Test Description

Peripheral blood sample required.

Diagnostic testing methodologies:(DNA sequencing) Mutation scanning in single direction confirmed in opposite direction and again in an exon specific separate assay. All primers are SNP and BLAST alignment checked. All mutations identified in a previous preliminary project were confirmed using this methodology. This methodology is well established in the laboratory for many disorders.

(MLPA) Use of an MLPA kit designed specifically to pick up deletions and duplications in the SCN1Agene. Exon 21 deletion control identified amongst 10 normal control samples, assay repeated to validate results. This methodology is well established in the laboratory for other disorders.

Public Health Importance

The estimated incidence of the disease in the UK population is difficult to ascertain as historically this group of epilepsy syndromes have been excluded from epidemiological studies as they have been difficult to diagnose in electro-clinical terms ¹ . A recent study based on a UK birth cohort suggested an incidence of at least 1 in 40,000 live births for SCN1A positive Dravet syndrome and 1 in 29.000 for Dravet syndrome as a whole ² . Dravet syndrome has been misdiagnosed as whooping cough vaccine damage or pertussis encephalopathy³ .

Where the mutation is inherited the inheritance pattern is autosomal dominant but most cases are found to be de novo. Familial cases most commonly arise in Genetic Epilepsy with Febrile Seizures plus (GEFS+). The majority of cases are sporadic and the great value of this test is providing an early diagnosis and allowing appropriate treatment. Penetrance is difficult to estimate.

A confirmed diagnosis has implications for treatment stragegies and genetic counseling. It can save many additional costly and invasive investigations. When a diagnosis confirms or supports a clinical suspicion, medication changes may result ⁴. Anti-epileptic medications such as carbamazepine, lamtrigine and pheytoin can worsen seizures in Dravets syndrome whereas there is evidence from placebo controlled trials that a medication called stirpentol in combination with valproate and clobazam may reduce seizures. Infants with Dravet syndrome suffer from developmental regression and there is good evidence that some of this is due to uncontrolled seizures and abnormal EEG activity (an epileptic encephalopathy). There is clinical justification and evidence from recent research on adults with the syndrome to hope that controlling seizures will reduce the cognitive impairment associated with the syndrome ⁵.

The clinical features of Dravet Syndrome develop over several years so without the support of molecular genetic testing the diagnosis may not be made until 2-4 years of age. By this time the child may have suffered years of uncontrolled seizures and already have significant cognitive impairment ² .

Published Reviews, Recommendations and Guidelines

Systematic evidence reviews: None identified

Recommendations by independent group: UKGTN – Gene Dossier

Guidelines by professional groups: None identified

Evidence overview

Analytic Validity: Test accuracy and reliability in measuring analytes or other entities measured (analytic sensitivity and specificity).

Diagnostic testing methodologies:(DNA sequencing) Mutation scanning in single direction confirmed in opposite direction and again in an exon specific separate assay. All primers are SNP and BLAST alignment checked. All mutations identified in a previous preliminary project were confirmed using this methodology. This methodology is well established in the laboratory for many disorders.

(MLPA) Use of an MLPA kit designed specifically for the SCN1A gene. Exon 21 deletion control identified amongst 10 normal control samples, assay repeated to validate results. This methodology is well established in the laboratory for other disorders.

Validation: Clinical Molecular Genetic Society (CMGS) Trainee project: 6/6 mutations were identified in “blind” analysis using conformation sensitive capillary electrophoresis (CSCE). A further panel of 20 patients with varied infantile epileptic encephalopathies, referred from a consultant paediatric neurologist, were screened using CSCE and DNA sequencing. Results were confirmed by bi-directional sequencing.

Analytical Sensitivity is estimated at >98% for bi-directional sequencing, 99.5% when MLPA included based upon our own laboratory test performance experience.

Clinical Validity: Test accuracy and reliability in
supporting clinical or public health assessment.

In all Dravet syndrome cases the clinical sensitivity is around 80%, rising to 90% in typical Dravet syndrome cases. In our series about 10% of individuals classified as typical Dravet syndrome were not found to have an SCN1A mutation.

The negative predictive value is estimated to be low. Approximately 1/100 of our patients thought to have SMEI/related syndrome (based on clinical, and electro-clinical data) were found not to have an SCN1A mutation. This is most likely to be due to allelic heterogeneity particularly for the related syndromes.

The negative predictive value is estimated to be low. Approximately 1/100 of our patients thought to have SMEI/related syndrome (based on clinical, and electro-clinical data) were found not to have an SCN1A mutation. this is most likely to be due to allelic heterogeneity particularly for the related syndromes.

Clinical Utility: Net benefit of test in improving health outcomes

When a pathogenic mutation is identified the diagnosis can be made and/or confirmed (i.e. some patients are so young that their epilepsy phenotype has not fully evolved enough for a clinical diagnosis to be made). A confirmed diagnosis has implications for treatment strategies and genetic counselling. It can save many additional costly and invasive investigations. When a genetic diagnosis confirms or supports a clinical suspicion, medication changes may result. Anti-epileptic medications such as carbamazepine, lamotrigine and phenytoin can worsen seizures in Dravet Syndrome whereas there is evidence from placebo controlled trials that a medication called stiripentol in combination with valproate and clobazam may reduce seizures. Infants with Dravet syndrome suffer from developmental regression and there is good evidence that some of this is due to the uncontrolled seizures and abnormal EEG activity (an epileptic encephalopathy). There is clinical justification and evidence from recent research on adults with the syndrome to hope that controlling seizures will reduce the cognitive impairment associated with the syndrome ⁵ .

We undertook a review of our service using questionnaires to ask carers and physicians their views on genetic testing. 187 carers and 163 physicians responded.

In the carers of the mutation positive group, 87% reported genetic testing helpful, 55% said it led to a change in treatment resulting in fewer seizures. 41% described other changes including improved access to therapies and respite care. In 48%, physicians reported that testing facilitated diagnosis earlier than with clinical and EEG data alone. Molecular testing prevented additional investigations in 67% of cases, altered treatment approach in 69%, helped medication choice in 74% and through medication change improved seizure control in 42%. Carer and physician views correlated significantly with regard to the clinical utility of genetic testing. In addition to confirming a clinical diagnosis, SCN1A genetic testing enabled early diagnosis, influenced treatment-choice and facilitated access to additional therapies in a significant proportion of cases⁴.

UKGTN Testing Criteria: Minimum criteria required for testing to be appropriate as stated in the Gene Dossier.

Electroclinical Phenotype of Dravet Syndrome or clinical subtypes – several seizure types in one individual with onset in infancy, refractory to medication and with generalised spike and wave on EEG OR Infants less than 1 year with 2 or more prolonged hemiclonic febrile seizures in early infancy

Links

Competing interests

The authors have declared that no competing interests exist.

Test Description

When needle passes are made for cytologic analysis of sonographically suspicious thyroid nodules, two dedicated passes also are made for Afirma GEC analysis and immediately stored in nucleic acid preservative solution. If the FNA cytopathology is nondiagnostic, benign, or malignant, the sample collected for the Afirma GEC is discarded. If the FNA cytopathology is indeterminate, the Afirma GEC sample undergoes RNA extraction and nucleic acid amplification. Processed Afirma GEC samples are hybridized to a custom Afirma Thyroid microarray and analyzed with a classification algorithm using linear support vector machine logic to produce either a “Benign” or “Suspicious” test result. About 10% of FNA samples have inadequate RNA yield or quality and are reported by the Afirma GEC as “No Result”.¹⁶

Public Health Importance

The incidence of thyroid cancer in the U.S. has risen dramatically. In 2009, there were 37,200 new cases of thyroid cancer, while in 2013, it is anticipated there will be 60,220 new cases.^17,18 At the same time, there has been an increase in the utilization of thyroid FNA and subsequent thyroid surgery.¹⁹ The prevalence of thyroid nodules increases with age and is more common in females. Approximately 50% of women ≥50 years have at least one thyroid nodule based on published ultrasound and autopsy series.²⁰ Two thirds of thyroid nodules have benign FNA cytopathology and monitoring is implemented, whereas those with indeterminate or malignant cytology are generally referred for surgery. Because thyroid nodules with indeterminate FNA cytopathology have a 25% risk of malignancy when resected, 75% of these operations will likely be on nodules determined to be benign post-operatively.^5,6 Thyroid surgery is associated with potential complications, including temporary and permanent hypocalcemia, recurrent laryngeal nerve injury (with voice change, dysphagia, and potentially airway compromise if bilateral), and bleeding, with an incidence as high as 2-10%.^21,22,23 While there is strong evidence that high volume thyroid surgeons on average have fewer complications than low volume counterparts, 50% of thyroid operations in the U.S. are still performed by surgeons who perform ≤5 thyroidectomies/year.²⁴ Hypothyroidism is an expected sequelae of thyroid surgery, with patients requiring life-long thyroid hormone supplementation or replacement therapy.

Published Reviews, Recommendations and Guidelines

Systematic evidence reviews. Palmetto Government Benefits Administrators (Palmetto GBA), the CMS Medicare Administrative Contractor with oversight for the Afirma GEC, has published its assessment of the test as an update to its local coverage article on molecular diagnostics. This review determined that the test meets criteria for analytical and clinical validity, and clinical utility as a reasonable and necessary Medicare benefit, effective January 1, 2012.²⁵

Recommendations by independent groups. As part of the CLIA Laboratory licensure process, the analytical and clinical validation data for the Afirma GEC were independently assessed by reviewers from the California Department of Public Health and the New York State Department of Health.^26,27 Both of these reviews resulted in a favorable licensure outcome.

Guidelines by professional groups. The National Comprehensive Cancer Network (NCCN) thyroid carcinoma guidelines were updated in December, 2012 to state “Molecular diagnostics may be useful to allow reclassification of follicular lesions (follicular neoplasm or follicular lesion of undetermined significance) as more likely to be benign or more likely to be malignant…If molecular testing predicts a risk of malignancy comparable to the risk of malignancy seen with a benign FNA cytology (approximately 5% or less), consider observation.” The NCCN guidelines for abnormal gene/gene expression profile testing are associated with Level of Evidence 2A (lower level evidence, uniform NCCN consensus that the intervention is appropriate).^8,28

Evidence Overview

Analytical Validity: test accuracy, reliability in measuring differences in expression of relevant genes (analytic sensitivity and specificity), and robustness.

• Building on an earlier study by Chudova et al.¹⁴, a large collaborative study by Walsh et al. reviewed over 30 sub-studies on the Afirma GEC, finding high analytic sensitivity and specificity after dilution of thyroid neoplasm FNA samples with adjacent normal tissue or benign neoplasms (such as nodular hyperplasia and lymphocytic thyroiditis), as well as dilution with blood and genomic DNA, respectively.^14,29

• High reproducibility was found in studies of interlaboratory concordance (R² 0.98), as well as intra-assay, inter-assay, and intra-nodule concordance (R² 0.99, 0.98, and 0.95, respectively).²⁹

• The assay was robust to a wide range of temperature, storage and stressed shipping conditions and was reproducible across different operators, runs, and reagent lots with routine use of control reagents/samples for in-process Quality Control monitoring.²⁹

Clinical Validity: test accuracy in correctly determining which indeterminate cytology FNA biopsies are benign compared to expert surgical histopathology.

• Two prospective multicenter studies evaluated the negative predictive value (NPV) for the Afirma GEC, which is the key diagnostic performance metric used to make a decision to monitor patients in lieu of referral for diagnostic thyroid surgery.^14,15 Both studies utilized diagnosis of the surgical pathology specimen by a central panel of blinded academic endocrine histopathologists as the reference standard for clinical validation. Approximately one quarter of study sites were academic and three quarters were community-based (total sites n=49).

• In both studies, NPV was >94% for cytologically indeterminate thyroid FNAs with atypia/follicular lesions of undetermined significance and follicular/Hürthle cell neoplasm diagnoses, but lower when the cytology was suspicious for malignancy. Overall, the NPV for the Afirma GEC was similar to the NPV for a resected thyroid nodule with benign cytopathology.^5,30

• In the second, larger study¹⁵, although the NPV was 95% for all indeterminate cytology nodules (at the 24% risk of malignancy expected in clinical practice), the NPV was 85% for FNAs with a cytopathology diagnosis suspicious for malignancy. Sensitivity was 92% for indeterminate nodules overall. Clinical specificity for indeterminate nodules rose from 0% for cytopathology alone to 52% with the Afirma GEC, meaning that half of the benign nodules with indeterminate cytopathology could be identified with this pre-operative test.

Clinical Utility: net benefit of test in improving health outcomes by allowing recommendation for monitoring instead of a diagnostic surgery on benign thyroid nodules.

• Duick et al. was a retrospective analysis based on chart review of 368 patients (395 indeterminate thyroid nodules) cared for by 51 physicians at 21 practice locations in 11 states. When compared to historical controls, the relative rate of operation on cytologically indeterminate nodules fell 90%, from 74% historically to 7.6% with Afirma GEC benign results (p < 0.001).¹⁶ The comparisons between historical controls and operative rates based on the binomial test achieved more than 99% power in detecting a less than 25% change.³¹ With Afirma GEC benign results, in 92.4% of the cases, physicians recommended watchful waiting in lieu of a diagnostic thyroid resection. The residual rate of surgery was similar to the historic 9% resection rate on cytologically benign thyroid nodules and may relate to other clinical parameters and issues of patient and physician preference, including issues related to nodule size.⁵

• These results were consistent with a previously conducted web- and mail-based opinion survey of 84 physician practices, with a mean of 89% of physicians reporting that they recommended watchful waiting for patients with cytologically indeterminate FNAs but benign Afirma GEC results.³²

Limitations. In the clinical validity studies, thyroid FNAs with indeterminate cytology diagnoses that were cytologically suspicious for malignancy did not have sufficiently low NPV to generally recommend monitoring.¹⁰ Secondly, the Duick et al clinical utility study used historical rather than contemporary controls.¹⁶ Although historical controls were used, these were appropriately validated based on a meta-review of 11 recent thyroid pathology studies where thyroid nodule evaluation was similar to current community practice.⁵

Conclusions

The clinical validity studies reviewed would fall into the Level 1 category in the Evaluation of Genomic Applications in Practice and Prevention (EGAPP) hierarchy of data sources and study designs, based on two prospective, multicenter, double blinded cohort studies.³³ A validated clinical decision rule was based on classification concordance of Afirma GEC benign results with blinded expert surgical pathology benign diagnosis. ¹¹ Sensitivity in most indeterminate FNAs was high enough (92%) to achieve a NPV of 94-95%, which is comparable to a thyroid nodule that is benign on cytopathology but undergoes surgical resection (93-94%).^5,30 However, the Afirma GEC NPV for FNAs where the cytopathology was suspicious for malignancy was only 85%. While this lowers the residual risk of malignancy from 62% to 15%, surgical consultation still should be planned in these patients. The test utilized interlaboratory comparisons in a large collaborative study, an EGAPP criterion for Level 1 hierarchy of analytic validity study design.³⁴ Improvement in the net health outcome is based on avoidance of surgery in patients with indeterminate thyroid FNA cytology that would have been found to be benign on surgical pathology. Clinical utility of the Afirma GEC was evaluated in clinical practice, outside of the investigational setting, in an opinion survey and a controlled study consistent with EGAPP Level 3 criteria for clinical utility study design. The Afirma GEC potentially can improve the net health outcome by providing an alternative to diagnostic thyroid surgery, and therein the risk of downstream complications of surgery, in patients with benign thyroid nodules but indeterminate FNA cytopathology. In summary, the studies reviewed here regarding clinical and analytic validity, and clinical utility support recommendation for offering patients the alternative of using the Afirma GEC to monitor patients in lieu of thyroid resection when applied in the specific case of thyroid FNAs where there is indeterminate cytopathology: atypia/follicular lesion of undetermined significance, and follicular/Hürthle cell neoplasm.

Competing interests

Dr. Lanman is an employee of Veracyte, Inc. The other co-authors have no conflict of interest.

Test Description

Cardiac ion channelopathies result from adverse alterations in genes that code for protein subunits of cardiac ion channels¹. The literature differentiates these channelopathies in terms of their subtypes (e.g., for long QT syndrome, LQT1, LQT2, LQT3, LQT4, LQT5, LQT6, …, LQT13) and the name of the gene affected (KCNQ1 for LQT1; KCNH2 for LQT2; SCN5A for LQT3; ANK2 for LQT4; KCNE1 for LQT5; KCNE2 for LQT6, …, KCNJ5 for LQT13). Genetic testing is available through a number of commercial and university-based genetic diagnostics laboratories^1141516. Transgenomic (formerly PGx Health and Genaissance), GeneDx, and Correlagen all offer genetic assays allowing genotyping of LQTS, SQTS, CPVT, and BrS. For example, panels offered by Transgenomic¹⁷ and GeneDx¹⁸ test for 13 and 12 LQTS genes, respectively, KCNQ1, KCNH2, andSCN5A (LQT1-3) mutations being most frequent; 3 SQTS genes – mutations in the KCNH2, KCNQ1, andKCNJ2 genes; 4 (Transgenomic) and 2 (GeneDx) CPVT genes, RYR2 mutations being most frequent; and 7 BrS genes, SCN5A mutations being most frequent. The Masonic Medical Research Laboratory covers these conditions for research purposes, but specializes in acquired as opposed to congenital long QT syndrome¹⁹. Samples are collected in an EDTA-containing tube; the DNA is isolated from fresh whole blood. DNA amplification by polymerase chain reaction (PCR) is then used to generate templates for direct sequencing. DNA, frozen blood, saliva, and other tissue samples such as buccal specimens have recently been accommodated by some labs¹⁷¹⁸. Detailed results are provided for each test panel, typically with a description of known literature on any identified mutation and its likelihood to cause disease. Focused testing for a single mutation relevant to a patient’s family, rather than a panel of potentially involved genes, is available at a lower cost.

Public Health Importance

The public health importance of these inherited arrhythmia syndromes is highlighted by their potential lethality, mostly due to ventricular tachyarrhythmias²⁰. Long QT syndrome may be responsible for as many as 3,000 unexpected deaths in children and young adults annually in the United States²¹. Catecholaminergic polymorphic ventricular tachycardia carries a high mortality in untreated cases⁵⁷. The current state of technology, however, does not enable arrhythmia screening of broad, potentially at-risk populations – athletes, all children, all patients exposed to QT-prolonging drugs, or all patients with a history of syncope⁶¹⁶. Subpopulations of sudden death victims are of great interest. It is estimated that approximately 25-35% of autopsy-negative or unexplained sudden cardiac death in the young, and 10% of sudden infant death (SIDS) may be attributable to mutations in either LQTS or CPVT susceptibility genes²². Brugada syndrome may be responsible for 4-15% of unexpected sudden deaths, particularly in individuals with an apparently normal heart²³²⁴²⁵.

In the case of LQTS, the specific genotype has prognostic value¹⁶. In studies ranging from 250 to 300,000 genotyped individuals through age 60 drawn from the International LQTS Registry, the LQT3 genotype has demonstrated a 5- to 8-fold higher risk for life-threatening events compared to the LQT1 and 2 genotypes³⁶. The LQT2 genotype displays a risk intermediate between LQT1 and LQT3³⁶, though the statistical strength and ordering of this relationship depends on age, sex, and source of the cohort¹⁰³⁷. Subgroup analyses have more precisely defined the impact of QTc duration and genotype in children and adolescents with LQT1 and 2³⁸.

Published Reviews, Recommendations and Guidelines

Systematic evidence reviews

The Hayes Inc. Genetic Test Evaluation (GTE) Program has prepared three evaluations of ion channelopathies: long QT syndrome, CPVT, and Brugada syndrome. These reviews were based on studies of primary literature retrieved from PUBMED and Embase in the following date ranges: LQTS – January 1, 1996 to June 16, 2009; CPVT – January 1, 1996 to February 10, 2010; and Brugada syndrome – January 1, 1996 to August 3, 2010³⁹. DNA Direct, Inc. has published two systematic reviews of LQTS genetic testing through its Genomic Medicine Institute. The reviews, which the company makes available with interactive tools and decision support, are based on studies of primary patient data published between 2001 and 2008⁴⁰. BlueCross BlueShield Association and Kaiser Permanente published a systematic review of genetic testing for long QT syndrome through the BlueCross BlueShield Technology Evaluation Center (BCBS Tec)⁴¹. The review was based on studies of primary patient data appearing in Medline and PUBMED published between 1990 and October 2007. The Australia and New Zealand Horizon Scanning Network released a Horizon Scanning Report on genetic testing for congenital long QT syndrome based principally on nine peer-reviewed articles, and detailing safety (e.g., false negatives), effectiveness, clinical utility (cost-effectiveness), and ethical (consent and privacy, harms from testing, access) considerations⁴². Investigative teams have published comprehensive clinical-epidemiologic reviews of both long QT syndrome as a family and several of the other channelopathies under Human Genome Epidemiology (HuGE) reviews (e.g., long QT syndrome²¹) and GeneReviews (e.g., Brugada syndrome⁴³).

Recommendations by independent groups

Two independent reviews have concluded that genetic testing has diagnostic value, including for the identification of asymptomatic heterozygotes, for all four syndromes; definitive prognostic value for LQTS and weak or contingent prognostic value (depending on the nature of the findings) for CPVT and BrS; and practical value in the determination of therapy for LQTS alone¹³⁰. One of the teams also published an “evidence-based” genetic testing scoring system to compare the relative clinical value of genetic testing over these conditions (LQTS > CPVT > BrS > SQTS)⁴⁴. SQTS’s low score is associated with paucity of data, this condition having first been described in 2000¹⁵. Criteria on when to screen for each channelopathy, based on consensus documents and primary literature from 1995 onward, are laid out by Tzou and Gerstenfeld¹⁵. This review concluded that genetic testing can lead to genotype-specific therapy in the case of LQTS, and decisions about ICD placement for malignant arrhythmias associated with CPVT and SQTS¹⁵.

Guidelines by independent groups

A 2007 consensus report by the U.S. National Heart, Lung, and Blood Institute and the Office of Rare Diseases on gene mutations affecting ion channel function concluded that genetic testing for LQTS must be combined with clinical evaluation, and noted lack of clarity in the proportion of SQTS cases that might be explained by the corresponding KCNH2, KCNJ2, and KCNQ1 genes⁴⁵. A 2011 Heart Rhythm Society (HRS) / European Heart Rhythm Association (EHRA) consensus statement further states that LQTS genetic testing is recommended for any asymptomatic patient with idiopathic (not attributable to QT prolonging disease states or conditions) QTc values > .48 s. (prepuberty) or > .50 s. (adult), and may be considered for QTc values >= .46 and .48, respectively⁶. (QTc = “heart rate-corrected QT interval,” as per the Bazett formula⁴.) The Heart Rhythm UK Familial Sudden Death Syndromes Statement Development Group published in 2008 a position statement on genetic testing for sudden cardiac death syndromes based on a comprehensive review of English language publications, grading of the evidence, and secondary review of the evidence by an external committee³. The Group followed with a position statement on ICD placement for these conditions based on risk of SCD⁴⁶. The first position statement and the more recent HRS/EHRA report recommend genetic testing for all patients with a firm diagnosis of congenital LQTS and those with clinical features of CPVT (due to its severity, despite an acknowledged lower clinical sensitivity), but that expert clinical and family history assessment are needed when genetic testing is undertaken for borderline LQTS cases and known or suspected cases of BrS. Practice guidelines from the American College of Cardiology / American Heart Association / European Society of Cardiology⁴⁷ have noted an evolving role for genetic testing of LQTS in risk stratification and clinical decision making. Both independent reviews and professional society guidelines agree that genetic testing by itself is not recommended in making a diagnosis or prognosis for BrS, though it may be used to support clinical diagnosis, and early detection of at-risk relatives⁶²⁵⁴⁸. Several HRS / EHRA consensus statements clarify that genetic testing can be useful in patients clinically suspected of having BrS with a type 1 (“coved” ST segment elevation) ECG pattern, but that it is not indicated in the setting of an isolated type 2 (less specific, “saddleback” ST elevation) or 3 (either shape but less pronounced elevation) pattern⁶²⁰²⁵.

Evidence Overview

Analytic Validity:

Commercially available channelopathy genetic testing (CGT) for LQTS, SQTS, CPVT, and BrS involves direct sequencing of protein-coding portions and flanking regions of targeted exons following PCR amplification. Sequencing is performed in both forward and reverse directions. Sequences are analyzed for heterozygous and homozygous variants using public reference sequences¹⁷⁴¹. For the FAMILION® test, variants detected in the initial analysis are confirmed by repeating the sequencing twice in the forward and three times in the backward direction. Transgenomic reports a CGT error rate of < 1% ¹⁷. GeneDx indicates a 98% “technical sensitivity” for assessment of each of the four conditions¹⁸. The John Welsh Cardiovascular Laboratory at Baylor College of Medicine makes available genetic diagnostic testing using DNA sequencing analysis for KCNJ2 (LQT7) and CAV3 (LQT9) mutations. This facility reports approximately 99% detection of the exons that are sequenced⁴⁹. Failure to detect variants in the laboratory setting is attributed to refractoriness of the amplicon to analysis by direct DNA sequencing or real-time PCR detection, sample mishandling, sample tracking errors, errors in data analysis, and other gene specific issues (see Clinical Validity)¹⁷. The FAMILION® analytic specificity approaches 100% for “Class I” (deleterious or probably deleterious) mutations, and approximates 95% for “Class II” mutations (of uncertain clinical significance – possibly deleterious)⁴¹.

Clinical Validity:

The near perfect analytic sensitivities of these assays must be contrasted with the ability to detect genetic variants in the clinical setting. The yield for the first 2,500 consecutive unrelated cases referred by physicians for commercially available long QT syndrome genetic testing (low-, intermediate-, and high-pretest probability) was 36%; values in the literature range between 33 and 39%⁵⁰. Tester et al. reported greater yields of 72 to 78% using the 5 major LQTS genes for patients with the highest clinical probability for LQTS (Schwartz-Moss score >= 4 ⁴)³⁷. Transgenomic reports a 75 to 80% yield for such patients based upon multiple data sets¹⁷⁵¹. False negatives may be explained by a number of factors, including the existence of private mutations³⁷, the presence of non-targeted exons¹⁷ and relevant introns outside tested splice sites¹⁶, and the existence of large deletion and duplication mutations (generally on the scale of a whole exon or more)⁷¹⁷. LQTS is in the upper range among cardiac conditions in terms of number of affected genes (13) and allelic mutations (>800) while BrS lies in the midrange (8 genes and >400 allelic mutations). By comparison, current compendia of arrhythmogenic right ventricular dysplasia/cardiomyopathy list 9 affected genes and >400 allelic mutations, while Marfan syndrome has ~600 allelic mutations in the FBN1 fibrillin gene.

Authors have reported various mutational hotspots in the case of LQTS arising either independently or due to founder effects, which because of their severity may aid risk stratification²¹. Interpretation of results for probands and family members is complicated by variable penetrance in family members with the same genotype⁵², and the possibility of compound mutations, observed in 4 to 10% of mutation positive individuals^37505354. Estimated positive predictive values (EPV – percent of mutations found in definite cases that would cause the condition) for LQTS genetic testing based on an assessment of > 1300 unrelated index cases with Schwartz score >= 4 or QTc >= .48 s. are: 96% (95% C.I., 94-98) for KCNQ1; 93% (89-95) for KCNH2; and 63% (40-77) for SCN5A⁵⁵. Nonmissense mutations have an EPV > 99% regardless of location⁵⁵. EPVs for missense mutations range from 0% in the interdomain linker of SCN5A to 100% in the transmembrane/linker/pore regions of KCNH2. Similar figures are lacking for the other channelopathies.

Several groups have investigated ECG genotyping (to be distinguished from use in initial diagnosis) through T-wave morphology as a means of reducing the cost of LQTS mutational genotyping⁵⁶⁵⁷⁵⁸. Sensitivity is highly dependent on the gene being assessed, and varies between 92% (LQT2) and 47% (LQT3)⁵⁷⁵⁸. QTc interval alone lacks predictive value for genotyping⁵⁶⁵⁷.

For channelopathy genetic testing (CGT) of the other inherited arrhythmia syndromes, Transgenomic reports a clinical sensitivity of 65-75% (CPVT), 25-40% (BrS), and 15-20% (SQTS)¹⁷. GeneDx¹⁸ and the Correlagen CardioGeneScan®⁵⁹ also test for these conditions. The GeneDx and Correlagen test information / FAQ sheets provide rough estimates of clinical sensitivity; figures from Correlagen are also reported with the assays. Some commercial laboratories only provide CPVT screening of a limited number of select exons encompassing critical RYR2 regions, which can contribute to false negatives³⁰. Disagreement exists on the impact of combining other variables with mutational results in order to increase clinical sensitivity; results vary by arrhythmia syndrome, genotype, and size of the study population¹⁰²⁷⁶⁰.

Mutational analysis of 27 SCN5A exons on cases from BrS databases at 9 international centers resulted in yields of 11-28%³⁵⁴³, lower than the 25-40% figure Transgenomic reports¹⁷⁴³⁶⁰. The literature suggests that for BrS about a quarter of the patients or fewer carry an SCN5A mutation (commercial assays also include other lower frequency mutations)¹²²⁵. The low clinical sensitivity of genetic testing for Brugada syndrome, due to the low gene frequency for the most prevalent type of mutation (involving SCN5A), incompleteness of known allelic variants, and the role of an organic substrate in many cases, limits its diagnostic capability¹⁵⁶¹⁶². Genetic testing for BrS is more often used for confirmatory purposes¹. Higher yields for the various channelopathies accrue in families with at least one recorded case of sudden unexpected death, permitting detection of potentially affected relatives²⁴.

Clinical Utility:

Genetic analysis has been found useful in risk stratification of LQTS patients and detection of at-risk family members, though current knowledge of CPVT genetic variants does not in itself does not yield prognostic information^{15222730364148}. “Intragenic risk stratification” based on mutation type and location, and cellular function, particularly for the LQT1 and 2 genotypes, has been an increasing part of large scale studies but is not yet in widespread practice^1611414763. A meta-analysis of 30 Brugada syndrome prospective studies by Gehi et al.⁶⁴ concluded family history of SCD and presence of an SCN5A mutation by themselves are insufficient to predict risk for cardiac events in BrS. Instead, debates about risk stratification for patients with this condition have taken place largely on the electrocardiographic front⁶⁵. The presence of spontaneous ST-segment elevation, particularly with a type-1 pattern, in conjunction with a history of syncope remains the strongest predictor of BrS risk².

Mutational information can serve as an adjunct to clinical and phenotypic assessment of genotype in therapeutic decisions for LQTS³⁰³⁶⁴⁸, though this role is not without controversy^1012164166. Clear evidence exists for genotype-specific therapy in management of the LQT1, 2, and 3 genotypes, but it is less substantiated for the other LQTS genotypes due to their rarity^111213366768. Disagreement exists on the use of intragenic, site-specific information to predict actual clinical phenotype or response to therapy for the LQTS genotypes¹¹³⁷. In analyzing retrospective data on 27 of 128 LQTS patients in the 3 to 13 year-old age range who received either pacemakers or ICDs as therapy, investigators in a 3-area study (Utah, British Columbia, and Arizona) found no association between device placement decisions and implementation of genetic testing. However, the authors also noted that all patients with SCN5A (LQT3) mutations had therapeutic devices, though only 1 of 30 with a KCNQ1 (LQT1) mutation had one⁶⁹. The use of genetic testing in therapeutic decision making for the other inherited arrhythmia syndromes is not yet substantiated but appears promising^16304870. ECG remains the principal tool for BrS diagnosis and therapeutic follow-up²¹⁶²⁵.

Several teams have evaluated step-wise or tiered strategies also including phenotypic assessment to increase the efficiency of LQTS and CPVT genetic testing^17537172. Targeted screening based on phenotypic information can lower the cost of more comprehensive LQTS genetic testing by ~60% ⁷¹. Phillips et al. found LQTS genetic testing more cost-effective than not testing for symptomatic index cases at an estimated cost of $2,500 per year of life saved ($50,000 per year of life saved is often used as a standard threshold)⁷³⁷⁴. Bai et al., in looking at 546 patients referred to a large consortium of LQTS research laboratories, found the highest yield (64%) and lowest cost ($8,418 per positive genotyping) for patients with ECG-confirmed LQTS, but diminishing yields and increasing costs for patients with borderline QTc intervals and those with normal intervals but a positive family history for SCD⁶⁰. They recommended genetic testing be prioritized to those with a “conclusive diagnosis” of LQTS. Genotyping of individuals with a conclusive diagnosis of CPVT, and of patients with type 1 BrS ECG with atrioventricular block was also found to be cost-effective. Apart from this report, the host of cost-effectiveness analyses for Brugada syndrome deal with the issue of implantation of ICDs.

Perez et al. used a Markov model to assess the cost-effectiveness of different strategies for testing then treating an asymptomatic 10 year-old first degree relative of a patient with clinically evident LQTS⁷⁵. They concluded that genetic testing is moderately expensive, at $67,400 per quality adjusted life year saved, but improves with higher clinical suspicion of the proband, number of relatives tested, and stronger family history of sudden death.

CGT is covered to different extents by insurance providers, ranging from denial to 100% coverage, with most covering at least 50 to 75% of the cost²². Some health plan policies explicitly limit genetic test reimbursement to LQTS while excluding the other channelopathies⁷⁶.

Public Health Ethical, Legal, and Social (PHELSI) Considerations:

Genetic testing for cardiac channelopathies is laden with ethical issues, including the search for a balance between individual privacy vs. alerting at-risk family members, as well as psychosocial issues inherent in informing individuals of their risk¹⁷⁷⁷⁸. A review by Wren provides general recommendations for genetic testing of children in families with a history of SCD, and offers ethical considerations in childhood genetic testing for Brugada syndrome⁷⁹. Use of pacemakers and ICDs in children, while often done under situations of medical necessity, can involve trade-offs between benefit and adverse effects⁴⁵⁶⁹⁸⁰. Screening for LQTS mutations in particular racial-ethnic groups deserves further ethical analysis²¹. Personalization of drug treatment through determination of individually specific genotype remains a hoped-for future direction in the field¹⁷⁵. Risk assessment for inherited heart arrhythmias is also becoming a part of direct-to-consumer genetic testing, an area subject to increasing attention by policy makers.

Links

The commercial and university-based laboratories cited above are approved under the Clinical Laboratory Improvement Amendments (CLIA). Several, but not all, are accredited by the College of American Pathologists (CAP). These assays are developed and validated in-house, thus do not require FDA approval.

Relevant web sites:

Competing Interests

The authors have declared that no competing interests exist.