Atypical presentation of dopa‐responsive dystonia in Taiwan

Abstract The typical clinical presentation of dopa‐responsive dystonia, which is also called Segawa disease, is a young age of onset, with predominance in females, diurnal fluctuation of lower limb dystonia, and fair response to low‐dose levodopa. This disease has both autosomal dominant and autosomal recessive inheritance. Autosomal dominant Segawa disease is caused by GCH1 mutation on chromosome 14q22.1‐q22.2. Here, we report the case of a male patient with genetically confirmed Segawa disease and atypical presentations including no diurnal symptom fluctuation and insufficient response to levodopa. The patient's father who had the same mutation presented parkinsonism in old age. We also review the literature to address the broad clinical heterogeneity of Segawa disease and the influence of onset age on clinical presentation.


| INTRODUCTION
Segawa disease, also called dopa-responsive dystonia (DRD), is an autosomal inherited disease classified as a dystonia-plus syndrome; it was first described by Segawa, Ohmi, Itoh, Aoyama, and Hayakawa (1971) as a disease with marked diurnal fluctuation of dystonia. Segawa disease is characterized by a typical clinical presentation of youngonset lower limb dystonia (Van Hove et al., 2006) and the dystonia becomes more severe near the end of the day (Nygaard, Marsden, & Duvoisin, 1988). Another characteristic is that the dystonia favorably responds to low-dose levodopa (Nygaard et al., 1988). The prevalence of this disease is approximately 0.5-1 per million (Segawa, Nomura, & Nishiyama, 2003;Tg, 1993;Zirn et al., 2008) and the data could be highly under-representative due to mild symptoms in some patients; because of a higher penetrance of the causative genes among females, the incidence of Segawa disease is 2-4 higher in females than in males (Furukawa et al., 1998;Steinberger et al., 1998). With time, the lower limb dystonia progresses to generalized dystonia (Segawa et al., 2003).

| CASE REPORT
A 43-year-old man, without underlying diseases, had left-foot dystonia for 3 years since 2011. The symptoms had progressed for 2 months before he sought medical help. Because of dystonic movement, he could not walk normally on flat ground; nevertheless, he could walk backward, run, and climb upstairs and downstairs without difficulty. No diurnal fluctuation of his left-foot dystonia was noted. The birth and developmental histories were normal for this patient. His personal, trauma, and drug exposure histories were also unremarkable. However, in elementary school, he was informed that he had a mildly abnormal walking stance, although this did not impair his daily function. His father was diagnosed of Parkinson' s disease since his 70s with the clinical presentations of rest tremor, bradykinesia and rigidity on the left side and then festinating gait. These symptoms responded to levodopa treatment.
He had regullarily follow-up in another medical center in Taipei and we had examined and interviewed him once in our outpatient clinic.
He also had REM sleep behavior disorder symptom. The patient's grandmother also had parkinsonian features according to patient's statement. Other family members are unremarkable ( Figure 1).
Extensive neurological examination of the patient did not reveal any significant abnormality including parkinsonian signs, except for inversion of the left ankle with dorsiflexion of the big toe when the patient walked on the ground. No Kayser-Fleischer ring or parkinsonian symptom was noted. His medication including trihexyphenidyl up to 4 mg and clonazepam 0.5 mg was ineffective. He was also administered levodopa with DTI (decarboxylase inhibitor) 100 mg per day, which alleviated his symptoms for only the first 10 days.
However, thereafter, even after increasing the dose to 300 mg per day, the patient subjectively felt no response; thus, he stopped medication use. His laboratory survey results including ceruloplasmin, blood and urine copper levels and thyroid, liver, and renal function were normal. The results of imaging studies including brain MRI and 99m Tc-TRODAT-1 SPECT were also unremarkable ( Figure 2).
For further diagnosis of dystonia, we conducted a genetic study; the results confirmed Segawa disease with GCH1 mutation at exon 6 c.670 A>G(p.Lys224Glu) ( Figure 3). We only conducted genetic study of DJ1, ATP13A2, PINK1, UCHL1, SNCA, PARKIN, and LRRK2 using multiplex ligation-dependent probe amplification (MLPA) assay for Parkinson's disease due to pure dystonia symptom for long time without parkinsonian symptom and normal TRODAT study.
There was no copy number variation in these genes. Genetic testing for dystonia-causing genes such as Tor1A, TH, or PRKRA was not F I G U R E 1 Pedigree of the study family. A genetic study was performed for the index case (arrow) and his father. Black indicates positive clinical signs; gray indicates suspicious clinical symptoms. Patient with parkinsosian symptoms was indicated with horizontal line symbol and patient with dystonia symptoms was indicated with dotted symbol Current age 46/onset age elementary school Current age 81 /onset age 70s onset age unknown F I G U R E 2 Image of 99mTc-TRODAT-1 SPECT of index patient. The image revealed bilateral symmetry of the normally shaped caudate nucleus and putamen. Symmetrical distribution of dopaminergic radioactivity was noted in the striatum. The result revealed no presynaptic uptake decreasing conducted as the subject's presentation was not consistent with the phenotype reported for each of these mutations. Because of the parkinsonian presentation observed in his father and possibility of inheritance, we arranged a genetic study for his father; he was also confirmed to have the same gene mutation at exon 6 c.670 A>G.
The patient's grandmother had already passed away; therefore, no further genetic study could be performed. We further conducted genetic study of c.670 in a cohort of 173 normal controls with 346 chromosomes, and no c.670 A>G was found. Moreover, because the patient did not have fair response to levodopa, we performed botulinum injection over his anterior tibialis, tibialis posterior, and extensor hallucis longus muscles. Consequently, his dystonia improved considerably from Global Dystonia Rating Scale (GDS) 5 to 1. The patient was satisfied with the treatment results.

| DISCUSSION
We report the case of an atypical presentation of Segawa disease including insufficient drug response and no diurnal fluctuation of dystonia. A good response to low-dose levodopa is considered characteristic for the diagnosis of Segawa disease (Müller, Steinberger, & Topka, 2002;Nygaard et al., 1988). Diurnal fluctuation is often present in the affected patients (Nygaard et al., 1988) and the rate of this fluctuation is approximately 94% (Liu et al., 2010). After searching PubMed for reports on Segawa disease, we found only a few reports mentioning poor or incomplete levodopa response (Bandmann et al., 1996;Grimes, Barclay, Duff, Furukawa, & Lang, 2002;Regula et al., 2007;Romstad et al., 2003;Tadic et al., 2012;Tassin et al., 2000).
One patient had tyrosine hydroxylase deficiency and other cases had GCH1 mutation. According to the previous reports, a daily levodopa dose of <20 mg/kg is sufficient for treating Segawa disease without a DTI (Segawa et al., 2003) whereas with a DTI, a daily dose of 300 mg, and sometimes 400 mg, is required (Nygaard, Snow, & Fahn, 1993;Steinberger et al., 2000). Although the ineffectiveness of levodopa may be explained by the extent of its decarboxylation in the intestines, patients with Segawa disease having insufficient drug response have rarely been reported thus far. In our patient, both the absence of diurnal fluctuation of dystonia and insufficient levodopa response rendered the diagnosis of Segawa disease increasingly difficult.
Moreover, the clinical course of Segawa disease is onset agedependent (Segawa, Hosaka, Miyagawa, Nomura, & Imai, 2000). In Trender-Gerhard's 2008 cohort study (Trender-Gerhard et al., 2009) 34 patients with Segawa disease were divided into four groups according to their clinical presentation: young-onset classic (onset age, 4.5 ± 4.6 years), young-onset mild (onset age, 7.3 ± 3.9 years), young onset with severe initial hypotonia (onset age, 0.3 ± 0.3 years), and adult onset (onset age, 37.0 ± 8.2 years). In the young-onset classic group, the female-to-male ratio was 2.3:1, a normal ratio for Segawa disease. However, in the young-onset mild and adult-onset groups, male predominance was noted (female:male = 1:3). This result reconsiders the basic concept of female predominance in Segawa disease. In addition, in the young-onset mild group (Trender-Gerhard et al., 2009) the patients' symptoms were mild, not even requiring treatment. In our case, the patient had mild symptoms and had no severe progression until adulthood; this presentation is similar to that reported in the literature regarding gender as well as clinical presentation (Trender-Gerhard et al., 2009). Furthermore, tremors or parkinsonism can be the main or only sign in adult-onset patients (Segawa et al., 2003). In the adult-onset group of Trender-Gerhard's study, all the adult patients presented tremors or parkinsonism as main symptoms. In our case, the patient's father was confirmed to have GCH1 mutation with typical parkinsonian features, which is also similar to the report in the previous study (Segawa et al., 2003;Trender-Gerhard et al., 2009). A similar classification has been summarized previously (Bernal-Pacheco et al., 2013). According to the literature and our observations, we conclude that mild symptoms of Segawa disease tend to occur in young-onset male patients and that pure parkinsonism could be a common feature in adult-onset male patients with CGH1 mutation.
For the completely different clinical presentation of Segawa disease in our report, we further searched Pubmed with the keywords "Parkinsonism" and "Segawa" or "DRD, " and we found a few familial study articles on Segawa disease with symptoms of both dystonia and parkinsonism that are similar to those in our familial study (Table 1). In these families, adult-onset cases apparently tended to present pure parkinsonism F I G U R E 3 Genetic study of the patient. A mutation was noted at exon 6 c.670 A>G (Table 2); in addition, the number of adult-onset male patients was higher than that of female patients (Table 2). Young-onset patients could have more varied presentation but still predominantly had pure dystonia or parkinsonism with dystonia (Table 3). In the meta-analysis study of Tadic et al., adult-onset patients could have more parkinsonian presentation than young-onset patients and this result is also not against our finding (Tadic et al., 2012). In a recent study of Mencacci et al. (2014) they found GCH1 gene could possibly have a correlation with Parkinson's disease via whole exome sequencing;. Another study of GCH1 single-nucleotide polymorphisms revealed possible increased risk of Parkinson's disease (Chen et al., 2016). Whether GCH1 mutation could be pathogenic to Parkinson's disease remained uncertain, but these results could possibly give us a new idea that parkinsonism may not necessary just be symptoms of DRD.
Until 2016, 229 mutations were collected in the Human Gene Mutation Database (http://www.hgmd.org/). The GCH1 mutation at exon 6 c.670 noted in our patient has been reported previously in patients with DRD (Bandmann et al., 1996;Jarman, Bandmann, Marsden, & Wood, 1997;Mencacci et al., 2014). Although this is not the first report regarding this mutation location, the previous reports reported either lysine to arginine or lysine to threonine change and cause conservative amino acid change or stop codon, whereas we discovered a lysine to glucine change in our case (Bandmann et al., 1996;Jarman et al., 1997). We also confirmed that lysine to glucine change is not a variant but a mutation in our study. The clinical presentations of each patient are different, and the mechanisms of this mutation to cause various clinical features are still not well understood.
In conclusion, we report a case with an atypical presentation of Segawa disease and aim to address the broad clinical heterogeneity of Segawa disease.

ACKNOWLEDGMENTS
We thank the patients who participated in this study. Portions of this work were supported by grants 104-2314-B-182A-025-MY3 from the Ministry of Science and Technology, Taipei, Taiwan.
T A B L E 2 Further evaluation of adult-onset cases with pure parkinsonism or dystonia symptom according to gender