To the best of our knowledge, this is the first study to estimate the prevalence and incidence of HD in Korea and the largest HD series of multi-center study in the Asian population. Most of the previous Asian epidemiologic studies of HD were performed on a regional basis, such as in Hong Kong, China or the San-in area of Japan [
10,
11]. In Taiwan, the average annual incidence rate was estimated at 0.1 and the prevalence at 0.42 per 100,000 persons, based on the National Health Insurance database [
12]. The annual incidence estimated in our study is about half of the average annual incidence of Taiwan; however, our estimate is similar to the annual incidence in China (0.046 per 100,000) [
10]. Our estimation of the prevalence of HD in Korea based on two data sources was 0.41 per 100,000 in 2013 (Korean RDR) and 0.38 per 100,000 bewteen 2009 to 2013 (National Health Insurance database), which are comparable to Taiwanese data and a meta-analysis [
11,
13]. It is known that HD is less prevalent in Asian countries compared with Western countries, in which the overall prevalence is 5.70 according to a meta-analysis [
13]. Recent data suggests that the prevalence of HD may be over 10 per 100,000 in the UK [
14]. These racial differences in prevalence are accounted for by the differences in CAG tract size, HTT haplotype and CCG polymorphism [
15–
17]. The mean CAG repeat length of the normal allele was reported to be significantly longer in the European (18.4 ± 3.7) population compared with the Chinese, Japanese, and Thai (16.4 ± 1.5, 16.6 ± 1.5, and 16.5 ± 1.9, respectively) population [
16,
18]. Furthermore, the distribution of CAG length was positively skewed in Europeans, which represents a length-dependent mutational bias towards longer alleles [
19]. In our collected dataset, we did not observe skewness in CAG length of normal alleles. The mean CAG repeat length of the normal allele in our HD patients was 17.4 ± 3.2, which is longer than that of other Asian populations, but shorter than that reported by a previous single center study in Korea [
20] and than those of Western populations. Recently, HTT haplogroups (types A, B, and C) were identified, and CAG instability was more likely to occur in the cis-acting haplogroup A [
15]. HTT haplogroups are known to show a regional difference, and the higher risk haplotype was infrequent in China, Japan, and Thailand [
18,
21,
22]. The CCG polymorphic region is located adjacent to the CAG triplet repeat of the HD gene, and there is a linkage disequilibrium between CCG polymorphism and the pathological CAG expansion according to the race [
17,
23]. HTT haplogroup and CCG polymorphism have not yet been studied in Korean HD patients.
The mean age at initial symptom onset (44.16 ± 14.08) and the inverse correlation between age at initial symptom onset and CAG repeat expansion size were compatible with previous reports [
20,
24–
26]. The mean length of CAG repeats were similar to the previously reported range of 44.7 to 49.0 (range: 36–70) in Korea and Western countries [
5–
7,
20,
27]. Moreover, the length of the expanded CAG repeat determines the age of onset in approximately 50–70% of patients in the 40 to 55 CAG repeat range, but possible roles of other genetic and environmental modifiers are currently being investigated [
28,
29].
Chorea is the most frequent initial symptom (60.3%) in our collected dataset as well as the chief complaint (85.3%), and the latter is compatible to the previously reported frequency (89%) at first visit to hospital [
20]. However, this finding also showed that approximately 40% of HD patients had initial symptoms other than chorea. Among the symptomatic triad of problems involving movement, cognition, and mood, the most prominent deficit may differ between patients, and recent studies have recognized that non-motor symptoms occur before overt motor features in a fraction of patients [
30,
31].
About one third of the patients comprising our collected dataset had no family history of HD, which is higher than reported in a study from British Columbia (29.8%) [
32] and in a previous report from a single center in Korea (20%).
20 Given that the probability of
de novo expansion of CAG repeats through paternal transmission was estimated to range from 1/6,241 to 1/951 [
33], the higher rate of HD patients with a negative family history might not be explained solely by this mechanism. In addition to
de novo expansion of CAG repeats, unstable familial transmission, reduced penetrance or underreported family history may be another cause of negative family history among HD patients [
34–
36]. Interestingly, our HD patients without family history showed a delayed age at onset of initial symptoms, a delayed age at genetic diagnosis and a prolonged interval between the age of initial symptom onset to the age of genetic diagnosis compared with the HD patients with a positive family history. Lower CAG repeat number in negative family history groups may be related with these findings; however, a clinician’s delay in suspicion of HD in choreic patients apparently lacking family history may have also contributed to this observation. Our study has some limitations. Because this research is based on retrospective chart review, extensive data gathering was impossible. Even after the diagnosis of HD, some patients may have refused to be registered into RDR because of social stigma and discrimination, which may have resulted in the underestimation of the real incidence and prevalence of HD [
37].
In conclusion, our estimated incidence and prevalence of HD were comparable with previous reports of Asian populations and were relatively low compared with estimates for Western populations. The clinical features showed similar phenotypes, and the effect of CAG repeat expansion on the age of disease onset was comparable. Our analyses using national registries and multi-center clinical data could be useful for future studies and large-scale investigations in HD patients.