Updated frequency analysis of spinocerebellar ataxia in China

Sir, We read with great interest the recent article published in Brain by (Coutelier et al., 2017) reporting a genetic screen of spinocerebellar ataxia (SCA) cases previously excluded for (CAG)n/polyglutamine expansions, via panel sequencing and TaqMan® PCR assay, and identifying a high frequency of channelopathies (CACNA1A, KCND3, KCNC3 and KCNA1) and possible risk factor variants in those cases. This may help elucidating causality and genotype–phenotype correlation in non-polyglutamine SCA. We would like to report our updated findings on genetic frequency analysis in 1489 index ataxia cases in Chinese Han cohort (including 1009 dominant cases and 480 sporadic cases) and investigation on causality in non-polyglutamine SCA cases in China. Patients were recruited from April 1994 to September 2017. Our study was approved by the ethics committee of Xiangya Hospital, Central South University. Blood samples were collected after written informed consent from all subjects. A list of genes implicated in SCAs were selected, including those carrying: (i) (CAG)n expansions: ATXN1 (SCA1), ATN1 (dentatorubral-pallidoluysian atrophy, DRPLA), ATXN2 (SCA2), ATXN3 (Machado-Joseph disease, MJD/SCA3), CACNA1A (SCA6), ATXN7 (SCA7), PPP2R2B (SCA12) and TBP (SCA17); (ii) other oligonucleotide expansions: ATXN8OS (SCA8), ATXN10 (SCA10), BEAN (SCA31) and NOP56 (SCA36); and (iii) conventional mutations: SPTBN2 (SCA5), CACNA1A, TTBK2 (SCA11), KCNC3 (SCA13), PRKCG (SCA14), ITPR1 (SCA15/16), KCND3 (SCA19/22), TMEM240 (SCA21), PDYN (SCA23), EEF2 (SCA26), FGF14 (SCA27), AFG3L2 (SCA28), ELOVL4 (SCA34), TGM6 (SCA35), ELOVL5 (SCA38), CCDC88C (SCA40), TRPC3 (SCA41) and CACNA1G (SCA42). As the most frequent cause of SCAs, (CAG)n expansions in ATXN1, ATN1, ATXN2, ATXN3, CACNA1A, ATXN7, PPP2R2B and TBP were firstly genotyped in 1489 index cases (dominant and isolated forms). The relative frequency of each subtype was consistent with our previous observations (Tang et al., 2000; Wang et al., 2011). Among 1009 dominant ataxia (SCA) probands, MJD/SCA3 (n = 632, 62.64%) remained the most common subtype, followed by SCA2 (n = 88, 8.72%), SCA1 (n = 73, 7.23%), SCA6 (n = 17, 1.68%) and SCA7 (n = 11, 1.09%), SCA17 (n = 3, 0.29%), SCA12 (n = 1, 0.09%) and DRPLA (n = 1, 0.09%), showing the predominance of (CAG)n/polyglutamine expansions. Among isolated patients, 18.91% could also be attributed to expanded repeats (Fig. 1). Workflow of the genetic screening performed, and relative frequency found for each subtype in Chinese dominant and isolated ataxia cases. After exclusion of (CAG)n expansions, genetic analyses for other oligonucleotide expansions and for conventional mutations were performed; in the latter, the pathogenic classification was as described by (Coutelier et al., 2017). Thus, (CTA)n/(CTG)n expansions in ATXN8OS of SCA8 (n = 6, 0.59%) and (GGCCTG)n expansions in NOP56 of SCA36 (n = 8, 0.79%) were detected in dominant cases, while SCA36 was diagnosed in one isolated patient. Unlike the observation of frequent channel genes accounting for non-polyglutamine ataxia in the SPATAX cohort (Battistini et al., 1999; Coutelier et al., 2017), only one known missense variant (c.1748G>A) was identified in CACNA1A, and no disease-causing variants were seen in KCNC3, KCND3 or CACNA1G underlined channel network. Pathogenic variants were found in PRKCG for SCA14 (n = 1, 0.09%) and TGM6 for SCA35 (n = 2, 0.19%) in the dominant cases; and de novo mutations were found in TMEM240 of SCA21 and CCDC88C of SCA40 in the isolated cases (Supplementary Table 1). According to these data, only 16.42% of dominant cases and as much as 80.49% of the isolated ones remained genetically undiagnosed (Fig. 1). We compared age-at-onset among patients with and without polyglutamine expansions (Supplementary Table 2). In contrast to the SPATAX study, patients with (CAG)n expansions had a lower mean age-at-onset (34.7 versus 40.9 years, P = 0.008), possibly due to the small percentage of earlier onset in patients with a channelopathy. Although this small size limits statistical power, the tendency observed imply further studies for genotype–phenotype correlations in non-polyglutamine SCA patients in the Chinese cohort. Here, we report updated relative frequency of SCA subtypes in dominant and isolated cases, in the largest sample described so far in China. As could be expected, the overall diagnostic yield is much higher in familial cases, though genes implicated in recessive ataxia may still need to be studied in the group of isolated patients. Besides, risk-factor variants also need to be taken into account, though they seem to be insufficient to trigger SCAs. Compared to undiagnosed dominant cases ranging 33–92% worldwide (Ruano et al., 2014), our study showed a relative high overall diagnostic yield in dominant ataxia; however, similar to the SPATAX cohort, the diagnostic rate (9.83%) in the 183 cases excluded for (CAG)n expansions was low, suggesting the low prevalence in the Chinese cohort of rarer subtypes. Thus, polyglutamine genes should be excluded before multigene panels and whole exome/genome sequencing is undertaken; these might lead to the identification of new causative genes. Furthermore, massively parallel sequencing could help determining multiple molecular diagnoses in a single patient, involving two or more independent loci, as noted by Posey et al. (2017). As the coexistence of two SCA subtypes in the same patient complicates genetic heterogeneity (Izumi et al., 2003; Kapur and Goldman, 2012; Roda et al., 2017), such high-throughput approach will be especially important for accurate diagnosis and to fully understand genotype–phenotype correlation in SCA. Also, the low frequency of channelopathies in this Chinese SCA cohort further confirms the variable distribution of SCA subtypes across ethnically and geographically different populations. This strengthens the need for a population-tailored genetic counselling. The authors would like thanking patients and their family members for their generous participation in this study. This study was supported by the National Natural Science Foundation of China (No. 81471156, 81771231 to Hong Jiang; No. 81430023, 81130021 to Beisha Tang; No. 31401135 to Rong Qiu), National Key Research and Development Program of China (No. 2016YFC0901504 and 2016YFC0905100 to Hong Jiang, 2016YFC1306000 to Beisha Tang), Clinical Research Funds of Xiangya Hospital (No. 2014L03 to Hong Jiang), the Clinical and rehabilitation fund of Peking University Weiming Biotech Group (No. xywm2015I10 to Hong Jiang), and Youth Foundation of Xiangya Hospital (No. 2017Q03 to Zhao Chen). Supplementary material is available at Brain online.