- Email: [email protected]
Clinical and ECG variables to predict the outcome of genetic testing in hypertrophic cardiomyopathy
European Journal of Medical Genetics xxx (xxxx) xxxx
Contents lists available at ScienceDirect
European Journal of Medical Genetics journal homepage: www.elsevier.com/locate/ejmg
Clinical and ECG variables to predict the outcome of genetic testing in hypertrophic cardiomyopathy Tomas Robynsa,b,c,∗, Jeroen Breckpotd, Dieter Nuyense, Bert Vandenberka,b, Anniek Corveleynd, Cuno Kuiperid, Lucas Van Aelsta,b, Johan Van Cleemputa,b, Rik Willemsa,b a
Department of Cardiovascular Diseases, University Hospitals Leuven, Belgium Department of Cardiovascular Sciences, University of Leuven, Belgium c The University Hospitals of Leuven are Member of the European Reference Network for Rare and Complex Diseases of the Heart (ERN GUARD-HEART), European Union d Center for Human Genetics, University Hospitals Leuven, Belgium e Department of Cardiology, Ziekenhuis Oost Limburg, Genk, Belgium b
A R T I C LE I N FO
A B S T R A C T
Keywords: Hypertrophic cardiomyopathy Genotype phenotype relation Genetic testing
Knowledge on the influence of specific genotypes on the phenotypic expression of hypertrophic cardiomyopathy (HCM) is emerging. The objective of this study was to evaluate the genotype-phenotype relation in HCM patients and to construct a score to predict the genetic yield based to improve counseling. Unrelated HCM patients who underwent genetic testing were included in the analysis. Multivariate logistic regression was performed to identify variables that predict a positive genetic test. A weighted score was constructed based on the odds ratios. In total, 378 HCM patients were included of whom 141 carried a mutation (global yield 37%), 181 were mutation negative and 56 only carried a variant of unknown significance. We identified age at diagnosis < 45 years, familial HCM, familial sudden death, arrhythmic syncope, maximal wall thickness ≥20 mm, asymmetrical hypertrophy and the absence of negative T waves in the lateral ECG leads as significant predictors of a positive genetic test. When we included these values in a risk score we found very high correlation between the score and the observed genetic yield (Pearson r = 0.98). MYBPC3 mutation carriers more frequently suffered sudden cardiac death compared to troponin complex mutations carriers (p = 0.01) and a similar trend was observed compared to MYH7 mutation carriers (p = 0.08) and mutation negative patients (p = 0.11). To conclude, a simple score system based on clinical variables can predict the genetic yield in HCM index patients, aiding in counseling HCM patients. MYBPC3 mutation carriers had a worse outcome regarding sudden cardiac death.
1. Introduction Hypertrophic cardiomyopathy (HCM) is the most common hereditary heart disease with an estimated prevalence of 1 in 500 to 1 in 200 (Maron et al., 1995; Semsarian et al., 2015). The first gene implicated in the disease was MYH7, identified in 1990 (Geisterfer-Lowrance et al., 1990). Since then multiple other genes have been labeled as causal for the disease (Maron et al., 2012). Albeit this tremendous progress, the yield of genetic testing remains incomplete with an estimate of 40% of patients in whom a mutation is identified (Lopes et al., 2013). Furthermore, interpretation of genetic variants remains difficult, especially in the less commonly mutated HCM genes. Genotype-phenotype relation for HCM has been studied in several cohorts, and were compiled in a meta-analysis (Lopes et al., 2013). Subsequently, two independent North American groups and an Australian group constructed a scoring
∗
system based on clinical variables to predict the chance of identifying a mutation in HCM patients (Murphy et al., 2016; Gruner et al., 2013; Ingles et al., 2013). However, not a single electrocardiographic parameter was included in these analyses. Furthermore, these tools have not been developed for European HCM patients. Yet, the factors that infer a higher risk of carrying a causal mutation might differ as much as the predictors of adverse outcome in HCM patients between European and American cohorts (Maron et al., 2015; Vriesendorp et al., 2015). Correct counseling of these patients regarding chances of identifying a causal mutation is essential to prevent unrealistic ideas on genetic testing and scoring systems therefore aid in this process. Therefore, we aimed to (1) explore genotype-phenotype relation in HCM patients in our Belgian cohort, (2) identify predictors of a positive genetic test and (3) construct a scoring system to predict genetic yield in HCM patients.
Corresponding author. University hospitals Leuven, Department of Cardiovascular diseases, Herestraat 49, 3000, Leuven, Belgium. E-mail address: [email protected] (T. Robyns).
https://doi.org/10.1016/j.ejmg.2019.103754 Received 8 November 2018; Received in revised form 23 August 2019; Accepted 8 September 2019 1769-7212/ © 2019 Elsevier Masson SAS. All rights reserved.
Please cite this article as: Tomas Robyns, et al., European Journal of Medical Genetics, https://doi.org/10.1016/j.ejmg.2019.103754
European Journal of Medical Genetics xxx (xxxx) xxxx
T. Robyns, et al.
2. Methods
2.4. ECG variables
2.1. Patient selection
Twelve lead ECG's were recorded using MAC 5500 (Marquette, GE Healthcare, Chicago, Illinois, USA). Automated analysis was done using the widely used ‘GE Marquette 12SLTM ECG Analysis Program’. As such, automated measurements of PR interval, QRS duration, RR interval and QT interval corrected for heart rate with Fridericia's (QTcF) formula were obtained. Index of cardio-electrophysiological balance (iCEB) was defined as QT interval divided by QRS duration (Robyns et al., 2016). All ECG leads were assessed for negative T waves > 1 mm. Anterior leads were defined as leads V1–V4, lateral leads as leads V5,V6, I, aVL and inferior leads as leads II, III and aVF. Presence of negative T waves in a specific area of the heart was defined as ≥ 2 leads in either anterior, lateral or inferior leads with negative T waves. Finally, left ventricular hypertrophy on the ECG was assessed by the sum of the maximum S wave amplitude in leads V1 or V2 and the maximum R wave amplitude in leads V5 or V6 (part of Framingham LVH score) (Levy et al., 1990).
The database of the center for hereditary heart disease of the university hospitals of Leuven was retrospectively analyzed to select all unrelated patients diagnosed with HCM, from the beginning of the database in 2002 until August 2016. HCM was defined according the 2014 ESC guidelines as unexplained left ventricular hypertrophy of ≥15 mm (Elliott et al., 2014). We selected only index patients for whom both ECG and echocardiography were digitally available. Our study complied with the Declaration of Helsinki, and the research protocol was approved by the local ethics committee.
2.2. Genotyping and variant interpretation Patients were only included if at least the 3 main genes (MYBPC3, MYH7 and TNNT2) that were selected at our center in the early days of HCM genotyping were evaluated. Genotyping was done (1) by Denaturing High Pressure Liquid Chromatography (DHPLC) of the 3 main genes followed by direct Sanger sequencing of abnormal results until 2009, (2) by direct Sanger sequencing of the 3 main genes between 2009 and 2013 and (3) by targeted gene panel testing from then on. In total, 217 patients were evaluated with the extended panel. The panel contained following genes: ACTC1, ACTN2, ANKRD1, CALR3, CASQ2, CAV3, CRYAB*, CSRP3, DES, FHL1*, FLNC*, GLA, JPH2, LAMP2, LDB3, MYBPC3, MYH6, MYH7, MYL2, MYL3, MYLK2, MYOZ2, MYPN*, NEXN*, PLN, PRKAG2, RYR2, TCAP, TNNC1, TNNI3, TNNT2, TPM1, TTR*, VCL. A subset of patients (N = 107) were evaluated using a panel that included 6 additional genes indicated by an asterisk (*). Moreover multiplex Ligation-dependent Probe Amplification (MLPA) was performed for MYBPC3 in all patients. All genetic variants were reassessed by the criteria proposed by the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG-AMP criteria) as pathogenic variant, likely pathogenic variant, variant of unknown significance (VUS), likely benign variant or benign variant (Richards et al., 2015). To evaluate the strength of co-segregation, we used the criteria proposed by Jarvik and Browning (2016). A mutation was defined as a pathogenic or likely pathogenic variant according these criteria.
2.5. Statistical analysis First we compared clinical, echocardiography and ECG parameters between mutation positive, mutation negative and VUS positive HCM patients using one way ANOVA with Tukey's post hoc testing for continuous variables and using chi square test for categorical variables. The genotype-phenotype relation was evaluated by comparing MYBPC3 mutation carriers, MYH7 mutation carriers and carriers of a mutation in the troponin complex (TNNC1, TNNI3 and TNNT2) using one way ANOVA with Tukey's post hoc testing for continuous variables and using chi square test for categorical variables. The number of mutation carriers in other genes was too small to include in this analysis. KaplanMeier curves with log rank test were constructed to evaluate survival regarding the composite endpoint with log rank test. Evaluation of predictors of carrying a mutation was done using binary logistic regression (enter method, variables with p-value < 0.1 were entered into the multivariate model). A score system was developed based upon the odds ratios (OR) of significant variables in the multivariate model: variables with OR > 3 were awarded 2 points, OR 1–3 were awarded 1 point and OR < 1 were awarded ‘-1’ point. A Pearson correlation coefficient was calculated to document the correlation between the risk score and the observed genetic yield. 3. Results
2.3. Definition clinical variables 3.1. Genetic background patient population Familial sudden death was defined as a sudden death before age 40 in a first or second degree relative, or as sudden death at any age in a family member with an established diagnosis of HCM. Familial HCM was defined as a family member with definite diagnosis of HCM or with an unexplained maximal wall thickness (MWT) ≥ 12 mm based upon a detailed family history and review of available clinical data (Elliott et al., 2014). MWT was either measured with transthoracic echocardiography (TTE) or cardiac MRI. Three different morphology types of hypertrophy were defined: asymmetrical, symmetrical and apical. Left ventricular outflow tract obstruction (LVOTO) was measured either in rest or after a Valsalva maneuver. Obstructive HCM was defined as LVOTO > 30 mmHg. Systolic anterior motion (SAM) of the mitral valve was also assessed from TTE. Arrhythmic syncope was defined as unexplained syncope with sudden onset. Heart failure episodes were episodes of clinical congestion necessitating diuretic therapy. The presence of arterial hypertension was extracted from the medical records. If 24 h holter recordings were available, non-sustained ventricular tachycardia (nsVT) was defined as 3 consecutive ventricular premature beats > 120 BPM. We defined sudden cardiac death as a composite endpoint of out of hospital cardiac arrest, appropriate ICD shock or sustained ventricular tachycardia.
Out of 381 identified HCM index patients in whom ECG and echocardiography was available at our institution, 378 (99%) underwent genetic testing for at least MYBPC3, MYH7 and TNNT2. A definite pathogenic mutation was identified in 141 patients corresponding to a yield of 37% in the total population. Fig. 1 illustrates the yield per gene indexed for the number of patients in whom the gene was sequenced. The vast majority of mutations were identified in MYBPC3 (N = 80; 57%), MYH7 (N = 27; 19%) and troponin complex subunits with TNNC1 (N = 11; 8%), TNNI3 (N = 9; 6%) and TNNT2 (N = 4; 3%). A minority of patients carried a mutation in other sarcomere proteins: MYL2 (N = 2; 1.4%), MYL3 (N = 1; 0.7%) and TPM1 (N = 1; 0.7%). Finally 6 index patients (4%) were carrier of a mutation in PRKAG2. Details of the mutations can be found in Supplementary Table 1. Some mutations were found in ≥5 patients, including 4 mutations in MYBPC3 (c.1404del (N = 20), c.676_701dup (N = 7), c.2373dup (N = 5) and a deletion of exon 23 to 26 (N = 5)), 1 mutation in TNNC1 (c.430A > G; N = 11) and 1 mutation in TNNI3 (c.433C > T (N = 6)). Sixty-nine variants of unknown significance were identified in 63 probands. Out of these 69 VUS, 7 were identified in a patient in combination with a mutation, 4 were carrier of 2 VUS and 1 was carrier 2
European Journal of Medical Genetics xxx (xxxx) xxxx
T. Robyns, et al.
Fig. 1. Genetic yield per gene. This figure illustrates the percentage of mutations (dark bars) per gene indexed for the number of patients in whom the gene was genotyped. The bright bars illustrate the percentage of patients carrying a variant of unknown significance (VUS) in a specific gene indexed for the number of patients in whom the gene was genotyped.
carriers during a mean follow-up of 101 months (Supplementary Table 4). Survival analysis showed a statistically significant difference in outcome between mutation negative patients, MYBPC3 mutation carriers, MYH7 mutation carriers and troponin complex mutation carriers (p = 0.03; supplementary figure). This was based on worse outcome for MYBPC3 mutation carriers compared to troponin complex mutation carriers (p = 0.01), while there was only a trend in survival difference between MYBPC3 mutation carriers and MYH7 mutation carriers (p = 0.08) and mutation negative patients (p = 0.11). There was no difference in survival between mutation negative patients and MYH7 mutation carriers (p = 0.31) or troponin complex mutation carriers (p = 0.10). LVH was more pronounced in MYBPC3 mutation carriers compared to troponin complex mutation carriers, while the documentation of nsVT occurred at the same frequency (Supplementary Table 4). However, in MYH7 mutation carriers, nsVT was less frequently observed compared to MYBPC3 mutation carriers. ECG parameters were similar in the 3 subgroups, apart from longer QRS duration in MYH7 mutation carriers compared to MYBPC3 mutation carriers. Negative T waves in the lateral leads were less frequently observed in MYBPC3 mutation carriers compared to troponin mutation carriers.
of a mutation and 2 VUS. Details of the VUS can be found in Supplementary Table 2. Of the 181 mutation negative HCM index patients, 57 (31%) were only tested for MYBPC3, MYH7 and TNNT2, and 124 (69%) underwent HCM panel testing. All genetic variants mentioned in this paper were submitted to ClinVar database and accession numbers for (likely) pathogenic and VUS are found in Supplementary Tables 1 and 2 respectively. 3.2. Mutation negative versus mutation positive HCM Comparison of clinical, echocardiographic and electrocardiographic variables between mutation positive (N = 141), mutation negative (N = 181) and VUS positive (N = 56) HCM index patients is summarized in Supplementary Table 3. In short, mutation positive HCM index patients were significantly younger, had more frequent occurrence of familial SD, familial HCM, arterial hypertension and syncope compared to mutation negative HCM patients and HCM patients carrying only a VUS. Wall thickness was more pronounced and morphology was more frequently asymmetrical in mutation positive HCM compared to mutation negative HCM. There was no difference in ECG variables between the 3 groups, apart from less prevalence of negative T waves in the lateral leads in mutation positive HCM patients. None of the variables differed between mutation negative HCM and HCM patients who carried only a VUS. No difference in survival regarding the composite endpoint of sudden death was observed between mutation positive patients and the combined group of mutation negative patients and VUS positive patients (Log rank test p = 0.38) or between the 3 groups separately (Log rank test p = 0.10). SCD occurred at similar ages between those patients in whom a mutation was identified (46 ± 19 years) and those who remained genetically elusive (49 ± 22 years; p = 0.68).
3.4. Predictors of a positive genotype We evaluated predictors of identifying a mutation in HCM patients. For this analysis, we merged mutation negative HCM patients and HCM patients who carried only a VUS as mutation negative. In multivariate analysis, several parameters predicted identification of a mutation (Table 1). Based on the odds ratios, 2 points were allocated to familial history of HCM, 1 point to familial SD, age at diagnosis < 45 years, MWT ≥20 mm, asymmetrical hypertrophy and syncope, while 1 point was subtracted for the presence of negative T waves in the lateral leads and arterial hypertension. This resulted in a high correlation between the genetic yield and the risk score (Pearson r = 0.98; p < 0.001) which can also be appreciated from the nice spread of genetic yields across the spectrum of combinations of clinical risk factors from 0% if only negative T waves in the lateral leads or arterial hypertension were present to 100% if all positive risk markers were present (Fig. 2). The sensitivity of genetic testing and the genetic yield depending on
3.3. Genotype-phenotype relation in MYBPC3, MYH7 and troponin complex mutation carriers SCD occurred in 13 out of 80 MYBPC3 mutation carriers during a mean follow-up of 74 months (annual risk 2.6%), while it only occurred in 1 out of 27 MYH7 mutation carriers during a mean follow-up of 83 months (annual risk 0.5%) and in none of the troponin mutation 3
European Journal of Medical Genetics xxx (xxxx) xxxx
T. Robyns, et al.
Table 1 Predictors of carrying a mutation in an HCM related gene.
Age diagnosis < 45 years Male Gender Familial HCM Familial SD Syncope Arterial hypertension MWT ≥ 20 mm Asymmetrical HCM negative T lateral leads
Multivariatea
Mutation pos (N = 141)
Mutation neg (N = 237)
univariate
Number (%)
Number (%)
p-value
OR
CI
p-value
OR
CI
96 (68) 96 (68) 79 (56) 46 (33) 28 (20) 27 (19) 81 (57) 128 (91) 56 (40)
77 (32) 165 (70) 55 (23) 31 (13) 18 (8) 115 (49) 95 (40) 170 (72) 135 (57)
< 0,001 0,75 < 0,001 < 0,001 0,001 < 0,001 0,001 < 0,001 0,001
4,43 0,93 4,22 3,22 3,02 0,25 2,02 3,88 0,50
2,84–6,93 0,59–1,46 2,69–6,61 1,92–5,39 1,60–5,68 0,15–0,41 1,32–3,08 2,05–7,34 0,33–0,76
< 0,001
2,83
1,66–4,83
< 0,001 0,047 0,01 0,006 0,03 0,02 0,002
4,18 1,94 2,69 0,44 1,85 2,48 0,44
2,43–7,17 1,01–3,73 1,26–5,74 0,25–0,79 1,08–3,18 1,18–5,18 0,26–0,75
a There was no significant interaction between the different variables; Pos = positive; neg = negative; SD = Sudden death; MWT = maximal wall thickness; OR = odds ratio; CI = confidence interval.
different cut-off of the score to decide upon genetic testing in our population are listed in Table 2.
Table 2 Different cut offs of the genotype score and their respective sensitivities and yield.
4. Discussion In this study, we describe the genotype-phenotype relation in our cohort of HCM patients. Using the stringent ACMG-AMP criteria, we identified a mutation in 37% of patients with the majority of them in MYBPC3, MYH7 and the troponin complex. The achieved genetic yield is very similar to what was previously published in a large cohort of almost 3000 unselected index patients (Alfares et al., 2015). Patients with a mutation were younger at diagnosis and had more severe disease including more pronounced hypertrophy and more frequent syncope. However, the rate of SCD was very similar. Patients carrying only a VUS were clinically indistinguishable from those who did not carry a mutation, suggesting that these variants might be bystanders rather than real mutations. We did not observe a difference in the occurrence of the composite endpoint of sudden death between mutation positive and mutation negative patients. Mutation positive patients were diagnosed at a younger age with similar follow-up duration compared to mutation negative patients, while events occurred at the same age in both groups. Therefore, mutation positive patients might still be at higher risk compared to mutation negative patients during long term follow-up. MYBPC3 mutation carriers had a worse outcome compared to troponin complex mutations and a trend towards worse outcome compared to MYH7 mutation carriers and mutation negative patients. This contrasts with earlier reports claiming worse survival in MYH7 mutation carriers compared to MYBPC3 (Charron et al., 1998). This might be explained
cutoff
Sensitivity (%)
Yield (%)
−2 −1 0 1 2 3 4 5 6 7
100 (141/141) 100 (141/141) 100 (141/141) 95 (134/141) 87 (122/141) 66 (93/141) 43 (60/141) 24 (34/141) 7 (10/141) 0.7% (1/141)
37 (141/378) 38 (141/370) 41 (141/345) 49 (134/273) 57 (122/208) 69 (93/134) 79 (60/76) 83 (34/41) 77 (10/13) 100 (1/1)
Sensitivities here reported represent the chance of identifying the mutation in a HCM proband carrying a definite mutation according different cutoffs of the proposed genotype score. The yield represents the chance of identifying a mutation in a cohort of patients with a score more than or equal to the cut-off.
by a more severe phenotype in MYBPC3 mutation carriers in our cohort as illustrated by increased MWT compared to troponin mutation carriers and increased occurrence of nsVT compared to MYH7 mutation carriers. Furthermore, this analysis only included 1 patient per family, which might have excluded other family members reaching the composite endpoint of SCD. We developed a score system based on clinical variables to predict the genetic yield, similarly to what was done previously by the group of the Mayo clinic in their North-American cohort (Murphy et al., 2016). On top of the clinical variables, we also included ECG variables. Surprisingly, presence of negative T waves in the lateral leads was a Fig. 2. Score to predict outcome of genetic testing. This figure illustrates the proposed score correlates with the genetic yield in HCM patients. The more risk markers a patient carries, the more chance you have to find a mutation (Pearson r = 0.98). The number in the bars illustrates the absolute number of patients with a specific score in the denominator and the number of patients in whom a mutation was identified in the numerator.
4
European Journal of Medical Genetics xxx (xxxx) xxxx
T. Robyns, et al.
Appendix A. Supplementary data
negative predictor of carrying a mutation. It is well known that giant negative T waves in multiple ECG leads is suggestive for apical hypertrophy and that this pattern is caused by apico-basal gradient of hypertrophy (Usui et al., 1993). Furthermore, it was suggested that regional negative T waves correlate with regional scarring demonstrated by late gadolinium enhancement on cardiac MRI (Fronza et al., 2016). This would suggest that presence of scarring in the lateral wall is less present in genotype positive HCM. However, further evaluation of this finding is needed before any firm conclusions can be drawn. The advantages of a tool that predicts the genetic yield based on clinical variables are manifold. First, such a prediction tool aids to counsel the patient and family appropriately and to create realistic pretest expectations. Since only one out of 44 index patients with a score of −1 or 0 appeared to have a pathogenic mutation, these patients should not be referred for genetic testing since the yield does not exceed the background genetic noise in HCM related genes (estimated at 9% in our cohort). Second, it helps in identifying those patients where phenotypic screening of relatives is highly recommended. Recently a novel nonfamilial subtype of HCM has been delineated based on the combination of absence of a familial history of HCM, absence of a mutation and a favorable clinical outcome (Ingles et al., 2017). Prediction tools might help to differentiate this specific subtype from the more malignant classical subtype since a low risk score already points in the direction of the nonfamilial subtype. Of course it does not rule out genetic testing in these patients. Third, in health care systems where access to genetic testing or resources are limited, this tool could help in selecting patients who will most likely carry a mutation. Therefore, cost-effective strategies can be proposed.
Supplementary data to this article can be found online at https:// doi.org/10.1016/j.ejmg.2019.103754. References Alfares, A.A., Kelly, M.A., McDermott, G., et al., 2015. Results of clinical genetic testing of 2,912 probands with hypertrophic cardiomyopathy: expanded panels offer limited additional sensitivity. Genet. Med. 17 (11), 880–888. Burns, C., Bagnall, R.D., Lam, L., Semsarian, C., Ingles, J., 2017. Multiple gene variants in hypertrophic cardiomyopathy in the era of next-generation sequencing. Circ Cardiovasc Genet 10 (4). Charron, P., Dubourg, O., Desnos, M., et al., 1998. Clinical features and prognostic implications of familial hypertrophic cardiomyopathy related to the cardiac myosinbinding protein C gene. Circulation 97 (22), 2230–2236. Elliott, P.M., Anastasakis, A., Borger, M.A., et al., 2014. ESC guidelines on diagnosis and management of hypertrophic cardiomyopathy: the task force for the diagnosis and management of hypertrophic cardiomyopathy of the European society of cardiology (ESC). Eur. Heart J. 35 (39), 2733–2779 2014. Fronza, M., Raineri, C., Valentini, A., et al., 2016. Relationship between electrocardiographic findings and cardiac magnetic resonance phenotypes in patients with hypertrophic cardiomyopathy. Int J Cardiol Heart Vasc 11, 7–11. Geisterfer-Lowrance, A.A., Kass, S., Tanigawa, G., et al., 1990. A molecular basis for familial hypertrophic cardiomyopathy: a beta cardiac myosin heavy chain gene missense mutation. Cell 62 (5), 999–1006. Gruner, C., Ivanov, J., Care, M., et al., 2013. Toronto hypertrophic cardiomyopathy genotype score for prediction of a positive genotype in hypertrophic cardiomyopathy. Circ Cardiovasc Genet 6 (1), 19–26. Ingles, J., Sarina, T., Yeates, L., et al., 2013. Clinical predictors of genetic testing outcomes in hypertrophic cardiomyopathy. Genet. Med. 15 (12), 972–977. Ingles, J., Burns, C., Bagnall, R.D., et al., 2017. Nonfamilial hypertrophic cardiomyopathy: prevalence, natural history, and clinical implications. Circ Cardiovasc Genet 10 (2). Jarvik, G.P., Browning, B.L., 2016. Consideration of cosegregation in the pathogenicity classification of genomic variants. Am. J. Hum. Genet. 98 (6), 1077–1081. Levy, D., Labib, S.B., Anderson, K.M., Christiansen, J.C., Kannel, W.B., Castelli, W.P., 1990. Determinants of sensitivity and specificity of electrocardiographic criteria for left ventricular hypertrophy. Circulation 81 (3), 815–820. Lopes, L.R., Rahman, M.S., Elliott, P.M., 2013. A systematic review and meta-analysis of genotype-phenotype associations in patients with hypertrophic cardiomyopathy caused by sarcomeric protein mutations. Heart 99 (24), 1800–1811. Maron, B.J., Gardin, J.M., Flack, J.M., Gidding, S.S., Kurosaki, T.T., Bild, D.E., 1995. Prevalence of hypertrophic cardiomyopathy in a general population of young adults. Echocardiographic analysis of 4111 subjects in the CARDIA Study. Coronary Artery Risk Development in (Young) adults. Circulation 92 (4), 785–789. Maron, B.J., Maron, M.S., Semsarian, C., 2012. Genetics of hypertrophic cardiomyopathy after 20 years: clinical perspectives. J. Am. Coll. Cardiol. 60 (8), 705–715. Maron, B.J., Casey, S.A., Chan, R.H., Garberich, R.F., Rowin, E.J., Maron, M.S., 2015. Independent assessment of the European society of cardiology sudden death risk model for hypertrophic cardiomyopathy. Am. J. Cardiol. 116 (5), 757–764. Murphy, S.L., Anderson, J.H., Kapplinger, J.D., et al., 2016. Evaluation of the Mayo clinic phenotype-based genotype predictor score in patients with clinically diagnosed hypertrophic cardiomyopathy. J Cardiovasc Transl Res 9 (2), 153–161. Richards, S., Aziz, N., Bale, S., et al., 2015. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of medical genetics and genomics and the association for molecular pathology. Genet. Med. 17 (5), 405–424. Robyns, T., Lu, H.R., Gallacher, D.J., et al., 2016. Evaluation of index of cardio-electrophysiological balance (iCEB) as a new biomarker for the identification of patients at increased arrhythmic risk. Ann. Noninvasive Electrocardiol. 21 (3), 294–304. Semsarian, C., Ingles, J., Maron, M.S., Maron, B.J., 2015. New perspectives on the prevalence of hypertrophic cardiomyopathy. J. Am. Coll. Cardiol. 65 (12), 1249–1254. Usui, M., Inoue, H., Suzuki, J., Watanabe, F., Sugimoto, T., Nishikawa, J., 1993. Relationship between distribution of hypertrophy and electrocardiographic changes in hypertrophic cardiomyopathy. Am. Heart J. 126 (1), 177–183. Vriesendorp, P.A., Schinkel, A.F., Liebregts, M., et al., 2015. Validation of the 2014 European Society of Cardiology guidelines risk prediction model for the primary prevention of sudden cardiac death in hypertrophic cardiomyopathy. Circ Arrhythm Electrophysiol 8 (4), 829–835.
5. Limitations Not all patients were tested with the extended gene panel. Some patients carrying a mutation in the 3 main sarcomere genes might be carrier of a second pathogenic variant. However, it was recently shown that carrying multiple (likely) pathogenic variants is very rare (Burns et al., 2017). On the other hand, 31% (57/181) of the mutation negative patients were only tested for the 3 main sarcomere genes. It can be estimated from our data that 7 out of these 57 patients (12%) carry a (likely) pathogenic variant in the additional genes that were tested with the extended panel. This might have affected the results. In a second step, the proposed model should be validated in an independent cohort. 6. Conclusions The genetic yield in this Belgian HCM cohort was 37% with the majority of mutations located in MYBPC3. We propose a tool that predicts the genetic yield based on clinical variables that could help in adequate counseling of an individual patient. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. The authors report no relationships that could be construed as a conflict of interest.
5
Contents lists available at ScienceDirect
European Journal of Medical Genetics journal homepage: www.elsevier.com/locate/ejmg
Clinical and ECG variables to predict the outcome of genetic testing in hypertrophic cardiomyopathy Tomas Robynsa,b,c,∗, Jeroen Breckpotd, Dieter Nuyense, Bert Vandenberka,b, Anniek Corveleynd, Cuno Kuiperid, Lucas Van Aelsta,b, Johan Van Cleemputa,b, Rik Willemsa,b a
Department of Cardiovascular Diseases, University Hospitals Leuven, Belgium Department of Cardiovascular Sciences, University of Leuven, Belgium c The University Hospitals of Leuven are Member of the European Reference Network for Rare and Complex Diseases of the Heart (ERN GUARD-HEART), European Union d Center for Human Genetics, University Hospitals Leuven, Belgium e Department of Cardiology, Ziekenhuis Oost Limburg, Genk, Belgium b
A R T I C LE I N FO
A B S T R A C T
Keywords: Hypertrophic cardiomyopathy Genotype phenotype relation Genetic testing
Knowledge on the influence of specific genotypes on the phenotypic expression of hypertrophic cardiomyopathy (HCM) is emerging. The objective of this study was to evaluate the genotype-phenotype relation in HCM patients and to construct a score to predict the genetic yield based to improve counseling. Unrelated HCM patients who underwent genetic testing were included in the analysis. Multivariate logistic regression was performed to identify variables that predict a positive genetic test. A weighted score was constructed based on the odds ratios. In total, 378 HCM patients were included of whom 141 carried a mutation (global yield 37%), 181 were mutation negative and 56 only carried a variant of unknown significance. We identified age at diagnosis < 45 years, familial HCM, familial sudden death, arrhythmic syncope, maximal wall thickness ≥20 mm, asymmetrical hypertrophy and the absence of negative T waves in the lateral ECG leads as significant predictors of a positive genetic test. When we included these values in a risk score we found very high correlation between the score and the observed genetic yield (Pearson r = 0.98). MYBPC3 mutation carriers more frequently suffered sudden cardiac death compared to troponin complex mutations carriers (p = 0.01) and a similar trend was observed compared to MYH7 mutation carriers (p = 0.08) and mutation negative patients (p = 0.11). To conclude, a simple score system based on clinical variables can predict the genetic yield in HCM index patients, aiding in counseling HCM patients. MYBPC3 mutation carriers had a worse outcome regarding sudden cardiac death.
1. Introduction Hypertrophic cardiomyopathy (HCM) is the most common hereditary heart disease with an estimated prevalence of 1 in 500 to 1 in 200 (Maron et al., 1995; Semsarian et al., 2015). The first gene implicated in the disease was MYH7, identified in 1990 (Geisterfer-Lowrance et al., 1990). Since then multiple other genes have been labeled as causal for the disease (Maron et al., 2012). Albeit this tremendous progress, the yield of genetic testing remains incomplete with an estimate of 40% of patients in whom a mutation is identified (Lopes et al., 2013). Furthermore, interpretation of genetic variants remains difficult, especially in the less commonly mutated HCM genes. Genotype-phenotype relation for HCM has been studied in several cohorts, and were compiled in a meta-analysis (Lopes et al., 2013). Subsequently, two independent North American groups and an Australian group constructed a scoring
∗
system based on clinical variables to predict the chance of identifying a mutation in HCM patients (Murphy et al., 2016; Gruner et al., 2013; Ingles et al., 2013). However, not a single electrocardiographic parameter was included in these analyses. Furthermore, these tools have not been developed for European HCM patients. Yet, the factors that infer a higher risk of carrying a causal mutation might differ as much as the predictors of adverse outcome in HCM patients between European and American cohorts (Maron et al., 2015; Vriesendorp et al., 2015). Correct counseling of these patients regarding chances of identifying a causal mutation is essential to prevent unrealistic ideas on genetic testing and scoring systems therefore aid in this process. Therefore, we aimed to (1) explore genotype-phenotype relation in HCM patients in our Belgian cohort, (2) identify predictors of a positive genetic test and (3) construct a scoring system to predict genetic yield in HCM patients.
Corresponding author. University hospitals Leuven, Department of Cardiovascular diseases, Herestraat 49, 3000, Leuven, Belgium. E-mail address: [email protected] (T. Robyns).
https://doi.org/10.1016/j.ejmg.2019.103754 Received 8 November 2018; Received in revised form 23 August 2019; Accepted 8 September 2019 1769-7212/ © 2019 Elsevier Masson SAS. All rights reserved.
Please cite this article as: Tomas Robyns, et al., European Journal of Medical Genetics, https://doi.org/10.1016/j.ejmg.2019.103754
European Journal of Medical Genetics xxx (xxxx) xxxx
T. Robyns, et al.
2. Methods
2.4. ECG variables
2.1. Patient selection
Twelve lead ECG's were recorded using MAC 5500 (Marquette, GE Healthcare, Chicago, Illinois, USA). Automated analysis was done using the widely used ‘GE Marquette 12SLTM ECG Analysis Program’. As such, automated measurements of PR interval, QRS duration, RR interval and QT interval corrected for heart rate with Fridericia's (QTcF) formula were obtained. Index of cardio-electrophysiological balance (iCEB) was defined as QT interval divided by QRS duration (Robyns et al., 2016). All ECG leads were assessed for negative T waves > 1 mm. Anterior leads were defined as leads V1–V4, lateral leads as leads V5,V6, I, aVL and inferior leads as leads II, III and aVF. Presence of negative T waves in a specific area of the heart was defined as ≥ 2 leads in either anterior, lateral or inferior leads with negative T waves. Finally, left ventricular hypertrophy on the ECG was assessed by the sum of the maximum S wave amplitude in leads V1 or V2 and the maximum R wave amplitude in leads V5 or V6 (part of Framingham LVH score) (Levy et al., 1990).
The database of the center for hereditary heart disease of the university hospitals of Leuven was retrospectively analyzed to select all unrelated patients diagnosed with HCM, from the beginning of the database in 2002 until August 2016. HCM was defined according the 2014 ESC guidelines as unexplained left ventricular hypertrophy of ≥15 mm (Elliott et al., 2014). We selected only index patients for whom both ECG and echocardiography were digitally available. Our study complied with the Declaration of Helsinki, and the research protocol was approved by the local ethics committee.
2.2. Genotyping and variant interpretation Patients were only included if at least the 3 main genes (MYBPC3, MYH7 and TNNT2) that were selected at our center in the early days of HCM genotyping were evaluated. Genotyping was done (1) by Denaturing High Pressure Liquid Chromatography (DHPLC) of the 3 main genes followed by direct Sanger sequencing of abnormal results until 2009, (2) by direct Sanger sequencing of the 3 main genes between 2009 and 2013 and (3) by targeted gene panel testing from then on. In total, 217 patients were evaluated with the extended panel. The panel contained following genes: ACTC1, ACTN2, ANKRD1, CALR3, CASQ2, CAV3, CRYAB*, CSRP3, DES, FHL1*, FLNC*, GLA, JPH2, LAMP2, LDB3, MYBPC3, MYH6, MYH7, MYL2, MYL3, MYLK2, MYOZ2, MYPN*, NEXN*, PLN, PRKAG2, RYR2, TCAP, TNNC1, TNNI3, TNNT2, TPM1, TTR*, VCL. A subset of patients (N = 107) were evaluated using a panel that included 6 additional genes indicated by an asterisk (*). Moreover multiplex Ligation-dependent Probe Amplification (MLPA) was performed for MYBPC3 in all patients. All genetic variants were reassessed by the criteria proposed by the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG-AMP criteria) as pathogenic variant, likely pathogenic variant, variant of unknown significance (VUS), likely benign variant or benign variant (Richards et al., 2015). To evaluate the strength of co-segregation, we used the criteria proposed by Jarvik and Browning (2016). A mutation was defined as a pathogenic or likely pathogenic variant according these criteria.
2.5. Statistical analysis First we compared clinical, echocardiography and ECG parameters between mutation positive, mutation negative and VUS positive HCM patients using one way ANOVA with Tukey's post hoc testing for continuous variables and using chi square test for categorical variables. The genotype-phenotype relation was evaluated by comparing MYBPC3 mutation carriers, MYH7 mutation carriers and carriers of a mutation in the troponin complex (TNNC1, TNNI3 and TNNT2) using one way ANOVA with Tukey's post hoc testing for continuous variables and using chi square test for categorical variables. The number of mutation carriers in other genes was too small to include in this analysis. KaplanMeier curves with log rank test were constructed to evaluate survival regarding the composite endpoint with log rank test. Evaluation of predictors of carrying a mutation was done using binary logistic regression (enter method, variables with p-value < 0.1 were entered into the multivariate model). A score system was developed based upon the odds ratios (OR) of significant variables in the multivariate model: variables with OR > 3 were awarded 2 points, OR 1–3 were awarded 1 point and OR < 1 were awarded ‘-1’ point. A Pearson correlation coefficient was calculated to document the correlation between the risk score and the observed genetic yield. 3. Results
2.3. Definition clinical variables 3.1. Genetic background patient population Familial sudden death was defined as a sudden death before age 40 in a first or second degree relative, or as sudden death at any age in a family member with an established diagnosis of HCM. Familial HCM was defined as a family member with definite diagnosis of HCM or with an unexplained maximal wall thickness (MWT) ≥ 12 mm based upon a detailed family history and review of available clinical data (Elliott et al., 2014). MWT was either measured with transthoracic echocardiography (TTE) or cardiac MRI. Three different morphology types of hypertrophy were defined: asymmetrical, symmetrical and apical. Left ventricular outflow tract obstruction (LVOTO) was measured either in rest or after a Valsalva maneuver. Obstructive HCM was defined as LVOTO > 30 mmHg. Systolic anterior motion (SAM) of the mitral valve was also assessed from TTE. Arrhythmic syncope was defined as unexplained syncope with sudden onset. Heart failure episodes were episodes of clinical congestion necessitating diuretic therapy. The presence of arterial hypertension was extracted from the medical records. If 24 h holter recordings were available, non-sustained ventricular tachycardia (nsVT) was defined as 3 consecutive ventricular premature beats > 120 BPM. We defined sudden cardiac death as a composite endpoint of out of hospital cardiac arrest, appropriate ICD shock or sustained ventricular tachycardia.
Out of 381 identified HCM index patients in whom ECG and echocardiography was available at our institution, 378 (99%) underwent genetic testing for at least MYBPC3, MYH7 and TNNT2. A definite pathogenic mutation was identified in 141 patients corresponding to a yield of 37% in the total population. Fig. 1 illustrates the yield per gene indexed for the number of patients in whom the gene was sequenced. The vast majority of mutations were identified in MYBPC3 (N = 80; 57%), MYH7 (N = 27; 19%) and troponin complex subunits with TNNC1 (N = 11; 8%), TNNI3 (N = 9; 6%) and TNNT2 (N = 4; 3%). A minority of patients carried a mutation in other sarcomere proteins: MYL2 (N = 2; 1.4%), MYL3 (N = 1; 0.7%) and TPM1 (N = 1; 0.7%). Finally 6 index patients (4%) were carrier of a mutation in PRKAG2. Details of the mutations can be found in Supplementary Table 1. Some mutations were found in ≥5 patients, including 4 mutations in MYBPC3 (c.1404del (N = 20), c.676_701dup (N = 7), c.2373dup (N = 5) and a deletion of exon 23 to 26 (N = 5)), 1 mutation in TNNC1 (c.430A > G; N = 11) and 1 mutation in TNNI3 (c.433C > T (N = 6)). Sixty-nine variants of unknown significance were identified in 63 probands. Out of these 69 VUS, 7 were identified in a patient in combination with a mutation, 4 were carrier of 2 VUS and 1 was carrier 2
European Journal of Medical Genetics xxx (xxxx) xxxx
T. Robyns, et al.
Fig. 1. Genetic yield per gene. This figure illustrates the percentage of mutations (dark bars) per gene indexed for the number of patients in whom the gene was genotyped. The bright bars illustrate the percentage of patients carrying a variant of unknown significance (VUS) in a specific gene indexed for the number of patients in whom the gene was genotyped.
carriers during a mean follow-up of 101 months (Supplementary Table 4). Survival analysis showed a statistically significant difference in outcome between mutation negative patients, MYBPC3 mutation carriers, MYH7 mutation carriers and troponin complex mutation carriers (p = 0.03; supplementary figure). This was based on worse outcome for MYBPC3 mutation carriers compared to troponin complex mutation carriers (p = 0.01), while there was only a trend in survival difference between MYBPC3 mutation carriers and MYH7 mutation carriers (p = 0.08) and mutation negative patients (p = 0.11). There was no difference in survival between mutation negative patients and MYH7 mutation carriers (p = 0.31) or troponin complex mutation carriers (p = 0.10). LVH was more pronounced in MYBPC3 mutation carriers compared to troponin complex mutation carriers, while the documentation of nsVT occurred at the same frequency (Supplementary Table 4). However, in MYH7 mutation carriers, nsVT was less frequently observed compared to MYBPC3 mutation carriers. ECG parameters were similar in the 3 subgroups, apart from longer QRS duration in MYH7 mutation carriers compared to MYBPC3 mutation carriers. Negative T waves in the lateral leads were less frequently observed in MYBPC3 mutation carriers compared to troponin mutation carriers.
of a mutation and 2 VUS. Details of the VUS can be found in Supplementary Table 2. Of the 181 mutation negative HCM index patients, 57 (31%) were only tested for MYBPC3, MYH7 and TNNT2, and 124 (69%) underwent HCM panel testing. All genetic variants mentioned in this paper were submitted to ClinVar database and accession numbers for (likely) pathogenic and VUS are found in Supplementary Tables 1 and 2 respectively. 3.2. Mutation negative versus mutation positive HCM Comparison of clinical, echocardiographic and electrocardiographic variables between mutation positive (N = 141), mutation negative (N = 181) and VUS positive (N = 56) HCM index patients is summarized in Supplementary Table 3. In short, mutation positive HCM index patients were significantly younger, had more frequent occurrence of familial SD, familial HCM, arterial hypertension and syncope compared to mutation negative HCM patients and HCM patients carrying only a VUS. Wall thickness was more pronounced and morphology was more frequently asymmetrical in mutation positive HCM compared to mutation negative HCM. There was no difference in ECG variables between the 3 groups, apart from less prevalence of negative T waves in the lateral leads in mutation positive HCM patients. None of the variables differed between mutation negative HCM and HCM patients who carried only a VUS. No difference in survival regarding the composite endpoint of sudden death was observed between mutation positive patients and the combined group of mutation negative patients and VUS positive patients (Log rank test p = 0.38) or between the 3 groups separately (Log rank test p = 0.10). SCD occurred at similar ages between those patients in whom a mutation was identified (46 ± 19 years) and those who remained genetically elusive (49 ± 22 years; p = 0.68).
3.4. Predictors of a positive genotype We evaluated predictors of identifying a mutation in HCM patients. For this analysis, we merged mutation negative HCM patients and HCM patients who carried only a VUS as mutation negative. In multivariate analysis, several parameters predicted identification of a mutation (Table 1). Based on the odds ratios, 2 points were allocated to familial history of HCM, 1 point to familial SD, age at diagnosis < 45 years, MWT ≥20 mm, asymmetrical hypertrophy and syncope, while 1 point was subtracted for the presence of negative T waves in the lateral leads and arterial hypertension. This resulted in a high correlation between the genetic yield and the risk score (Pearson r = 0.98; p < 0.001) which can also be appreciated from the nice spread of genetic yields across the spectrum of combinations of clinical risk factors from 0% if only negative T waves in the lateral leads or arterial hypertension were present to 100% if all positive risk markers were present (Fig. 2). The sensitivity of genetic testing and the genetic yield depending on
3.3. Genotype-phenotype relation in MYBPC3, MYH7 and troponin complex mutation carriers SCD occurred in 13 out of 80 MYBPC3 mutation carriers during a mean follow-up of 74 months (annual risk 2.6%), while it only occurred in 1 out of 27 MYH7 mutation carriers during a mean follow-up of 83 months (annual risk 0.5%) and in none of the troponin mutation 3
European Journal of Medical Genetics xxx (xxxx) xxxx
T. Robyns, et al.
Table 1 Predictors of carrying a mutation in an HCM related gene.
Age diagnosis < 45 years Male Gender Familial HCM Familial SD Syncope Arterial hypertension MWT ≥ 20 mm Asymmetrical HCM negative T lateral leads
Multivariatea
Mutation pos (N = 141)
Mutation neg (N = 237)
univariate
Number (%)
Number (%)
p-value
OR
CI
p-value
OR
CI
96 (68) 96 (68) 79 (56) 46 (33) 28 (20) 27 (19) 81 (57) 128 (91) 56 (40)
77 (32) 165 (70) 55 (23) 31 (13) 18 (8) 115 (49) 95 (40) 170 (72) 135 (57)
< 0,001 0,75 < 0,001 < 0,001 0,001 < 0,001 0,001 < 0,001 0,001
4,43 0,93 4,22 3,22 3,02 0,25 2,02 3,88 0,50
2,84–6,93 0,59–1,46 2,69–6,61 1,92–5,39 1,60–5,68 0,15–0,41 1,32–3,08 2,05–7,34 0,33–0,76
< 0,001
2,83
1,66–4,83
< 0,001 0,047 0,01 0,006 0,03 0,02 0,002
4,18 1,94 2,69 0,44 1,85 2,48 0,44
2,43–7,17 1,01–3,73 1,26–5,74 0,25–0,79 1,08–3,18 1,18–5,18 0,26–0,75
a There was no significant interaction between the different variables; Pos = positive; neg = negative; SD = Sudden death; MWT = maximal wall thickness; OR = odds ratio; CI = confidence interval.
different cut-off of the score to decide upon genetic testing in our population are listed in Table 2.
Table 2 Different cut offs of the genotype score and their respective sensitivities and yield.
4. Discussion In this study, we describe the genotype-phenotype relation in our cohort of HCM patients. Using the stringent ACMG-AMP criteria, we identified a mutation in 37% of patients with the majority of them in MYBPC3, MYH7 and the troponin complex. The achieved genetic yield is very similar to what was previously published in a large cohort of almost 3000 unselected index patients (Alfares et al., 2015). Patients with a mutation were younger at diagnosis and had more severe disease including more pronounced hypertrophy and more frequent syncope. However, the rate of SCD was very similar. Patients carrying only a VUS were clinically indistinguishable from those who did not carry a mutation, suggesting that these variants might be bystanders rather than real mutations. We did not observe a difference in the occurrence of the composite endpoint of sudden death between mutation positive and mutation negative patients. Mutation positive patients were diagnosed at a younger age with similar follow-up duration compared to mutation negative patients, while events occurred at the same age in both groups. Therefore, mutation positive patients might still be at higher risk compared to mutation negative patients during long term follow-up. MYBPC3 mutation carriers had a worse outcome compared to troponin complex mutations and a trend towards worse outcome compared to MYH7 mutation carriers and mutation negative patients. This contrasts with earlier reports claiming worse survival in MYH7 mutation carriers compared to MYBPC3 (Charron et al., 1998). This might be explained
cutoff
Sensitivity (%)
Yield (%)
−2 −1 0 1 2 3 4 5 6 7
100 (141/141) 100 (141/141) 100 (141/141) 95 (134/141) 87 (122/141) 66 (93/141) 43 (60/141) 24 (34/141) 7 (10/141) 0.7% (1/141)
37 (141/378) 38 (141/370) 41 (141/345) 49 (134/273) 57 (122/208) 69 (93/134) 79 (60/76) 83 (34/41) 77 (10/13) 100 (1/1)
Sensitivities here reported represent the chance of identifying the mutation in a HCM proband carrying a definite mutation according different cutoffs of the proposed genotype score. The yield represents the chance of identifying a mutation in a cohort of patients with a score more than or equal to the cut-off.
by a more severe phenotype in MYBPC3 mutation carriers in our cohort as illustrated by increased MWT compared to troponin mutation carriers and increased occurrence of nsVT compared to MYH7 mutation carriers. Furthermore, this analysis only included 1 patient per family, which might have excluded other family members reaching the composite endpoint of SCD. We developed a score system based on clinical variables to predict the genetic yield, similarly to what was done previously by the group of the Mayo clinic in their North-American cohort (Murphy et al., 2016). On top of the clinical variables, we also included ECG variables. Surprisingly, presence of negative T waves in the lateral leads was a Fig. 2. Score to predict outcome of genetic testing. This figure illustrates the proposed score correlates with the genetic yield in HCM patients. The more risk markers a patient carries, the more chance you have to find a mutation (Pearson r = 0.98). The number in the bars illustrates the absolute number of patients with a specific score in the denominator and the number of patients in whom a mutation was identified in the numerator.
4
European Journal of Medical Genetics xxx (xxxx) xxxx
T. Robyns, et al.
Appendix A. Supplementary data
negative predictor of carrying a mutation. It is well known that giant negative T waves in multiple ECG leads is suggestive for apical hypertrophy and that this pattern is caused by apico-basal gradient of hypertrophy (Usui et al., 1993). Furthermore, it was suggested that regional negative T waves correlate with regional scarring demonstrated by late gadolinium enhancement on cardiac MRI (Fronza et al., 2016). This would suggest that presence of scarring in the lateral wall is less present in genotype positive HCM. However, further evaluation of this finding is needed before any firm conclusions can be drawn. The advantages of a tool that predicts the genetic yield based on clinical variables are manifold. First, such a prediction tool aids to counsel the patient and family appropriately and to create realistic pretest expectations. Since only one out of 44 index patients with a score of −1 or 0 appeared to have a pathogenic mutation, these patients should not be referred for genetic testing since the yield does not exceed the background genetic noise in HCM related genes (estimated at 9% in our cohort). Second, it helps in identifying those patients where phenotypic screening of relatives is highly recommended. Recently a novel nonfamilial subtype of HCM has been delineated based on the combination of absence of a familial history of HCM, absence of a mutation and a favorable clinical outcome (Ingles et al., 2017). Prediction tools might help to differentiate this specific subtype from the more malignant classical subtype since a low risk score already points in the direction of the nonfamilial subtype. Of course it does not rule out genetic testing in these patients. Third, in health care systems where access to genetic testing or resources are limited, this tool could help in selecting patients who will most likely carry a mutation. Therefore, cost-effective strategies can be proposed.
Supplementary data to this article can be found online at https:// doi.org/10.1016/j.ejmg.2019.103754. References Alfares, A.A., Kelly, M.A., McDermott, G., et al., 2015. Results of clinical genetic testing of 2,912 probands with hypertrophic cardiomyopathy: expanded panels offer limited additional sensitivity. Genet. Med. 17 (11), 880–888. Burns, C., Bagnall, R.D., Lam, L., Semsarian, C., Ingles, J., 2017. Multiple gene variants in hypertrophic cardiomyopathy in the era of next-generation sequencing. Circ Cardiovasc Genet 10 (4). Charron, P., Dubourg, O., Desnos, M., et al., 1998. Clinical features and prognostic implications of familial hypertrophic cardiomyopathy related to the cardiac myosinbinding protein C gene. Circulation 97 (22), 2230–2236. Elliott, P.M., Anastasakis, A., Borger, M.A., et al., 2014. ESC guidelines on diagnosis and management of hypertrophic cardiomyopathy: the task force for the diagnosis and management of hypertrophic cardiomyopathy of the European society of cardiology (ESC). Eur. Heart J. 35 (39), 2733–2779 2014. Fronza, M., Raineri, C., Valentini, A., et al., 2016. Relationship between electrocardiographic findings and cardiac magnetic resonance phenotypes in patients with hypertrophic cardiomyopathy. Int J Cardiol Heart Vasc 11, 7–11. Geisterfer-Lowrance, A.A., Kass, S., Tanigawa, G., et al., 1990. A molecular basis for familial hypertrophic cardiomyopathy: a beta cardiac myosin heavy chain gene missense mutation. Cell 62 (5), 999–1006. Gruner, C., Ivanov, J., Care, M., et al., 2013. Toronto hypertrophic cardiomyopathy genotype score for prediction of a positive genotype in hypertrophic cardiomyopathy. Circ Cardiovasc Genet 6 (1), 19–26. Ingles, J., Sarina, T., Yeates, L., et al., 2013. Clinical predictors of genetic testing outcomes in hypertrophic cardiomyopathy. Genet. Med. 15 (12), 972–977. Ingles, J., Burns, C., Bagnall, R.D., et al., 2017. Nonfamilial hypertrophic cardiomyopathy: prevalence, natural history, and clinical implications. Circ Cardiovasc Genet 10 (2). Jarvik, G.P., Browning, B.L., 2016. Consideration of cosegregation in the pathogenicity classification of genomic variants. Am. J. Hum. Genet. 98 (6), 1077–1081. Levy, D., Labib, S.B., Anderson, K.M., Christiansen, J.C., Kannel, W.B., Castelli, W.P., 1990. Determinants of sensitivity and specificity of electrocardiographic criteria for left ventricular hypertrophy. Circulation 81 (3), 815–820. Lopes, L.R., Rahman, M.S., Elliott, P.M., 2013. A systematic review and meta-analysis of genotype-phenotype associations in patients with hypertrophic cardiomyopathy caused by sarcomeric protein mutations. Heart 99 (24), 1800–1811. Maron, B.J., Gardin, J.M., Flack, J.M., Gidding, S.S., Kurosaki, T.T., Bild, D.E., 1995. Prevalence of hypertrophic cardiomyopathy in a general population of young adults. Echocardiographic analysis of 4111 subjects in the CARDIA Study. Coronary Artery Risk Development in (Young) adults. Circulation 92 (4), 785–789. Maron, B.J., Maron, M.S., Semsarian, C., 2012. Genetics of hypertrophic cardiomyopathy after 20 years: clinical perspectives. J. Am. Coll. Cardiol. 60 (8), 705–715. Maron, B.J., Casey, S.A., Chan, R.H., Garberich, R.F., Rowin, E.J., Maron, M.S., 2015. Independent assessment of the European society of cardiology sudden death risk model for hypertrophic cardiomyopathy. Am. J. Cardiol. 116 (5), 757–764. Murphy, S.L., Anderson, J.H., Kapplinger, J.D., et al., 2016. Evaluation of the Mayo clinic phenotype-based genotype predictor score in patients with clinically diagnosed hypertrophic cardiomyopathy. J Cardiovasc Transl Res 9 (2), 153–161. Richards, S., Aziz, N., Bale, S., et al., 2015. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of medical genetics and genomics and the association for molecular pathology. Genet. Med. 17 (5), 405–424. Robyns, T., Lu, H.R., Gallacher, D.J., et al., 2016. Evaluation of index of cardio-electrophysiological balance (iCEB) as a new biomarker for the identification of patients at increased arrhythmic risk. Ann. Noninvasive Electrocardiol. 21 (3), 294–304. Semsarian, C., Ingles, J., Maron, M.S., Maron, B.J., 2015. New perspectives on the prevalence of hypertrophic cardiomyopathy. J. Am. Coll. Cardiol. 65 (12), 1249–1254. Usui, M., Inoue, H., Suzuki, J., Watanabe, F., Sugimoto, T., Nishikawa, J., 1993. Relationship between distribution of hypertrophy and electrocardiographic changes in hypertrophic cardiomyopathy. Am. Heart J. 126 (1), 177–183. Vriesendorp, P.A., Schinkel, A.F., Liebregts, M., et al., 2015. Validation of the 2014 European Society of Cardiology guidelines risk prediction model for the primary prevention of sudden cardiac death in hypertrophic cardiomyopathy. Circ Arrhythm Electrophysiol 8 (4), 829–835.
5. Limitations Not all patients were tested with the extended gene panel. Some patients carrying a mutation in the 3 main sarcomere genes might be carrier of a second pathogenic variant. However, it was recently shown that carrying multiple (likely) pathogenic variants is very rare (Burns et al., 2017). On the other hand, 31% (57/181) of the mutation negative patients were only tested for the 3 main sarcomere genes. It can be estimated from our data that 7 out of these 57 patients (12%) carry a (likely) pathogenic variant in the additional genes that were tested with the extended panel. This might have affected the results. In a second step, the proposed model should be validated in an independent cohort. 6. Conclusions The genetic yield in this Belgian HCM cohort was 37% with the majority of mutations located in MYBPC3. We propose a tool that predicts the genetic yield based on clinical variables that could help in adequate counseling of an individual patient. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. The authors report no relationships that could be construed as a conflict of interest.
5