Genetic testing is part of the clinical diagnosis in IAS, including SQTS or phenotype-like entities. However, only variants with a conclusive role should be actionable in clinical practice (Landstrom et al. 2023). One of the main challenges is a precise interpretation of rare variants, the remaining large part with an unknown role following current ACMG recommendations. In addition, current guidelines recommend that genetic diagnosis in patients with SQTS or any phenotype-like variants, should be restricted only to genes with a definite disease-association (Wilde et al. 2022). This restricted approach helps to avoid overinterpretation of rare variants found in genes with little evidence for disease causality. For this reason, at our point of view, the main step before performing a genetic test is to conclude a definite clinical diagnosis (Martinez-Barrios et al. 2022). In our study, we updated all rare variants previously reported as the cause of SQTS or any phenotype-like variant. We used the available tools, but with a particular evidence-based framework to perform an exhaustive reinterpretation. We identified 34 rare variants but only nine variants still played a deleterious role associated with a definite SQTS phenotype after our reanalysis. These variants were located in the four principal genes (KCNQ1, KCNH2, KCNJ2 or SLC4A3) currently associated with SQTS (Walsh et al. 2022). Additional rare variants situated in other genes were associated with other conditions with phenotypic shortened QT intervals, but did not lead definite diagnosis of SQTS. It is mandatory to clarify the clinical burden of variants in SQTS, especially those previously classified as VUS, helping to conclude a definite role and therefore, solve the dilemma for clinical teams and reduce the ambiguity in families carrying many rare variants, even in currently well-established genes. In consequence, recontact to families should be mandatory if a reclassification of variant occurs, mainly if it results in a change in clinical actionability.

Nowadays, only rare variants located in four genes have been associated with a definite phenotype of SQTS (Walsh et al. 2022; Wilde et al. 2022), as observed in our study. Concretely, out of 24 rare variants previously associated with SQTS, our reanalysis observed that only 20 rare variants were related to a definite phenotype of SQTS. The four remaining variants were located in the KCNH2 gene but were associated with a phenotype of BrS with stnQT. Therefore, the four rare variants in this gene remain classified as VUS (p.Thr152Ile, p.Arg164Cys, p.Trp927Gly and p.Arg1135His), according to current available data of BrS patients with shorter QTc (Wang et al. 2014). In total, nine variants remain currently classified as LP following ACMG recommendations after the update (KCNH2: p.Ile560Thr, p.Asn588Lys and p.Thr618Ile. KCNJ2: p.Asp172Asn and p.Glu299Val. KCNQ1: p.Arg259His, p.Ala287Thr and p.Phe299Val. SLC4A3: p.Arg370His). It is imperative to remark that despite a LP role reported, clinical translation should be done with caution, especially without conclusive familial segregation. Therefore, a personalized genetic interpretation is also mandatory in actionable variants.

The main gene currently associated with SQTS is KCNH2, as observed in our study. We identified five variants, three of which classified as LP (p.Ile560Thr, p.Asn588Lys and p.Thr618Ile) (Butler et al. 2019; Du et al. 2022; Huang et al. 2021; Shiti et al. 2023; Zhang et al. 2022; Zhao et al. 2019), and the p.Ser631Ala (Akdis et al. 2018), variants remain as a definite deleterious role following the ACMG recommendations due to lack of sufficient data. The clinical role of both these variants may be underestimated and may lead to confusion in clinical translation. For this reason, our approach focused on definite diagnosis, definite gene and no MAF concluded a highly potential deleterious role as the cause of SQTS. In the KCNJ2 gene, a total of four rare missense variants were identified in cases with definite SQTS, one in relation with isolated SQTS (p.Asp172Asn) (Du et al. 2021), two in patients with SQTS concomitant to other entities such as AF (p.Glu299Val and p.Met301Lys) (Deo et al. 2013; Hasegawa et al. 2014), or ASD (p.Lys346Thr) (Ambrosini et al. 2014). The first two rare variants remain classified with a definite LP role following the ACMG recommendations but the last two remain as VUS due to the lack of data. None showed MAF, but with a definite diagnosis and located in a definite gene, suggesting a highly potential deleterious role, as predicted by our approach. In the KCNQ1 gene, we identified six rare missense variants related to SQTS, two associated with isolated SQTS (p.Phe279Ile and p.Ala287Thr) (Schneider et al. 2021), and four concomitant to AF (p.Val141Met, p.Arg259His, p.Phe299Val and p.Val307Leu) (Hong et al. 2005; Mazzanti et al. 2014; Moreno-Manuel et al. 2024; Wu et al. 2015). Three showed a definite LP role and three remained as VUS, following the ACMG recommendations, but three remained as VUS due to the lack of data (p.Arg259His, p.Ala287Thr and p.Phe299Val). Our approach predicted a highly suspected deleterious role due to be presented in patients with a definite diagnosis, being in a definite gene and showing a very low or unavailable MAF. The fourth gene is SLC4A3, and we identified five rare missense variants associated with a definite SQTS (p.Arg370His, p.Arg600Cys, p.Arg621Trp, p.Glu852Asp and p.Arg952His) (Christiansen et al. 2023; Thorsen et al. 2017). Only the first variant is classified as LP, following the ACMG recommendations. The other variants remain as VUS due to lack of adequate data to conclude a deleterious role. Our approach, supported by a definite diagnosis, being in a definite gene, and having a very low or unavailable MAF, concluded a highly potential deleterious role as a cause of SQTS.

All updated deleterious variants located in definite genes associated with SQTS are related to a high risk of malignant arrhythmias. Therefore, there is an urgent need to identify compounds that allow us to adopt effective pharmacological and non-pharmacological therapies in diagnosed patients. However, pharmacological treatment must be personalized in the SQTS subtype, depending on the altered gene. In recent years, studies using the human induced pluripotent stem cells (hiPSCs) derived to cardiomyocytes (CMs) have allowed to identify appropriate drugs for SQTS, especially type 1 (due to pathogenic variants in the KCNH2 gene) (El-Battrawy et al. 2018; Zhao et al. 2019). In addition, other approaches have been proposed to help in the study of the potential drug response in clinical practice of a novel compound identified using hiPSC-CMs. The most promising tool has been mathematical modeling to provide insight into the drug effects on mechanisms of shortening (Jaeger et al. 2019; Jaeger et al. 2020). This dual-component therapy, also known as SupRep, uses cloning into a single construct of a custom-designed short hairpin KCNH2 RNA with an approximately 80% reduction of the mutated allele (deletion) and a KCNH2 cDNA with the correct version of the “short hairpin RNA-immune” allele (substitution). This therapy has been shown to effectively correct APD and rescue the LQT2 and SQT1 disease phenotypes in patient-derived iPSC-CM (in vitro)(Bains et al. 2022). However, these gene therapies are still in pre-clinical phases, and they still need to be tested in humans.

Our update showed that 14 rare variants should be considered with a deleterious or potentially deleterious role in entities with a corrected QT interval shorter than normal but not definite SQTS. Four of these rare variants are located in the KCNQ1 gene (as mentioned before). Other 10 rare variants are placed on three genes: four in CACNA1C, one in CACNB2 and five in the SLC22A5 gene. These three last genes have been potentially associated with similar phenotypes to SQTS, in concordance to current data (Walsh et al. 2022; Wilde et al. 2022), but never with a conclusive definite diagnosis of SQTS, at least to date. Of these rare variants, only four remain classified with a definite pathogenic role following ACMG recommendations: one in CACNB2 (p.Ser480Leu) associated with BrS and stnQT (El-Battrawy et al. 2021; Zhong et al. 2022), and three in the SLC22A5 gene (p.Phe17Leu, p.Phe23del and p.Arg289Ter) associated with RCTD and stnQT following an autosomal recessive pattern of inheritance (Roussel et al. 2016). As mentioned, despite these pathogenic variants being associated with SQTS phenotype-like entities (BrS + stnQT, ASD + stnQT and RCTD + stnQT), translation into clinical practice should be performed in a personalized way. If adoption of treatment is carried out, recent published data should be taken into account, recent published data but also family segregation, and, most importantly, clinical symptoms of the patient. All other rare variants remain as VUS due to lack of data after a comprehensive update, therefore not actionable following the current guidelines (Wilde et al. 2022). However, no available MAF suggest a high potential deleterious role in each of phenotype-like variant identified, and all variants should be re-analysed periodically and not discarded as a potential cause of the reported phenotype.

Finally, it is important to note that rare variants located in two genes previously associated with SQTS phenotype-like, should not be included in the potential cause of disease due to updated data suggesting a non-deleterious role, mainly due to a non-conclusive phenotype or a high increase of MAF (SCN5A_p.Arg689His and CACNA2D1_p.Ser755Thr). Our update agrees with recent results supporting the disputed association of these variants/genes with SQTS phenotype-like cases (Walsh et al. 2022). Other variants in additional genes have been proposed as the cause of potential alterations in the reduction of the QT interval (ANK2_pArg3634Asp, PKP2_p.Asp26Asn and ABCC9_p.Arg633Cys) (Treat et al. 2021), but some of them have a MAF extremely high to be consider a causative variant of SQTS, and others require further studies to conclude a definite role in the SQTS or phenotype-like entities. In addition, an in vivo study has suggested that the ion channel modulator nitric oxide synthase 1 adaptor protein (Nos1ap) overexpression causes targeted subcellular localization of Nos1 to the CaV1.2 with a subsequent decrease of the QT interval in transgenic mice (Jansch et al. 2023). However, no clinical studies have been published to date to conclude a definite association in humans.

Our interpretation has some limitations. Firstly, the limited number of families reported worldwide with a definite diagnosis of SQTS impedes conducting a large and significant study. Second, the lack of available information for some analyzed variants impeded a comprehensive reanalysis of these rare variants, especially those classified as VUS. It is important to note that additional data may be available in the next future that may change the current reanalysis performed in our study. Lastly, those who should perform the reanalysis and assume the economic cost are not analyzed in our study, despite being a key point from our point of view.