Clin Genet 2007
Printed in Singapore. All rights reserved
#2007 The Authors
Journal compilation#2007 Blackwell Munksgaard
Survival estimates for patients with
Machado–Joseph disease (SCA3)
Kieling C, Prestes PR, Saraiva-Pereira ML, Jardim LB. Survival
estimates for patients with Machado–Joseph disease (SCA3).
Clin Genet 2007.#Blackwell Munksgaard, 2007
Machado–Joseph disease (MJD), one of the most prevalent autosomal
dominant cerebellar ataxias, is a neurodegenerative disease that starts
during adulthood, with patients showing difficulties in gait, later
becoming bedridden, and ultimately presenting premature death. There
is, however, scarce data quantifying disease impact on patient survival.
We investigated the overall survival of a large series of MJD patients and
compared it with the survival of their asymptomatic relatives. A total of
412 affected and 413 unaffected individuals were ascertained from
a consecutive sample of 82 families with a molecular diagnosis of MJD.
Estimated mean survival time was 63.96 years [95% confidence interval
(CI), 62.09–65.83] for the affected group and 78.61 years (95% CI, 74.75–
82.47) for the unaffected group (p , 0.001). For a subset of 366 patients,
mean age at onset was 36.37 years (95% CI, 35.21–37.53) and survival
after disease onset was estimated as 21.18 years. Early onset and
large CAG length predicted shorter overall survival times. This study
presents quantitative data on the impact of MJD on overall survival,
a phenomenon that is related to CAG length, age at onset, and year of
C Kielinga, PR Prestesb,
and LB Jardima,d
aMedical Genetics Service, Hospital de
Clinicas de Porto Alegre, Porto Alegre,
Brazil,bGenetics and Molecular Biology
anddInternal Medicine Department,
Universidade Federal do Rio Grande do
Sul, Porto Alegre, Rio Grande do Sul,
Key words: Machado–Joseph
disease – polyglutamine diseases –
SCA3 – spinocerebellar ataxia – survival
Corresponding author: Laura Bannach
Jardim, Medical Genetics Service,
Hospital de Clinicas de Porto Alegre,
Rua Ramiro Barcelos, 2350, 90035-903
Porto Alegre, Brazil.
Tel.: 155 51 2101 8011;
fax: 155 51 2101 8010;
Received 30 May 2007, revised and
accepted for publication 24 August 2007
Machado–Joseph disease (MJD, also known as
spinocerebellar ataxia type 3, or SCA3) is one of
the most prevalent autosomal dominant cerebel-
lar ataxias (1). The condition is related to a CAG
expansionat MJD1 locusin chromosome14q32.1
disease range is from 62 to 86 (1)] and is
process. Clinical manifestations usually start
during adulthood, with patients presenting diffi-
and later becoming bedridden (3). Notwithstand-
there is scarce data quantifying disease impact on
patient survival. We therefore investigated the
overall survival of a large series of MJD patients
and compared it with the survival of their
The present study was approved by the ethics
committee of our institution (Hospital de Clinicas
de Porto Alegre). After consent, an interview was
or older whose families had a molecular diagnosis
of MJD. All cases belonged to MJD families
alreadyregistered inour hospital.Eachinformant
was personally interviewed and provided data on
his/her relatives (children, siblings, parents,
uncles, cousins, and grandparents of the affected
lineage) by following a structured questionnaire
about each family member. Data were double-
checked by a second (or a third, and so on)
informant of the same family; only concordant
informationwasincluded. RioGrandedoSul, the
origin of all families (4). All studied cases were
carriers of the A-C-A haplotype (5).
Individuals were classified into two groups as
follows: MJD affected cases, if they were molec-
ularly diagnosed or if they were reported as
symptomatic by the informants, and unaffected
individuals, if no disease symptoms were reported
by the informants. Only healthy brothers and
sisters of affected individuals could be recruited in
the unaffected group. Variables, such as year of
birth, gender, disease onset, and CAGn (in the
affected group), were also determined.
Survival was estimated through the Kaplan–
Meier product-limit method and compared using
a log-rank test. Cox regression model was
employed for inclusion of covariates and calcula-
tion of hazard ratios (HR). Proportional hazard
survival curve and Schoenberg residuals for
individual covariates against time.
participate in this study. Reliable information on
birth and survival was available for 825 individ-
uals from 82 families; all individuals were born
between 1903 and 1996. There were 142 deaths
among the 412 individuals presenting MJD and
15 events among their 413 unaffected relatives.
Estimated mean survival time was 63.96 years
[95% confidence interval (CI), 62.09–65.83] for
the affected group and 78.61 years (95% CI,
Compared with the unaffected individuals, pa-
(HR, 4.92; 95% CI, 2.86–8.46; Fig. 1).
in the entire population. Stratification revealed
(95% CI, 1.02–1.05) only in the affected group,
towards protection. For a subset of 105 patients
with information regarding number of CAG
repeats, tract length also influenced survival, each
extra repeat representing an HR of 1.80 (95% CI,
1.22–2.65). Gender did not significantly alter
results and was not included in the final model.
Age at disease onset was available for 366
individuals and was estimated as 36.37 years
(95% CI, 35.21–37.53). As we expected, early
after disease onset was calculated for each
individual, and the mean period between onset
of symptoms and death was 21.18 years (95% CI,
19.85–22.51). Although year of birth and age at
disease onset were not significant predictors of
survival time after disease onset (p ¼ 0.288 and
0.130, respectively), the number of CAG repeats
remained a risk factor for disease survival time
(HR, 1.35; 95% CI, 1.09–1.68).
This study brings further evidence of a reduced
survival in MJD, as well as on the role of CAG
with MJD. To our knowledge, no comparison
between life expectancy in MJD patients and in
unaffected relatives has ever been published. This
quantitative data measure the impact of MJD on
overall survival and survival after disease onset,
investigating the influence of CAG length, age at
onset, and year of birth.
Previous studies have estimated mean survival
time after disease onset in series of cases that
included only deceased patients, with results
ranging from 14 to 17 years (3, 6, 7). However,
these figures overestimate the effect of MJD as
they ignore the contribution of living patients to
an increased survival time. Klockgether et al. (8)
performed a sophisticated graphic analysis of the
natural history of degenerative ataxias and
reported survivaltimes afterdisease onset ranging
and SCA3. Unfortunately, the authors did not
Fig. 1. Kaplan–Meier overall survival curves by affection
status. Survival probabilities were significantly different
between individuals affected by Machado–Joseph disease
and their unaffected relatives (log-rank test, p , 0.001).
Kieling et al.
provide objective estimates for survival times in Download full-text
MJD. For a sample of 110 MJD patients, they
was a risk factor for disease progression (no
whereas gender did not have any significant
influence on progression or survival.
A comparison of survival between disease and
control groups has evident advantages because it
truly evaluates the impact of the disease in a given
community. The present strategy of using unaf-
fected siblings as a comparison group aimed at
reducing the interference of confounding varia-
bles while measuring the magnitude of the disease
impact. Also, external validity was corroborated
survival time and the life expectancy of the MJD
individuals’ contemporaries in the general pop-
ulation [76.59 years of age according to the latest
Brazilian census (9)].
We acknowledge that among asymptomatic
carriers of the MJD mutation. These individuals
therefore, potentially increase the survival esti-
mates of the affected group. To assess this
limitation, we performed an additional analysis
including only individuals older than 45 years,
a subgroup in which the proportion of pre-
symptomatic carriers would be reduced. Results
were similar to those obtained with the entire
sample (data not shown).
The risk imparted to MJD-affected individuals
by their year of birth deserves some comments.
Whereas each additional year exhibited a trend
towards protection in the control group, it
increased the HR of death in the affected group.
This finding probably reflects the lack of specific
treatment for MJD and the large time span
included in this study, in which the anticipation
phenomenon could be operating, that is, recently
born individuals would present the disease earlier
and, thus, have shorter survival times. Interest-
ingly,neither year of birthnorage atdisease onset
was associated with different survival times after
disease onset, which indicates that these variables
could be reducing overall survival time mostly by
reducing disease-free survival time. This was not
the case regarding CAG repeats (which are risk
factors for reduced overall survival and survival
after disease onset).
In conclusion, MJD conferred an almost five-
fold increase in the risk of death to the affected
individuals compared with their unaffected rela-
tives, and this effect was modulated by CAG
length, year of birth and age of symptoms onset.
Webelieve that,accordingtotheresults presented
herein, estimates of the magnitude of premature
death among MJD patients will assist clinicians in
counseling patients regarding the nature of their
disease and will guide researchers in the evalua-
tion of the long-term impact of future therapies.
We are grateful to the patients and their relatives who agreed to
participate in this project. We also thank Professor Jorge
Sequeiros and Isabel Silveira for the molecular studies of the
first families with MJD. This study was partially supported by
CAPES, through a scholarship granted to P. R. P., and by
FAPERGS. Professor L. B. J. is supported by CNPq, Brazil.
1. Schols L, Bauer P, Schmidt T et al. Autosomal dominant
cerebellar ataxias: clinical features, genetics, and pathogenesis.
Lancet Neurol 2004: 3: 291–304.
2. Kawaguchi Y, Okamoto T, Taniwaki M et al. CAG
expansions in a novel gene for Machado-Joseph disease at
chromosome 14q32.1. Nat Genet 1994: 8: 221–228.
3. Sequeiros J, Coutinho P. Epidemiology and clinical aspects
of Machado-Joseph disease. In: Harding A, Deufel T,
Chamberlain S, eds. Hereditary ataxias. New York: Raven
Press, 1993: 139–153.
4. Trott A, Jardim LB, Ludwig HT et al. Spinocerebellar
ataxias in 114 Brazilian families: clinical and molecular
findings. Clin Genet 2006: 70: 173–176.
5. Bressel T. [Analysis of intragenic haplotypes of MJD1 gene in
Machado-Joseph disease patients]. [MSc Dissertation] [Portu-
guese]. Porto Alegre: Genetics and Molecular Biology Depart-
ment Universidade Federal do Rio Grande do Sul, 2003.
6. Barbeau A, Roy M, Cunha L et al. The natural history of
Machado–Joseph disease. Can J Neurol Sci 1984: 11: 510–525.
7. Jardim LB, Pereira ML, Silveira I, Ferro A, Sequeiros J,
Giugliani R. Machado-Joseph disease in South Brazil:
clinical and molecular characterization of kindreds. Acta
Neurol Scand 2001: 104: 224–231.
8. Klockgether T, Ludtke R, Kramer B et al. The natural
history of degenerative ataxia: a retrospective study in 466
patients. Brain 1998: 121: 589–600.
9. The 2000 Brazilian demographic census. IBGE. Retrieved
25 May 2007, from http://www.ibge.gov.br/home/estatistica/
Survival in MJD