nonstandard dosing recommendations, 11.7% (14/119) were taking an associated medication at return of results, or had previously taken an associated medication. Of these participants, 35.7% (5/14) had required a dose adjustment, alternative medication, or discontinuation before learning SFs, which their SFs may explain. These 5 participants all had variants in CYP2C19 (3 rapid metabolizers, 1 poor metabolizer, and 1 intermediate metabolizer) and had taken escitalopram (n = 2), amitriptyline (n = 3), pantoprazole (n = 1), and lansoprazole (n = 1); 2 participants had taken multiple associated medications (Table 3).

Discussion

Our study demonstrates the clinical utility of opportunistic screening for a broad range of SFs. All 139 participants who chose to learn SFs had at least one SF returned. These were most frequently pharmacogenomic variants, risk variants for common diseases, and carrier status. Fewer participants harbored variants related to medically actionable disease risks, early-onset neurodegenerative disease risks and other Mendelian disease risks. A considerable proportion of participants had suggestive features or a family history, potentially related to their SFs. Importantly, 28.1% of participants had a change in their medical management prompted by their SFs, including SFs not a priori categorized as medically actionable. In the context of genomic sequencing, it can be unfeasible to measure the impact of test results on morbidity and mortality—traditional measures of clinical benefit—because it can take years or decades for these outcomes to occur.⁴¹ Proximate endpoints, such as changes in management, can indicate potential benefits.⁴¹

Our high yield of reportable SFs is unsurprising given the broad scope of our analysis. Previous studies have similarly found that up to ~100% of individuals harbor carrier status variants^42–44 and up to ~95% harbor pharmacogenomic variants,^42,43 reflecting common genetic variation. Our 1.4% yield of ACMG-recommended medically actionable SFs (noncancer genes only) falls within the range of ~1% to 4% reported by previous studies.^11,45–48 It should be noted that because cancer was the primary indication in our cohort, cancer-related genes were not included on the SFs gene lists. Many of the genes on the ACMG-recommended SFs list are cancer genes, and as such, the yield of reportable ACMG-recommended SFs has been found to be higher in cohorts in which cancer genes were also interrogated.^11,45–48 The SFs that we observed were extremely heterogeneous. Findings ranged from variants in genes associated with highly penetrant disorders with severe manifestations (eg, MYBPC3) to variants associated with a modest absolute risk of morbidity (eg, F5 heterozygotes) or relatively mild manifestations (eg, FLG heterozygotes). Importantly, the identification of an SF is not equivalent to a clinical diagnosis.⁴⁹ Many variants exhibit incomplete or age-dependent penetrance and variable expressivity,⁵⁰ which complicates risk assessment and patient management. The heterogeneity among the SFs identified when going beyond the predefined ACMG-recommended gene list may present challenges for nongenetics practitioners ordering genomic sequencing. Inclusion of certified GCs and medical geneticists, and a dedicated referral structure for medical genetics or disease specialist consultation³² is critical to enable medical actionability and promote clinical utility from the return of a broad range of SFs.^51,52

As a consequence of the high yield of reported findings, a relatively high proportion of our participants (28.1%) had a change in management prompted by their SFs. A systematic review found that medically actionable SFs prompted management changes for 20% (144/709) of patients.¹⁰ This is lower than what we observed (SFs led to changes in management among 71.4% of our participants with medically actionable results), possibly attributable to the incomplete ascertainment of outcomes among studies included in the systematic review.¹⁰ In studies that returned a broader range of SFs (eg, nonmedically actionable results), up to 34% of participants were found to have a SF-prompted change in their management.^14,44 Variability across studies is likely in part due to differences in the nature of the SFs, study methods for outcome ascertainment, and patients’ access to follow-up care (eg, because of health insurance coverage). In our cohort, although some participants had recommendations for the initiation of surveillance or medications based on their SFs (eg, statins for LDLR variant), many of the reported SFs did not lead to immediate recommendations for intervention, and instead required ongoing awareness about possible symptoms or risk factors (eg, awareness of blood clot risk factors and symptoms among those with heterozygous P F5 variants). The clinical relevance of SFs may also change over a participant’s lifetime, for example, carrier status findings may have the greatest utility among patients who have yet to complete family planning because results could be used in reproductive decision making. Carrier status results had more limited utility for the participants in our cohort, among whom the average age was over 50 years old.

SFs provided a possible explanation for a considerable proportion of our participants’ personal and family disease histories, a further potential benefit of returning these results. There is an increasing appreciation that genetic conditions are underascertained in the general population.⁵³ Other studies have similarly found suggestive clinical features or a family history among some but not all participants with SFs,^11,14,44 which is to be expected given incomplete and age-dependent penetrance. It is possible that additional participants who were negative for a phenotype during the study may manifest features of the condition in the future. Our study extends the current literature by reporting on phenotypic and family history concordance for a broader range of SFs than has been investigated by most previous studies. However, these findings should not be interpreted as penetrance estimates because this study was not designed nor statistically powered to assess penetrance.⁵⁴ Large-scale sequencing studies among unselected populations with linked phenotypic data are needed to clarify penetrance and expressivity for a broad range of genetic conditions in unselected populations, to enhance patient counseling and management in genomic screening.

Although considerable scholarship has addressed clinical utility and metrics of value in genomic medicine,^8,55–61 the field has yet to come to consensus on what endpoint or what level of evidence is necessary or sufficient to demonstrate clinical utility. There is considerable heterogeneity in how clinical utility is measured and defined by decision makers.⁶² When genomic testing is used in a diagnostic context for a symptomatic patient, a molecular diagnosis alone may demonstrate value, because even in the absence of therapeutic options, it can end a diagnostic odyssey and provide information about prognosis and reproductive risk.^8,63 Indeed, many studies of the utility of diagnostic genomic sequencing report only the diagnostic yield.⁶⁴ However, in the context of genomic screening among asymptomatic patients, a reportable molecular finding is a less tenable measure of value. SFs, as with the results of any genomic or diagnostic test, provide information and only influence health outcomes indirectly through informing medical actions, which may or may not improve health outcomes.⁹ When a P/LP variant is identified in an asymptomatic patient, there can be uncertainty around if or when the patient will develop the associated condition, as well as uncertainty around relative benefits and harms of any subsequent interventions. In some cases, identification of SFs in an asymptomatic patient may afford opportunities for early detection or prevention, but changes in traditional health outcomes, such as improved morbidity or mortality from screening, may take years or decades to occur after testing.⁴¹ In such cases, endpoints such as changes in management can be considered as proxies, which may be linked to changes in health outcomes. We defined a change in management broadly, including referrals for specialist evaluation, investigations, surveillance, and changes to medication, in line with suggested metrics of clinical utility.^56,65 Although our findings suggest that there could be potential benefits from returning a broad range of SFs, further work is needed to determine whether the observed management changes translate into health outcomes, such as improvements in morbidity, mortality, and quality of life.

Genomic results may also benefit patients for reasons of personal utility that go beyond medical actions, such as enabling future planning or for the inherent value of information.^8,66,67 The personal utility of genomic screening requires further investigation. Potential benefits of genomic screening must also be considered in light of possible harms. These include adverse psychological effects, overdiagnosis, overtreatment, and unnecessary health system costs.⁶⁸ Reassuringly, quantitative studies of the psychological outcomes of genomic screening have not found evidence of lasting negative psychological impacts,^11,69–71 although, qualitative research has found that some patients may experience shock in the short term, or challenges in coping with their risk genomic risk information.⁷² Future analyses from the Incidental Genomics RCT will address personal utility, quality of life, psychosocial outcomes, cascade effects on family members, health service utilization, costs, and cost-effectiveness,¹⁶ providing further information to inform decision making about SFs.

Our study has several limitations. Our study population was fairly homogeneous, and our participants were predominately female; however, this is reflective of the population we recruited from (patients who had received standard-of-care multigene panel testing for suspicion of a hereditary cancer syndrome); hereditary breast/ovarian/prostate cancer panels comprise the bulk of testing performed among Ontario cancer patients,⁷³ and the majority of patients who receive these panels are female.⁷⁴ Several factors could have led to an underestimation of the utility of returning SFs. First, exome sequencing technology limited our reportable pharmacogenomic variants, and we did not construct polygenic risk scores. We did not analyze copy-number variants, structural variants, mitochondrial variants, or repeat expansions. Analyzing additional variant types may have increased our yield and subsequent clinical utility. Second, we used patient-reported outcomes to ascertain specific types of changes in management (eg, general practitioner appointments). Participants may not have recalled all changes in management during the study period, leading to an underestimation of this outcome. Finally, we used a narrow definition of clinical utility, focused on medical outcomes for the proband. Analyses that incorporate personal utility, lifestyle changes, and cascade effects on relatives could reveal benefits missed by this study.

Conclusions

We found a high yield of reportable SFs, largely due to reporting carrier status, pharmacogenomic variants, and common disease risk variants. SFs across all categories demonstrated clinical utility by prompting changes in management for a high proportion of participants (28.1%), including results not a priori categorized as medically actionable. SFs also provided possible explanations for participants’ personal and family disease histories. These findings suggest there are potential benefits from opportunistic screening for a broad spectrum of SFs.

Supplementary Material

The online version of this article (https://doi.org/10.1016/j.gim.2024.101323) contains supplemental material, which is available to authorized users.

Members of the Incidental Genomics Study Team

Yvonne Bombard, Susan Randall Armel, Melyssa Aronson, Nancy N. Baxter, Kenneth Bond, José-Mario Capo-Chichi, June C. Carroll, Timothy Caulfield, Marc Clausen, Tammy J. Clifford, Iris Cohn, Irfan Dhalla, Craig C. Earle, Andrea Eisen, Christine Elser, Michael Evans, Emily Glogowski, Tracy Graham, Elena Greenfeld, Jada G. Hamilton, Wanrudee Isaranuwatchai, Monika Kastner, Raymond H. Kim, Jordan Lerner-Ellis, Chantal F. Morel, Michelle Mujoomdar, Abdul Noor, Kenneth Offit, Seema Panchal, Mark E. Robson, Stephen W. Scherer, Adena Scheer, Kasmintan A. Schrader, Terrence Sullivan, Kevin E. Thorpe

Data Availability

The data sets used and/or analyzed during the current study are available from the corresponding author on reasonable request. Variant classifications were shared in ClinVar, and all reported variants are included in the article or Supplemental Materials.