Movement Disorders

Vol. 23, No. 15, 2008, pp. 2232–2238

2008 Movement Disorder Society

Early Symptoms in Spinocerebellar Ataxia Type 1, 2, 3, and 6

Christoph Globas, MD,1 Sophie Tezenas du Montcel, MD, PhD,2,3 Laslo Baliko, MD,4 Syliva Boesch, MD,5
Chantal Depondt, MD,6 Stefano DiDonato, MD,7 Alexandra Durr, MD,8,9 Alessandro Filla, MD,10
Thomas Klockgether, MD,11 Caterina Mariotti, MD,7 Bela Melegh, MD, PhD,12 Maryla Rakowicz, MD,13
Pascale Ribai, MD,8 Rafal Rola, MD,14 Tanja Schmitz-Hubsch, MD,11 Sandra Szymanski, MD,15
Dagmar Timmann, MD,16 Bart P. Van de Warrenburg, MD,17 Peter Bauer, MD,18 and Ludger Schols, MD1*
1
Department of Neurology and Hertie-Institute for Clinical Brain Research, University of Tübingen, Tübingen, Germany
2
Modelling in Clinical Research, EA 3974, Pierre and Marie Curie Paris 6 University, Paris, France
3
AP-HP, Biostatistics and Medical Informatics Unit, Pitie´-Salpeˆtrie`re Hospital, Paris, France
4
Department of Neurology and Stroke, County Hospital, Veszpre´m, Hungary
5
Department of Neurology, University of Innsbruck, Innsbruck, Austria
6
Department of Neurology, University of Brussels, Brussels, Belgium
7
Division of Biochemistry and Genetics, Fondazione Instituto Neurologico C. Besta, Milan, Italy
8
INSERM, Hôpital de la Salpeˆtrie`re, UMR 679, Paris, France
9
APHP, De´partement de Ge´ne´tique et Cytoge´ne´tique, Hôpital de la Salpeˆtrie`re, Paris, France
10
Department of Neurology, University of Naples, Naples, Italy
11
Department of Neurology, University of Bonn, Bonn, Germany
12
Department of Medical Genetics and Child Development, University of Pe´cs, Pe´cs, Hungary
13
Department of Clinical Neurophysiology, Institute of Psychiatry and Neurology, Warsaw, Poland
14
First Department of Neurology, Institute of Psychiatry and Neurology, Warsaw, Poland
15
Department of Neurology, St. Josef Hospital, Ruhr-University Bochum, Bochum, Germany
16
Department of Neurology, University of Essen, Essen, Germany
17
Department of Neurology, University of Nijmegen, Nijmegen, Netherlands
18
Department of Medical Genetics, University of Tübingen, Tübingen, Germany

Abstract: Onset of genetically determined neurodegenerative 1 year or more in 44% of cases. Influence of repeat length
diseases is difficult to specify because of their insidious and on age of onset was maximum when onset was defined as
slowly progressive nature. This is especially true for spino- beginning of permanent gait disturbance and cases with
cerebellar ataxia (SCA) because of varying affection of many symptoms for more than 10 years were excluded. Under
parts of the nervous system and huge variability of symp- these conditions, CAG repeat length determined 64% of
toms. We investigated early symptoms in 287 patients with onset variability in SCA1, 67% in SCA2, 46% in SCA3, and
SCA1, SCA2, SCA3, or SCA6 and calculated the influence 41% in SCA6 demonstrating substantial influence of nonrep-
of CAG repeat length on age of onset depending on (1) the eat factors on disease onset in all SCA subtypes. Identifica-
definition of disease onset, (2) people defining onset, and (3) tion of these factors is of interest as potential targets for
duration of symptoms. Gait difficulty was the initial symptom disease modifying compounds. In this respect, recognition of
in two-thirds of patients. Double vision, dysarthria, impaired early symptoms that develop before onset of ataxia is manda-
hand writing, and episodic vertigo preceded ataxia in 4% of tory to determine the shift from presymptomatic to affected
patients, respectively. Frequency of other early symptoms did status in SCA.
2008 Movement Disorder Society
not differ from controls and was regarded unspecific. Data Key words: spinocerebellar ataxia; early symptoms; deter-
about disease onset varied between patients and relatives for minants of age at onset; CAG repeat expansion

*Correspondence to: Dr. Ludger Schöls, Department of Neurology Received 4 June 2008; Revised 21 July 2008; Accepted 27 July
and Hertie Institute for Clinical Brain Research, University of Tübin- 2008
gen, Hoppe-Seyler-Str. 3, D-72076 Tübingen, Germany. Published online 29 August 2008 in Wiley InterScience (www.
E-mail: ludger.schoels@uni-tuebingen.de interscience.wiley.com). DOI: 10.1002/mds.22288
Potential conflict of interest: Nothing to report.

2232
 EARLY SYMPTOMS IN SCA 2233

Spinocerebellar ataxia (SCA) comprises a group of We assessed early symptoms other than gait ataxia
neurodegenerative multisystem disorders, which present in the most frequent subtypes of SCA. To optimize
with progressive ataxia as their key feature. It is caused accuracy of specifications, we combined information
by mutations in more than 25 genes of which 14 have from both patients and close relatives in the assessment
been cloned so far.1–3 The expansion of a CAG trinucleo- of onset of gait ataxia. Correlations of repeat length
tide repeat in the coding region of the respective gene and disease onset were calculated for (1) occurrence of
causes the disease in seven subtypes including the most the first disease related symptom and (2) the onset of
prevalent genotypes in Europe (SCA1, SCA2, SCA3, gait disturbance.
and SCA6). SCA is a phenotypically heterogeneous in-
sidious disease characterized by slowly progressive gait
ataxia and variable additional symptoms including visual PATIENTS AND METHODS
problems, dysarthria, dysphagia, limb ataxia, spasticity, Patients were recruited in the clinical network of
Parkinsonism, dystonia, peripheral neuropathy, restless EUROSCA, an international consortium funded by the
legs syndrome, and urge incontinence. Because of its he- European Union for clinical and basic research in SCA
reditary nature, the pathogenic process is likely to start (http://www.eurosca.org). Two hundred and eighty-
early in life or even prior to birth, but early development seven patients were recruited in specialized ataxia clinics
is generally normal and often patients remain clinically in Bochum (Germany), Bonn (Germany), Brussels (Bel-
healthy far beyond the second, third, or even seventh gium), Essen (Germany), Innsbruck (Austria), Milan
decade of their life. The exact disease onset, however, of- (Italy), Naples (Italy), Nijmegen (The Netherlands),
ten remains unclear and has only rarely been well defined Paris (France), Pec (Hungary), Tübingen (Germany),
in previous studies. Determination of disease onset might and Warsaw (Poland) including 78 patients with SCA1,
differ whether patients are asked for onset of gait ataxia 97 patients with SCA2, 62 with SCA3, and 50 with
or, alternatively, for onset of any other kind of behavioral SCA6. Data on age, disease severity as assessed by the
or neurological problem. Moreover, it might vary scale for the assessment and rating of ataxia (SARA14),
depending on the person that is asked—either the patient and CAG repeat length are given in Table 1. Addition-
or his/her relatives. Despite lack of standardization in the ally, 122 age- and sex-matched control subjects without
assessment of age at onset, significant correlations have a history of neurological disease were interviewed for
been found between the age of onset and the number of symptoms that may occur early in SCA. All patients and
the CAG motifs in the expanded allele in all of the more controls gave their written informed consent prior to
common SCA subtypes (SCA1, SCA2, SCA3, and inclusion. The study was approved by the Ethics Com-
SCA6) and repeat length can account for 50 to 80% of mittees of the participating centers.
variability in age of onset,4–10 it still remains unknown to Structured interviews were performed with patients
what extend differences in the ascertainment of disease and their close relatives concerning the year of onset
onset contribute to the unexplained portions of variabi- of permanent gait disturbance, double vision, reduced
lity. Apart from CAG repeat expansions, alternative visual acuity, dysarthria, frequent throat clearing
genetic or environmental factors influencing age of onset reflecting early dysphagia, problems with hand writing,
in SCA have rarely been identified11–13—although they episodic vertigo, neuropathic symptoms like weakness
are of major interest due to their likely function as modi- or sensory complaints, cramps, restless legs syndrome,
fiers of disease progression. sleep disturbances, or urinary urgency (Appendix).

TABLE 1. Biographic, genetic, and clinical data of patients included in this study
Whole cohort N 5 287 SCA1 N 5 78 SCA2 N 5 97 SCA3 N 5 62 SCA6 N 5 50
Age at examination (yr) 50.0 6 14.0 (18–84) 45.6 6 12.4 (25–76) 47.2 6 13.7 (18–84) 48.8 6 12.5 (24–72) 63.6 6 10.2 (38–83)
Age at onset of gait 39 6 13 (7–77) 37 6 11 (16–66) 35 6 12 (7–66) 38 6 11 (18–60) 54 6 11 (34–77)
ataxia (yr)
Disease duration (yr) 10.1 6 6.1 (1–33) 8.8 6 5.2 (1–22) 11.1 6 6.4 (1–30) 10.4 6 6.0 (1–25) 9.9 6 6.5 (1–33)
Sex (male/female) 150/137 49/29 42/55 32/30 27/23
SARA 14.9 6 7.9 (0–40) 15.8 6 9.4 (2–40) 15.6 6 7.5 (2–36) 13.9 6 8.1 (0–35.5) 13.6 6 5.8 (1–31)
CAG expanded NA 47 6 5 (39–62) 39 6 3 (33–47) 68 6 4 (56–75) 22 6 1 (22–28)
CAG normal NA 30 6 2 (27–36) 22 6 2 (20–33) 21 6 5 (14–34) 13 6 1 (8–14)

Disease duration is given as years with gait disturbance as fixed by the patients after discussion with their relatives. Higher SARA sum scores
indicate more severe disease.14 Data are presented as mean 6 standard deviation and range (in parenthesis).
NA, not applicable.

Movement Disorders, Vol. 23, No. 15, 2008

2234 C. GLOBAS ET AL.

Additionally, we asked for other early symptoms that relatives for 1 year or more. In 8.5% of patient-relative
may be related to SCA from the patient’s point of pairs these differences exceeded 5 years. SCA2
view. In case of incongruent information, we asked patients gave a later year of onset compared to their
patients to discuss these differences with their relatives. relatives (1.3 years 6 4.8), whereas SCA3 patients
Ultimately, the year of onset of symptoms was fixed recalled earlier onset than relatives (21.1 years 6 2.7;
by the patients. SCA2 and SCA3: ANOVA with Tukey Kramer adjust-
CAG repeat length was analyzed in DNA extracted ment for pairwise comparisons, P < 0.002). In their
from EDTA blood samples. DNA was available from final decision, 17.4% of patients changed their appraisal
259 patients (SCA1: 73, SCA2: 87, SCA3: 53, SCA6: about onset of permanent gait difficulties in favor of
46). To optimize comparability of repeat lengths, all their relatives’ estimation. Differences between patients
analyses were performed in the same lab (Human and care givers increased with longstanding disease
Genetics, Tübingen). A multiplex PCR assay (described (r 5 0.33, P < 0.0001; Pearson’s correlation co-
in Ref. 15) was further optimized for robust amplifica- efficient) but not with disease severity as assessed by
tion of all SCA mutations in one PCR assay: Genomic SARA (r 5 0.05, P 5 0.46).
DNA of 250 to 500 ng was used per PCR reaction. Gait difficulty was reported as the initial symptom
Primer sequences, PCR conditions, and details of frag- in two-thirds (66%) of all SCA patients. Symptoms
ment analysis are provided on request. As CAG repeats preceding gait ataxia were in the order of frequency
do not perfectly result in a 3-bp spacing, expected frag- cramps (9%), dysarthria (5%), sleep disturbance (5%),
ment lengths were compared with known (sequenced) double vision (4%), problems with hand writing (4%),
genotype standards and the allele calling was adapted episodic vertigo (4%), neuropathic symptoms like
accordingly. weakness or sensory complaints (3%), restless legs
syndrome (3%), urinary urgency (3%), reduced visual
acuity (2%), frequent throat clearing suggesting begin-
Statistics
ning dysphagia (1%), and other symptoms preceding
To test whether the reported onset of permanent gait gait disturbance (1%). In comparison to the control
disturbance varies between patients and relatives, a group, only double vision, dysarthria, problems with
paired Student’s t-test was performed. Correlations hand writing, and episodic vertigo occurred more fre-
between quantitative variables were assessed using quently in SCA patients (Table 2). Restricted to these
Pearson’s correlation coefficient. Differences for quan- items, 12% of SCA1, 13% of SCA2, 15% of SCA3,
titative variables between genotypes were compared and 24% of SCA6 patients started the disease with
using ANOVA with pairwise comparisons after a sig- other symptoms but gait ataxia. Episodic vertigo was
nificant global ANOVA (Tukey-Kramer correction of especially frequent as initial symptom in SCA6
P-value) except for the delay between early symptom patients when compared with other SCA subtypes
and gait ataxia that were compared using a Kruskal– (12.2% vs. 1.6–2.6%; P < 0.03; Fisher’s Exact Test).
Wallis test. Frequency of symptoms in patients and Frequency of other early signs did not differ between
controls are compared using a Fisher’s Exact Test with SCA genotypes.
adjustment for age (year of birth) and sex (logistic Double vision occurred up to 8 years before gait dis-
regression). Correlations between repeat length and turbance in SCA1, up to 23 years in SCA2, 16 years in
other variables were performed after exclusion of SCA3, and 35 years in SCA6 (median 6 years). Epi-
patients with extreme numbers of CAG repeats defined sodic vertigo preceded gait ataxia by up to 10 years in
as follows: expanded alleles SCA6 ‡ 28 CAG (1 SCA1, 1 year in SCA2, 2 years in SCA3, and 15 years
patient); normal alleles SCA1 ‡ 33 CAG (3 patients), in SCA6 (median 4 years). Four percent of SCA
SCA2 ‡ 27 CAG (2 patients), SCA3 ‡ 33 CAG (2 patients experienced these symptoms more than 5 years
patients), and SCA6  8 CAG (1 patient). All tests before onset of gait ataxia (5% in SCA1, 3% in SCA2,
were performed two-sided. P < 0.05 was considered 2% in SCA3, and 8% in SCA6). Dysarthria and prob-
significant. Statistical analyses were performed using lems with hand writing were reported not more than
SAS 9.1 (SAS Institute, Cary, NC). 5 years before onset of gait ataxia.
CAG repeat length in the expanded allele was re-
sponsible for about 60% of variability in age at onset
RESULTS of gait ataxia in SCA1 and SCA2, for about 25% in
Data about onset of permanent gait disturbance var- SCA3, and about 20% in SCA6. No major differences
ied frequently (44%) between SCA patients and their in correlation with repeat length were observed when

Movement Disorders, Vol. 23, No. 15, 2008

 EARLY SYMPTOMS IN SCA 2235

TABLE 2. Symptoms preceding gait abnormalities in SCA

Co All SCA SCA1 SCA2 SCA3 SCA6
,a
Double vision – 4.3 1.3 3.2 10.0** 4.2
Reduced visual acuity 4.1 2.1 5.1 1 – 2
Dysarthria – 4.7 2.6 5.6* 3.3 8.2*
Frequent throat clearing – 1.4 2.6 1.1 1.6 –
Problems with hand writing – 4 5.3 5.6* 1.7 2.1
Episodic vertigo 1.6 3.9 2.6 2.1 1.6 12.2*
Neuropathic symptoms 4.1 2.9 4 – 4.8 4.1
Cramps 8.2 8.9 11.5 9.9 4.9 8
Restless legs syndrome 6.3 2.8 5.3 2.1 1.6 2
Sleep disturbances 9.0 4.7 2.6 2.2* 5 12.5
Urinary urgency 1.6 2.8 3.9 3.2 1.6 2
Other preceding symptoms 0.7 1.3 – 1.7 –

Proportion of patients (%) who reported onset of the respective symptom prior to gait ataxia.
*P < 0.05; **P < 0.01 (comparison of SCA vs. control group; Fisher’s Exact Test after adjustment for age and sex).
a
P < 0.05 (comparison to SCA1, 2, and 3).

information about age of onset was provided by differences in number of CAG repeats were found con-
patients or by their relatives or after discussion of both cerning first symptoms in any SCA genotype (Wil-
(Table 3). Under the hypothesis that long disease dura- coxon two-sample test).
tion may hamper accuracy of memory concerning
onset of the disease, we repeated correlations after
exclusion of patients with the disease for more than 10
years. This improved correlations especially in SCA3 DISCUSSION
and SCA6 (Table 3, Fig. 1A–D). Exclusion of patients This systematic study on disease onset in SCA dem-
with more severe disease (SARA sum score above 20 onstrated that gait ataxia is the initial complaint in
points14) had a similar effect. Consideration of nongait only two-thirds of patients. When rather unspecific
symptoms did not further improve the predictive value symptoms like cramps, restless legs, and sleep disturb-
of CAG repeat for age of onset (Table 3). No effect of ance were excluded still 16% of SCA patients report
normal alleles on age at onset was found. other problems than gait as the earliest symptom. Espe-
We tested the hypothesis that initial symptoms of cially, diplopia and episodic vertigo but also dysrthria
SCA may be determined by repeat length. No group and clumsiness occurred prior to onset of gait ataxia.
None of the early clinical signs had specificity for a
certain SCA subtype but episodic vertigo was more
common in SCA6. This reflects the close relationship
TABLE 3. Correlation of CAG repeat length
and age of onset
of SCA6 and episodic ataxia type 2 (EA2), both of
which are caused by mutations in the a1A-subunit
SCA1 SCA2 SCA3 SCA6 of the voltage-gated neuronal calcium channel.16
Age of onset defined by (%) (%) (%) (%)
Although patients with EA2 frequently develop cere-
Patients onlya 57*** 59*** 27*** 19** bellar atrophy on MRI and permanent gait ataxia after
Relatives onlya 51*** 61*** 24*** 17*
Patients 1 relativesa 57*** 60*** 26*** 20** longstanding disease, episodic ataxia has rarely been
Patients 1 relatives 64*** 67*** 46*** 41*** reported in patients with SCA6.17–21
1 duration < 10 yra Detection of early signs is of major importance for
Patients 1 relatives 53*** 67*** 47*** 21**
1 SARA < 20a consecutive treatment management. If disease modulat-
Occurrence of diplopia, 55*** 54*** 25** 2 ing compounds become available, they may be most
dysarthria, clumsiness, efficient when introduced to patients in early stages of
episodic vertigo, or gait
disturbance the disease. On the other hand, treatment before dis-
ease onset may not be advisable in drugs with potential
Percentages to which repeat length determined age of onset are side effects. Thus, reliable diagnosis of disease onset is
given for different definitions of disease onset.
*P < 0.05; **P < 0.01; ***P < 0.001. a major challenge and recognition of nonataxia symp-
a
Onset of progressive gait ataxia. toms might be a promising approach.

Movement Disorders, Vol. 23, No. 15, 2008

2236 C. GLOBAS ET AL.

FIG. 1. Linear regression analysis of age at onset and length of expanded CAG repeat. Onset has been defined as beginning of progressive gait
ataxia. Patients with disease durations of more than 10 years have been excluded.

Our results suggest that first symptoms may occur a of 5 to 20 years with mean differences between 1.2
decade or more before the onset of gait instability. The and 4.5 years depending on the genotype. The largest
list of symptoms analyzed in this study is by no means differences between patients and relatives were
complete but had to be restricted to problems that observed when disease duration exceeded 10 years.
could be remembered with acceptable accuracy also These findings underline the importance of standar-
after many years. In this respect, it is important to note dized assessment to yield most reliable results and
that not only patients in late stages of the disease stress the necessity of prospective studies of disease
described early symptoms many years before onset of onset in SCA. Why SCA3 patients recognized onset of
gait difficulties, but also patients with short disease du- gait ataxia about 1 year earlier than their relatives,
ration and low SARA scores indicating less advanced whereas SCA2 patients dated the beginning of gait
disease reported problems such as diplopia to precede difficulties 1 year later than their relatives remains
gait ataxia for up to 15 years. unclear. It would be reasonable to expect that patients
Assessment of disease onset was poorly standardized feel changes in gait stability before it becomes obvious
in former studies.4,5 In this study, the view of close from outside. Whether cognitive changes that are more
relatives concerning the onset of gait ataxia differed pronounced in SCA2 than in other subtypes22,23 con-
from the patient’s report in 17% of cases with a span tribute to the shift in recognition or memory in SCA2

Movement Disorders, Vol. 23, No. 15, 2008

 EARLY SYMPTOMS IN SCA 2237

has to be assessed in prospective studies including neu- from biography by asking for ataxia at major events
ropsychological testing. in former years (e.g., 50th birthday party). Addition-
Our data confirm that CAG repeat length can only par- ally, patient is confronted with onset data mentioned
tially explain variability in age of onset of SCA. Our in former records if available.
data revealed the closest correlation with CAG repeat c Close relatives are asked in the same manner. If no
length when (1) onset is defined by beginning of perma- relatives accompany the patient call spouses or chil-
nent gait disturbance and (2) patients with longstanding dren.
disease were excluded. Correlation did not improve fur- c Then, ask the patient to settle potential discrepancies
ther when nongait symptoms were taken into account. with the statement of his/her relatives for final deter-
Interviews with caregivers did not improve correlation of mination of age at onset.
onset age and CAG repeat length, although dates for c Afterward, ask the patient for onset of gait ataxia in
onset varied substantially between patients and relatives. parents, sibs, and children. List these statements.
Under all conditions, CAG repeat length explained less
than 50% of variability in age of onset in SCA3 and
Early Symptoms of SCA
SCA6. In SCA6, the influence of repeat length may be
masked by the rather uniform size of the expanded allele c After determination of onset of permanent gait insta-
(22 CAG in 74% of SCA6 patients). However, our data bility ask the patient and his/her relatives for other
show that onset variability in all SCA subtypes is driven symptoms of SCA that may have preceded the onset
by other genetic or environmental factors that remain to of gait ataxia. List spontaneous recall.
be identified. Given the enormous variability in onset c Then, offer a list of potential early symptoms of
data depending on the assessment strategy, prospective SCA like
studies with standardized evaluation procedures of disease
onset are necessary to identify disease modifiers that  Double vision
have minor effects than repeat length in expanded alleles.  Reduced visual acuity (If present, verify that
Recent progress in the understanding of disease mech- this is due to retinopathy or optic atrophy)
anisms in polyglutamine disorders and promising results  Dysarthria
in animal models24–26 offer chances for clinical trials of  Problems with hand writing
potentially disease modifying compounds in the near  Episodic vertigo
future. If such compounds aim to delay disease, a precise  Weakness or sensory complaints related to pe-
prediction of onset age is warranted. This may not be ripheral neuropathy
possible for individual patients but is feasible for larger  Restless legs syndrome
cohorts. To this end, recognition of early symptoms that  Sleep disturbance (specify if possible)
may develop before the onset of gait ataxia is mandatory  Urinary urgency or incontinence.
and will require prospective studies.
c Ask for the year of first occurrence of symptoms
Acknowledgments: This study has been funded by the Eu- that preceded gait ataxia.
ropean Union by the EUROSCA grant (LSHM-CT-2004-
503304). We thank all the patients for their kind cooperation.

REFERENCES
1. Schols L, Bauer P, Schmidt T, Schulte T, Riess O. Autosomal
dominant cerebellar ataxias: clinical features, genetics, and
APPENDIX pathogenesis. Lancet Neurol 2004;3:291–304.
2. Waters MF, Minassian NA, Stevanin G, et al. Mutations in volt-
Instructions for the Structured Interview of Early age-gated potassium channel KCNC3 cause degenerative and de-
Symptoms in SCA velopmental central nervous system phenotypes. Nat Genet
2006;38:447–451.
3. Ikeda Y, Dick KA, Weatherspoon MR, et al. Spectrin mutations
Definition of Age at Onset cause spinocerebellar ataxia type 5. Nat Genet 2006;38:184–190.
4. van de Warrenburg BP, Sinke RJ, Verschuuren-Bemelmans CC,
et al. Spinocerebellar ataxias in the Netherlands: prevalence
c Onset is defined as the beginning of permanent and and age at onset variance analysis. Neurology 2002;58:702–
progressive gait instability. 708.
5. van de Warrenburg BP, Hendriks H, Durr A, et al. Age at onset
c Patients are asked for the onset of permanent and variance analysis in spinocerebellar ataxias: a study in a Dutch-
progressive gait instability. Orientation is provided French cohort. Ann Neurol 2005;57:505–512.

Movement Disorders, Vol. 23, No. 15, 2008

2238 C. GLOBAS ET AL.

6. Orr HT, Chung MY, Banfi S, et al. Expansion of an unstable tri- tions in the Ca21 channel gene CACNL1A4. Cell 1996;87:
nucleotide CAG repeat in spinocerebellar ataxia type 1. Nat 543–552.
Genet 1993;4:221–226. 17. Denier C, Ducros A, Vahedi K, et al. High prevalence of CAC-
7. Stevanin G, Durr A, Brice A. Clinical and molecular advances in NA1A truncations and broader clinical spectrum in episodic
autosomal dominant cerebellar ataxias: from genotype to pheno- ataxia type 2. Neurology 1999;52:1816–1821.
type and physiopathology. Eur J Hum Genet 2000;8:4–18. 18. Baloh RW, Yue Q, Furman JM, Nelson SF. Familial episodic
8. Schols L, Amoiridis G, Buttner T, Przuntek H, Epplen JT, Riess ataxia: clinical heterogeneity in four families linked to chromo-
O. Autosomal dominant cerebellar ataxia: phenotypic differen- some 19p. Ann Neurol 1997;41:8–16.
ces in genetically defined subtypes? Ann Neurol 1997;42:924– 19. Yabe I, Sasaki H, Yamashita I, et al. Initial symptoms and mode
932. of neurological progression in spinocerebellar ataxia type 6
9. Lorenzetti D, Bohlega S, Zoghbi HY. The expansion of the CAG (SCA6). Rinsho Shinkeigaku 1998;38:489–494.
repeat in ataxin-2 is a frequent cause of autosomal dominant 20. Jodice C, Mantuano E, Veneziano L, et al. Episodic ataxia type
spinocerebellar ataxia. Neurology 1997;49:1009–1013. 2 (EA2) and spinocerebellar ataxia type 6 (SCA6) due to CAG
10. Yabe I, Sasaki H, Matsuura T, et al. SCA6 mutation analysis in repeat expansion in the CACNA1A gene on chromosome 19p.
a large cohort of the Japanese patients with late-onset pure cere- Hum Mol Genet 1997;6:1973–1978.
bellar ataxia. J Neurol Sci 1998;156:89–95. 21. Koh SH, Kim HT, Kim SH, Lee GY, Kim J, Kim MH. Spinocer-
11. Pulst SM, Santos N, Wang D, et al. Spinocerebellar ataxia ebellar ataxia type 6 and episodic ataxia type 2 in a Korean fam-
type 2: polyQ repeat variation in the CACNA1A calcium channel ily. J Korean Med Sci 2001;16:809–813.
modifies age of onset. Brain 2005;128 (Part 10):2297–2303. 22. Globas C, Bosch S, Zuhlke C, Daum I, Dichgans J, Burk K. The
12. Durr A, Stevanin G, Cancel G, et al. Spinocerebellar ataxia 3 cerebellum and cognition. Intellectual function in spinocerebellar
and Machado-Joseph disease: clinical, molecular, and neuropath- ataxia type 6 (SCA6). J Neurol 2003;250:1482–1487.
ological features. Ann Neurol 1996;39:490–499. 23. Burk K, Globas C, Bosch S, et al. Cognitive deficits in spinocer-
13. Kubis N, Durr A, Gugenheim M, et al. Polyneuropathy in auto- ebellar ataxia type 1, 2, and 3. J Neurol 2003;250:207–211.
somal dominant cerebellar ataxias: phenotype-genotype correla- 24. Sarkar S, Perlstein EO, Imarisio S, et al. Small molecules
tion. Muscle Nerve 1999;22:712–717. enhance autophagy and reduce toxicity in Huntington’s disease
14. Schmitz-Hubsch T, du Montcel ST, Baliko L, et al. Scale for the models. Nat Chem Biol 2007;3:331–338.
assessment and rating of ataxia: development of a new clinical 25. Sarkar S, Davies JE, Huang Z, Tunnacliffe A, Rubinsztein DC.
scale. Neurology 2006;66:1717–1720. Trehalose, a novel mTOR-independent autophagy enhancer,
15. Konieczny M, Bauer P, Tomiuk J, et al. CAG repeats in Restless accelerates the clearance of mutant huntingtin and a-synuclein.
Legs syndrome. Am J Med Genet B Neuropsychiatr Genet J Biol Chem 2007;282:5641–5652.
2006;141:173–176. 26. Watase K, Gatchel JR, Sun Y, et al. Lithium therapy improves neu-
16. Ophoff RA, Terwindt GM, Vergouwe MN, et al. Familial hemi- rological function and hippocampal dendritic arborization in a spino-
plegic migraine and episodic ataxia type-2 are caused by muta- cerebellar ataxia type 1 mouse model. PLoS Med 2007;4:e182.

Movement Disorders, Vol. 23, No. 15, 2008