Author + information
- Peter J. Schwartz, MD∗ ( )( and )
- Maria-Christina Kotta, MSc, PhD
- IRCCS Istituto Auxologico Italiano, Center for Cardiac Arrhythmias of Genetic Origin and Laboratory of Cardiovascular Genetics, Milan, Italy
- ↵∗Address for correspondence:
Dr. Peter J. Schwartz, IRCCS Istituto Auxologico Italiano, Center for Cardiac Arrhythmias of Genetic Origin, Via Pier Lombardo 22, 20135 Milan, Italy.
The relationship between channelopathies and sudden infant death syndrome (SIDS) has always been complex and, for reasons that are interesting but would be out of place here, is a controversial one. Although one may currently refer to channelopathies in general, most of the initial work and discussion has centered on the role in SIDS of the long QT syndrome (LQTS). Therefore, it is appropriate to briefly review how the story unfolded.
The hypothesis that “some” cases of SIDS might have been due to LQTS was first presented in an RO1 grant application funded by the National Institutes of Health in 1974 and was more formally published in February 1976 (1). The same year, a small study on the QT interval in parents of SIDS victims supported the concept (2). To obtain solid data, it was necessary to perform a large electrocardiographic prospective study in newborns to compare the QT interval of SIDS cases, when they were still healthy, with that of the babies who developed normally. We enrolled >33,000 infants and demonstrated that a QTc prolongation in the first week of life increased the risk of SIDS by >40-fold (3). Two years later, the turning point: a 2-month-old healthy infant had a cardiac arrest at home due to a documented ventricular fibrillation and, after defibrillation, manifested a constant and major QT prolongation (QTc: 650 ms). Our genetic testing revealed a de novo SCN5A-p.S941N mutation that increased the late inward sodium current and that was absent in both parents (paternity confirmed) (4). If the child had died, which would have happened without prompt cardioversion, the normal QT interval of the parents would have prompted the diagnosis of SIDS. This case provided the “proof-of-concept” for the relationship between LQTS and SIDS, and was then followed by others (5).
What remained to be determined was the prevalence of LQTS-causing genetic mutations among SIDS cases. This question has prompted a growing number of genetic studies, beginning with the one by Tester and Ackerman (6), who found a prevalence of 5.2% of LQTS-causing mutations among 58 white SIDS cases (ethnic origin is important in these studies because the prevalence of LQTS among blacks is very low). Subsequently, a much larger study in 201 Norwegian SIDS victims provided the first data-driven estimate of prevalence (7). By considering only rare genetic variants with an in vitro functional effect, 15 variants present in 19 SIDS cases were regarded as likely contributors to sudden death (9.5%; 95% confidence interval [CI]: 5.8% to 14.4%). Among the latter, 8 variants were deemed as probable disease-causing mutations (4%) according to the pathogenicity criteria in effect at the time. These scientific efforts of >10 years ago (4–7) provided the genetic evidence that firmly established the link between LQTS and SIDS. Much has changed since then, and yet, many things still remain the same.
In this issue of the Journal, Tester et al. (8), using whole-exome sequencing analysis of 90 genetic heart disease (GHD)−associated genes in the largest SIDS cohort published to date (n = 419), explored the relative role of GHD in SIDS pathogenesis. This painstaking study has 3 strong aspects: it assembled a uniquely large cohort to enable significant conclusions; it interrogated most GHD-associated genes known to date; and it occurred at a time when indispensable tools for interpreting genetic findings are largely available. The investigators, by using a conservative minor allele frequency (MAF) cutoff of <0.005%, identified ultra-rare genetic variants in 46.3% of SIDS cases, of which 12.6% hosted at least 1 variant considered to be potentially informative. This yield did not correlate with sex, sleep position, or co-sleeping, but it was correlated with age, by increasing significantly after the age of 4 months. Upon stringent variant classification of all ultra-rare variants identified according to the American College of Medical Genetics and Genomics guidelines (9), the investigators arrived at 4.3% of SIDS cases either carrying pathogenic or likely pathogenic genetic variants.
Several important messages and considerations have emerged. In the era of whole genome population databases, stringent classification guidelines, and increased awareness over the questionable pathogenicity of variants previously associated to GHDs, it seems that at least 4.3% (95% CI: 2.6% to 6.7%) of SIDS may be relatively safely attributed to GHDs. This is because although we cannot exclude the possibility that some of the likely pathogenic variants reported may prove not to be pathogenic as knowledge evolves, we also expect at least a few of the variants of unknown significance (VUS) among the remaining 8.3% of cases with potentially informative variants to be upgraded to a likely pathogenic status. This particularly applies to the 29 novel VUSs found in 7% of SIDS cases that have mostly no evidence against pathogenicity, but just not enough in its favor. Therefore, 4.3% is probably as low as it gets. Interestingly, by applying the criteria of Tester et al. to the genetic findings of our SIDS study (7), and analyzing only the LQTS genes in common, we arrived at an almost identical yield of potentially informative variants (3.5% vs. 3.1% in this study; p = 0.81).
Although the conservative MAF cutoff corresponding to 1:10,000 individuals used in this study may provide the safety net for identifying true causative and highly penetrant variants, considering the SIDS incidence (1:5,000 live births) and the prevalence of LQTS (1:2,000) or hypertrophic cardiomyopathy (1:500), it can be inferred that this cutoff leaves a large genetic gap unexplored. Although it is true that a variant with a MAF falling just below the prevalence of these syndromes is not expected to be causative for SIDS, what about a variant with a MAF of 0.006%? We do not imply that the investigators' stringency is not justified. When we studied idiopathic ventricular fibrillation, we chose the same MAF cutoff (10); however, we also interrogated all variants within a higher MAF range in the 21 major GHD genes. Since MAF cutoffs are relatively arbitrary, an inherent limitation of the current study is that it did not firmly demonstrate the absence of likely pathogenic variants, nor did it explore the existence of possibly clinically relevant variants in a higher MAF range.
One specific example supporting the concept that MAF might not be the only clinically relevant aspect for such a study was provided by the investigators themselves. A described LQTS-associated variant (KCNH2-p.R176W) was filtered out due to the MAF criterion since its frequency exceeds the cutoff. This variant was traditionally considered to be a founder LQT2 mutation in the Finnish population (11) and was demonstrated to have a functional effect in a patient-derived induced pluripotent stem cell assay (12) but not in a zebrafish assay (13). This example highlights the merit of exploring higher frequency variants, because this information is clinically relevant and, furthermore, aids to upgrade or downgrade a variant in the classification scale. Remarkably, this variant seems to have been left floating from a “risk factor,” “pathogenic,” “VUS,” to a “likely benign,” classification status even among experienced professionals (8,14), thus underlining the persistent challenges in the interpretation of variants and the large margin of interpretational error in best-practice classification implementation.
Although this study focused on potential monogenic causes of SIDS, thus justifying stringent analyses, nevertheless, it did explore the existence of a possible rare variant burden in SIDS cases. With potential relatedness, ancestry, and genetic admixture confounders elegantly eliminated, an ultra-rare variant case−control analysis was performed among SIDS cases of European ancestry and European control subjects. Interestingly, ultra-rare variants in the 4 major channelopathies genes were found to be significantly overrepresented in SIDS cases with respect to control subjects (6.5% vs. 3.1%; p = 0.013). The investigators therefore stated that “channelopathies may represent the underlying pathological basis for some of SIDS cases,” but they seemed reluctant to dig deeper. We firmly believe that SIDS, apart from its monogenic causal origin, as dictated by penetrant variants able per se to cause arrhythmias, may also be the outcome of particular genetic signatures, an aggregation of GHD variants acting synergistically and operating under different levels of control from environmental and other factors.
Although the percentage of SIDS that may now be safely attributed to channelopathies may be somewhat lower than previously assumed, the early identification of infants affected by LQTS remains essential for their protection through the effective therapies available (15) because those who do not die in the first year of life remain at high risk for sudden death in childhood or adolescence. Therefore, the value of neonatal electrocardiographic screening (16) remains unchallenged, also because there is 8.3% of SIDS cases still seeking genetic justice. In the latter, over- and under-attribution of genetic causality would be the mirror image of 2 evils.
The authors thank Pinuccia De Tomasi, BS, for expert editorial support.
↵∗ Editorials published in the Journal of the American College of Cardiology reflect the views of the authors and do not necessarily represent the views of JACC or the American College of Cardiology.
Both authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- 2018 American College of Cardiology Foundation
- Maron B.J.,
- Clark C.E.,
- Goldstein R.E.,
- Epstein S.E.
- Arnestad M.,
- Crotti L.,
- Rognum T.O.,
- et al.
- Tester D.J.,
- Wong L.C.H.,
- Chanana P.,
- et al.
- Leinonen J.T.,
- Crotti L.,
- Djupsjöbacka A.,
- et al.
- Lahti A.L.,
- Kujala V.J.,
- Chapman H.,
- et al.
- Jou C.J.,
- Barnett S.M.,
- Bian J.T.,
- Weng H.C.,
- Sheng X.,
- Tristani-Firouzi M.
- ↵NCBI. ClinVar Public Archive. NM_000238.3(KCNH2):c.526C>T (p.Arg176Trp). Available at: https://www.ncbi.nlm.nih.gov/clinvar/variation/67509/#clinical-assertions. Accessed January 17, 2018.