|
Correction
to
Rosenberg,
|
| |||||||||||||||||||||||||||||||||||||||||
For many years the diagnosis of Friedreich's ataxia has been based on clinical criteria. According to these criteria, 103 of their patients had typical Friedreich's ataxia, whereas 74 did not have at least one essential feature and are described as having atypical Friedreich's ataxia. We are concerned that it is not clear how many of the patients in each group were included in the analysis that led to the identification of mutations in X25 as the cause of Friedreich's ataxia.2 Furthermore, we doubt the pathogenetic relevance of the GAA-repeat expansion because 46 percent of the patients with atypical Friedreich's ataxia were homozygous for the repeat and 24 percent of those homozygous for the repeat expansion had atypical Friedreich's ataxia. It is inconsistent of the authors to raise the possibility that point mutations in one allele may have pathogenetic effects (when paired with an allele with a large GAA expansion) while stating that "38 patients had two alleles with GAA repeats in the normal size range, which excludes the diagnosis of Friedreich's ataxia." The pathogenetic relevance of the GAA repeat is further called into question by the report of Carvajal and colleagues,3 who found that the X25 gene, with its intronic GAA repeat, constitutes the 3' end of a much larger gene, STM7, which encodes a protein with 1-phosphatidylinositol-4-phosphate 5-kinase activity; many of the splice variants terminate 5' of the GAA repeat.
We believe that there is no justification for extending the definition of Friedreich's ataxia on the basis of the nature of the GAA-repeat expansion. In particular, we do not believe that "direct molecular diagnosis through determination of the size of the GAA expansion should become an essential tool in clinical practice" until this genetic change is proved beyond a reasonable doubt to be the pathogenetic abnormality that causes Friedreich's ataxia. As the authors admit, "more than 30 percent of the [phenotypic] variance results from unknown factors, so that the predictive value of expansion size is limited."
David P. Dutka, M.D.
Derek J.R. Nunez, M.D.
Hammersmith Hospital
London W12 0NN, United Kingdom
References
Under these conditions the appropriate method of analysis is multiple regression. We used it in our study and found that the contribution of the larger allele was not significant. Thus, according to this analysis, unknown factors account for approximately 50 percent of the variability in the age at onset. A large, worldwide study that includes all groups involved in the study of Friedreich's ataxia is needed to clarify the role of the larger GAA expansion of each pair of alleles and of other potential genetic and environmental factors in determining the phenotype.
Alessandro Filla, M.D.
Giuseppe De Michele, M.D.
Sergio Cocozza, M.D.
University of Naples Federico II
1-80131 Naples, Italy
References
Hiten D. Madhani, M.D., Ph.D.
Whitehead Institute for Biomedical Research
Cambridge, MA 02142-1479
References
To the Editor: Dutka and Nunez doubt that the GAA-repeat expansion has a causal role in Friedreich's ataxia because 46 percent of the patients with atypical Friedreich's ataxia in our study were homozygous for the expansion and because Carvajal et al.1 suggested that the frataxin gene constitutes the 3' end of the STM7 gene. As in the study in which the gene for Friedreich's ataxia was initially identified,2 only patients with typical Friedreich's ataxia were included. Ninety-four percent were homozygous for the GAA expansion; the remaining were compound heterozygotes with an expansion of one allele and a point mutation on the other. No expansions were found, however, in 98 normal controls. These results strongly support a causal role of the GAA expansion in Friedreich's ataxia. Consequently, we consider that all patients with GAA expansions on both alleles have the same disease, even though there is clinical heterogeneity. In addition, since allelic heterogeneity is well documented in many genetic disorders, particularly in diseases caused by trinucleotide-repeat expansions, the molecular diagnosis of Friedreich's ataxia in patients with GAA expansions on one allele can also be established if a point mutation is detected on the other allele. In contrast, only 0.1 percent of patients with Friedreich's ataxia can be expected to carry two point mutations (one on each allele), since point mutations in the frataxin gene are found in 3 percent of the mutant chromosomes.2 Therefore, the 38 patients with clinically atypical disease who did not have GAA expansions were very unlikely to have point mutations on both alleles of the frataxin gene.
The discovery of transcripts that link the frataxin and STM7 genes does not challenge the causal role of the GAA expansion in Friedreich's ataxia. Point mutations have been found in the frataxin part of the molecule2,3 (and unpublished data) but not in the STM7 moiety,1 indicating that only transcripts corresponding to frataxin alone or to the putative STM7–frataxin fusion protein are relevant to the disease. In addition, the importance of these rare fusion transcripts in Friedreich's ataxia is highly questionable.3
Our results, which were corroborated by two independent studies,4,5 show that the clinical spectrum of Friedreich's ataxia extends beyond that covered by the usual clinical criteria, and that testing for the GAA expansions in the frataxin gene is important in clinical practice. Although the size of the expansion is only of limited prognostic value, this simple molecular analysis allows an early and accurate diagnosis to be made without the need for additional investigations and is of major importance for genetic counseling.
We agree with the comments of Filla and colleagues.
Alexandra Dürr, M.D.
Alexis Brice, M.D.
INSERM Unité 289
75651 Paris Cedex 13, France
Michel Koenig, M.D., Ph.D.
Institut de Génétique et de Biologie Moléculaire et Cellulaire
67404 Illkirch CEDEX, Strasbourg, France
References
Roger N. Rosenberg, M.D.
University of Texas Southwestern Medical Center
Dallas, TX 75235-9036
| |||||||||||||||||||||||||||||||||||||||||
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | TERMS OF USE | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2009 Massachusetts Medical Society. All rights reserved. |
Previous
