Abstract
Objective: To review the direct DNA testing for Huntington’s disease (HD) in Germany, Switzerland, and Austria from 1993 to 1997, and to analyze the population with regard to age structure, gender, and family history.
Methods: Twelve laboratories (nine in Germany, two in Austria, and one in Switzerland) recorded data pertaining to repeat number, gender, age at molecular diagnosis, and family history of probands. The molecular test was categorized as either diagnostic (for symptomatic individuals), presymptomatic (for individuals at risk), and prenatal (for pregnancies at risk).
Results: A total of 3,090 HD patients, 992 individuals at risk, and 24 fetuses were investigated using DNA analysis. The clinical diagnosis was confirmed in 65.6% of patients. A total of 38.5% of individuals at risk inherited an expanded CAG repeat. The female-to-male ratio showed a distinct predominance of women both in the diagnostic and presymptomatic groups. Of the fetuses tested, six were carriers of an expanded CAG repeat. Two pregnancies were interrupted; one pregnancy was not. No information about the parents’ decision was obtained for the remaining three pregnancies.
Conclusions: Approximately 20% of the estimated 10,000 HD patients living in Germany, Switzerland, and Austria have been identified by DNA analysis (total population, approximately 100 million; incidence of HD, 1:10,000). Assuming a ratio of HD patients to individuals at risk of 1:3, approximately 30,000 individuals are, in principle, eligible for a presymptomatic test. Less than 3 to 4% of individuals at risk have requested a presymptomatic test. This shows that the assumed enormous request of predictive testing has not occurred. More surprisingly, prenatal diagnoses were found to be rare.
The discovery of the Huntington’s disease (HD) gene (IT15) and of the underlying mutation in 1993 led rapidly to the development of molecular tests capable of confirming the diagnosis in patients and identifying carriers among individuals at risk.1 To cope with the technical problems and ethical issues raised by the molecular testing, and to develop identical strategies of the different laboratories offering the test to the community, a consortium of clinicians, human geneticists, patient support groups, and all laboratories offering DNA analysis for HD was founded in Germany, Switzerland, and Austria in 1993.
HD belongs to the continuously expanding group of “CAG repeat disorders,” including at least eight neurodegenerative diseases.2 Ranges of nonexpanded and expanded CAG repeats have been defined in HD by molecular testing: Individuals carrying alleles with as many as 35 CAG repeats harbor no risk of developing HD, whereas individuals with 40 or more CAG repeats are affected or will develop HD during their lifetime. The range of 27 to 35 repeats is referred to as “mutable normal allele,” although there is no general agreement on the exact limits.3,4 Furthermore, alleles with 36 to 39 repeats have been found both in affected as well as in nonaffected individuals.5 Therefore, it has been suggested that these alleles have reduced penetrance and have to be considered carefully with respect to diagnostic significance and presymptomatic value.5 However, meiotic instability of alleles with 27 to 35 repeats is a widely accepted phenomenon, with consequences and implications for patients and individuals at risk that have been discussed recently.6 Despite these difficulties regarding the clinical and prognostic implications of alleles harboring 27 to 39 CAG repeats, molecular genetic testing has spread quickly and has proved to be a reliable diagnostic tool for neurologists, psychiatrists, and geneticists. The worldwide testing of more than 2,800 individuals7 has demonstrated the specificity and sensitivity of the test. In fact, CAG expansion in the IT15 gene has been found only in HD patients and not in patients affected with other neurologic or psychiatric conditions. We report the results of data collected retrospectively from the molecular testing of 3,090 affected patients, 992 individuals at risk, and 24 fetuses performed nationwide in 12 laboratories in Germany, Switzerland, and Austria.
Methods.
Twelve laboratories of the Consortium for DNA Analysis of HD provided their data from the molecular testing for HD conducted from 1993 to 1997. For each individual tested, the age at molecular diagnosis, CAG repeat number, gender, and family history were provided. Family history was defined as positive (FHP) if relatives with movement disorders, cognitive impairment, or psychiatric symptoms were known, and as family history uncertain (FHU) if relatives with movement or psychiatric conditions could not be excluded. Family history was considered negative (FHN) if there were no relatives with a neurodegenerative disease or psychiatric symptoms. Data were taken mostly from patients’ files and not from direct interviews. Molecular testing and test availability were carried out according to international guidelines.8
CAG repeat sizes were determined after PCR amplification of genomic DNA from peripheral white blood cells. In general, the CAG repeat was amplified by the method described by Warner et al.9 or Riess et al.10 with slight modifications. We considered 40 or more CAG repeats to be pathologic alleles, whereas alleles with 26 or less CAG repeats were considered to be not expanded. The range of 27 to 35 repeats comprises the mutable normal alleles, and alleles with 35 to 39 repeats were considered to have reduced penetrance. A retrospective distinction between data obtained with PCR, including the polymorphic CCG located at the 3′ end of the CAG repeat,10 and PCR amplification of the CAG repeat alone9 was not possible for all subjects retrospectively. However, all alleles in the range of 30 to 39 CAG repeats were reassessed in one laboratory (O.R.) excluding the flanking CCG repeats.9 In case of detection of only one allele with at least two independent PCR methods, Southern blot analysis was carried out to confirm homozygosity of the identified allele.
Results.
Diagnostic testing.
Data from the molecular analysis of 3,090 individuals with suspected HD were collected from 12 laboratories (2,825 individuals from Germany, 199 patients from Switzerland, and 66 patients from Austria). Testing was generally requested by neurologists, psychiatrists, or genetic counselors. The diagnosis of HD was confirmed in 65.6% of all patients referred for differential diagnosis. We found HD alleles with more than 39 CAG repeats in 2,028 individuals. A total of 971 individuals (31.4%) had alleles with less than 27 CAG repeats, 55 individuals (1.8%) had mutable normal alleles, and 36 individuals (1.2%) carried alleles in the reduced penetrance range (table 1). Only one allele was detected by PCR in 140 patients (4.5%) who were considered to be homozygous for the nonexpanded allele. The range of expansion was 40 to more than 100 CAG repeats with a mean of 45.1 CAG repeats (SD, ±4.6) and a median of 44 CAG repeats (figure 1). The nonexpanded alleles (<27 CAG repeats) in HD patients ranged from 4 to 26 CAG repeats with a mean of 20.2 ± 2.6 CAG repeats. The most frequent expansion in our cohort was 43 CAG repeats (15% of the expanded alleles), and the most common nonexpanded allele had 17 repeats (22.2% of all alleles that were not expanded). Thirty-six patients harbor an allele in the reduced penetrance range and 55 patients carry a mutable normal allele. Twenty-three HD patients with more than 39 CAG repeats had a second allele in the mutable normal allele/reduced penetrance range (31 through 39). Two patients of the latter group had two alleles with 39 and 43 CAG repeats, and 39 and 62 CAG repeats respectively. The distribution of expanded alleles is depicted in figure 1. Our results show an inverse correlation between repeat number and age at molecular diagnosis (r = −0.67), which is comparable with the correlation between repeat number of expanded alleles and age at onset of HD.11 Overall, 75% of the patients were examined between age 34 and 70 years, and 50% were between 41 and 62 years. Considering exclusively the patients with expanded alleles, 75% were examined between age 36 and 67 years, and 50% were examined between age 43 and 60 years, whereas patients without expansion had a broader age of distribution (75% were between 29 to 72 years and 50% of the patients were between 39 to 66 years).
Summary of diagnostic and presymptomatic DNA tests for HD in Germany, Switzerland, and Austria, 1993–1997
Figure 1. Distribution of expanded alleles in patients with Huntington’s disease.
Seventeen patients were found who had HD before age 20 (juvenile cases), and represent 0.9% of all patients with an expansion. A total of 135 patients (6.6% of all patients with an expansion mutation) had a late-onset form of HD (age at diagnosis, 70 years of age or older). Molecular testing confirmed the diagnosis in 24.6% of the juvenile group, in 69.1% of patients with the classic form of HD, and in 43.4% of the late-onset patients.
The sex ratio of the whole group shows a significant predominance of women (ratio of women to men, 1.4; p < 0.02). The mean age of the female and male groups are 52.2 years and 50.0 years respectively. Considering the group of patients without expanded alleles, we found a significant difference between men and women with respect to the age at diagnosis. The mean age of this group was 53.1 years for women and 48.9 years for men (t = 3.7, df 917, p < 0.001). In contrast, the group of HD patients was homogeneous with no remarkable differences regarding gender, CAG repeat number, or age at diagnosis.
Family history.
In 1,527 of 2,028 patients with an expansion it was possible to retrieve information on the family history (table 2). In FHP patients, HD was confirmed in 946 of 1,017 (93.0%) patients. HD was confirmed in 445 of 833 patients (53.7%) with FHU, and in 136 of 435 patients (31.3%) with FHN.
Distribution of the affected patients in relation to their family history
Within the group of HD patients, 62.0% had FHP, 29.1% had FHU, and 8.9% had FHN. The mean age at diagnosis in the three groups was FHP, 48.9 years; FHU, 53.9 years; and FHN, 57.1 years, and the mean CAG expansion was FHP, 45.6 CAG repeats; FHU, 44.7 CAG repeats; and FHN, 43.8 CAG repeats, which is in agreement with the inverse correlation of age at diagnosis (figure 2). The difference in the age at onset between the three groups carrying a CAG repeat expansion is significant by t-test analysis: FHP versus FHU patients: t = 7.1, df = 852, p < 0.0001; FHP versus FHN patients: t = 7.3, df = 182, p < 0.0001; and FHU versus FHN patients: t = 2.5, df = 235, p < 0.05. Considering the groups of patients without expansion, the statistical analysis is significant for the FHP patients versus the FHU patients (t = 3.6, df = 61, p < 0.001) and versus the FHN patients (t = 3.8, df = 64, p < 0.001), whereas there is no significance between the group of patients with FHU and FHN (t = 0.5, df = 602, p = 0.6). Investigating the size of the expanded alleles we also found a significant difference between the three groups (FHP versus FHU: t = 3.7, df = 900, p < 0.0005; FHP versus FHN: t = 5.9, df = 231, p < 0.0001; and FHU versus FHN: t = 2.5, df = 319, p < 0.05).
Figure 2. Chart illustrating the cumulative frequency of identified patients grouped in relation to the family history in each age group. The x-axis represents the upper limit of each age group. The y-axis is the cumulative frequency of detected patients in percent. ♦ = family history positive; ▪ = family history uncertain; ▴ = family history negative.
Presymptomatic and prenatal diagnosis.
A total of 992 individuals at risk requested presymptomatic testing (Germany, 886; Switzerland, 89; and Austria, 17). A total of 370 individuals (37.3%) were found to be carriers of alleles of more than 39 CAG repeats, and were therefore considered to have a high risk of developing HD. Fifty-two individuals have a mutable normal allele, and 30 individuals carry an allele in the reduced penetrance range. A total of 540 individuals have alleles in the nonexpanded range less than 27 CAG repeats (see table 1). The mean age of the whole group was 34.8 years. However, approximately 50% of the population at risk requested molecular testing between 23 and 33 years, with a clear predominance of women to men (ratio, 1.37). The remaining individuals at risk show a women-to-men ratio of 1.30.
We carried out 24 prenatal diagnoses. Fourteen parents were carriers of a CAG expansion for a total of 15 pregnancies. One woman requested a prenatal diagnosis in three subsequent pregnancies. The result of the first prenatal molecular test identified her as a carrier of the HD mutation. Nine parents had a 50% risk of being a carrier. The resulting average of the a priori risk for the fetuses was approximately 40%. Six fetuses were identified as carriers. The difference between the observed and expected number of carrier fetuses is not relevant (χ2 = 1.5; p = 0.2). One pregnancy continued; two subsequent pregnancies in the same woman, who herself carried the expansion, were terminated. No further information was available for the remaining three pregnancies.
Discussion.
Estimating the population of HD patients to be approximately 10,000 in Germany, Austria, and Switzerland, approximately 20% of the HD patients have been tested for the CAG repeat expansion of the HD gene. The population investigated also contained patients for whom HD was not considered to be the primary cause for the condition but was requested to be excluded (exclusion analysis). This could explain the lower rate of confirmation of diagnosis in our population compared with that of previously published studies.11 In 2,285 patients it was possible to correlate the presence of an expanded allele with the family history. A total of 93% of the patients with FHP carried a CAG expansion, which highlights the significance of a pedigree analysis for the diagnosis. The rate of patients having a CAG expansion decreased to 53.4% in the FHU group and decreased to 31.2% in the FHN group. We identified two patients with two expanded alleles. Unfortunately we do not have enough clinical data to describe their phenotype exhaustively. However, the age at diagnosis for the patient harboring 39 and 43 CAG repeats was 55 years, and for the patient having 39 and 62 CAG repeats was 22 years. This fits perfectly with the mean age of the patients carrying 43 repeats (mean age, 55.8 years; n = 304) and 62 repeats (mean age, 22.6 years; n = 5), and it might suggest that the phenotype is determined predominantly by the larger expanded allele.
The confirmation of the diagnosis in FHN patients is the lowest described until now. This is probably due to the composition of our group, which was not selected clinically and is therefore quite heterogeneous. However, the identification of the CAG repeat expansion in approximately one-third of the FHN group clearly demonstrates that the diagnosis of HD has to be considered for clinically affected individuals even in the absence of a positive family history. It remains to be determined which factors in particular contribute to such a high rate of HD within the FHN group. Nonpaternity, which is thought to be as high as 10% in high industrialized populations, new mutations, or early death of progenitors before manifestation of the disease are adding to this number. It can be hypothesized that alleles in the reduced penetrance/intermediate range progress to the range of full penetrance through moderate meiotic expansions. Alternatively, alleles in the low pathologic range (39 through 41 CAG repeats) may become fully penetrating relative to age. It is possible that parental generations were not affected because they did not reach the age of full penetrance of the alleles. Both hypotheses might explain the lowest average level of CAG repeats and the highest mean age, which was characteristic for our group of FHN HD patients. Analogous characteristics have been found in a group of 36 patients with FHN.12 Our assessment cannot address the question of the new mutation rate in HD. In this respect it is interesting, however, that patients with FHN make up 8.9% of the whole group. This frequency was proposed prior to direct molecular testing as a realistic percentage of new mutations in HD.13 Several reports have confirmed the existence of new mutations in HD.4,6,14,15 Except for some cases, in which the analysis of the parental CAG demonstrated the presence of fresh mutations definitively,16,17 most cases of recently reported new mutations4,6,14,15 are biased by the lack of information regarding parental DNA or by the lack of exclusion of slight symptoms of HD in the parents. In our opinion the issue of the new mutation rate of HD cannot be cleared completely. As a matter of fact, considering the demographic profile of patients with FHN, it is practically impossible to analyze the DNA of the patients’ parents in most cases to discriminate between cases caused by new mutations and cases of “pseudo” new mutations due to the early death of the carrier parent. However, our data are likely to portray the realistic probability of finding a CAG expansion in an unselected population of patients with FHN. Because details of the symptoms of these patients were not always available, it was not possible to define the phenotype in the patients with FHN. Additional studies characterizing the phenotype of patients who are carriers of the HD mutation and of patients who are clinically considered as having HD but do not have a CAG expansion would be worthwhile.
In the juvenile-onset group, 17 probands (25%) were found to carry an expanded HD allele, which was much lower than in the classic or late-onset groups. This stresses the rarity of juvenile forms, although in our group they may have been underestimated because we did not consider age at onset, but rather age at diagnosis. Furthermore, this indicates the difficulty of an accurate selection of symptoms considered as first signs of manifestation of HD for patients in this group. Of the 13 juvenile cases with FHP who were tested, only 6 had an expansion mutation. The confirmation of the diagnosis in this group was much lower (46.2%) than that observed for the whole group of FHP patients. It can be hypothesized that nonspecific symptoms have been attributed mistakenly to HD in the presence FHP or even that molecular testing is requested under the pretense of a differential diagnosis to obtain a presymptomatic test for children. There is sometimes a conflict between the parents’ wish to test their children and the refusal of the testing laboratory according to the guidelines set for DNA analysis.8 In fact, we identified a supposedly affected child who was 7 years old with 17 and 42 CAG repeats. Despite repeated requests, it was not possible to obtain detailed documentation attesting to the clinical state of the child. A similar situation has been described recently in a child carrying an allele with 43 CAG repeats with EEG changes and autism, who was tested at age 11 years.18 In both cases the family history was positive. It is unlikely that these symptoms are early manifestations of the disease due to the low pathologic range of the repeat expansion. These two examples stress the necessity for the insistence on obtaining clinical documentation on children referred for an HD diagnostic test, particularly if the family history is positive.
The demand for presymptomatic testing has been low, considering that only 3% of the eligible population decided to be tested. The relatively little interest in the test has already been observed, and our data confirm this attitude of the population at risk.19,20 The predicted demand for presymptomatic testing after the cloning of the gene was not apparent. In our countries, the number of requests for a presymptomatic test plateaued during the last 2 years (approximately 160 requests per year). We do not expect a dramatic change in the demand for predictive testing in the future unless a decisive therapy for delaying the onset of the disease is found. Women are, in general, more motivated in requesting the test, particularly from 23 to 33 years of age, during which time the majority of women make decisions regarding family planning. This prevalence of women is not related to the female-to-male ratio in the general population, because there are slightly more men than women in this age group.21
We found that 38.5% of individuals at risk are carriers of an expanded CAG repeat. This percentage of carriers is definitively different from the expected 50%. Most probably this is due to the presence of probands with a 25% or less a priori risk in our group rather than due to a selective force acting on individuals at risk. Alternatively, it is possible that individuals at risk with moderate symptoms did not request the molecular test. Prenatal diagnoses were a rare request. Unfortunately, we do not know the total number of pregnancies at risk during which parents at risk for HD had the option of a prenatal diagnosis to draw any conclusions.
Acknowledgments
Acknowledgment
The authors thank Prof. W. Engel, as initiator of the project, for his continuous encouragement, incitement, and critical advice. They also thank Profs. J.T. Epplen, U. Müller, and E. Schwinger for their support; and Profs. R. Witkowski, A. Schinzel, and B. Zoll, as well as Dr. R. Maiwald for critical review of the manuscript. Furthermore they thank Drs. H.N. Aschauer, D. Emmerich, and L. Pfeiffer, who collected data on numerous patients. They are grateful to all members of the consortium and to the patients’ support organization and are very obliged to the patients’ families.
- Received December 14, 1998.
- Accepted in final form April 3, 1999.
References
Letters: Rapid online correspondence
REQUIREMENTS
You must ensure that your Disclosures have been updated within the previous six months. Please go to our Submission Site to add or update your Disclosure information.
Your co-authors must send a completed Publishing Agreement Form to Neurology Staff (not necessary for the lead/corresponding author as the form below will suffice) before you upload your comment.
If you are responding to a comment that was written about an article you originally authored:
You (and co-authors) do not need to fill out forms or check disclosures as author forms are still valid
and apply to letter.
Submission specifications:
- Submissions must be < 200 words with < 5 references. Reference 1 must be the article on which you are commenting.
- Submissions should not have more than 5 authors. (Exception: original author replies can include all original authors of the article)
- Submit only on articles published within 6 months of issue date.
- Do not be redundant. Read any comments already posted on the article prior to submission.
- Submitted comments are subject to editing and editor review prior to posting.
