Presymptomatic testing in myotonic dystrophy: genetic counselling approaches
- Institute of Medical Genetics, University of Wales College of Medicine, Heath Park, Cardiff, CF4 4XN, UK
- Professor Harper,
Editor—We report the genetic counselling approaches used in a series of 72 presymptomatic genetic tests for myotonic dystrophy undertaken in our centre over an eight year period. The study has identified factors which influenced the counsellor's approach, and which can provide a basis for further, more systematic research.
Genetic counselling in myotonic dystrophy has always been difficult and complex, owing to the extreme variability of the disorder, in both severity and age at onset, with anticipation between generations and influence of the sex of the affected parent.
The identification of a CTG repeat expansion within the 3′ untranslated region of the myotonic dystrophy protein kinase gene on chromosome 191 as the primary molecular defect has transformed our understanding of the genetic aspects of this disorder and provides the basis for an accurate and specific diagnostic and presymptomatic test. The broad correlation of the size of the CTG expansion with age at onset and severity of the phenotype allows a limited degree of prognosis to be given to those found to have the mutation, particularly for very large or minimal gene expansions. Subjects carrying a minimal expansion (less than 100 repeats) usually show few or no muscle symptoms,2 but may develop cataract in later life; they contribute to a pool of mutation carriers who may transmit clinically significant disease to their offspring as a result of anticipation.
Presymptomatic genetic testing for late onset dominantly inherited disorders first became possible for Huntington's disease (HD),3 for which extensive experience has resulted in widely accepted guidelines for genetic counselling protocols; these comprise two pre-test sessions for information and preparation combined with post-test support.4 With appropriate adaptation, this has become a model for other late onset genetic disorders of the nervous system and to some extent also for the familial cancers. Although direct presymptomatic testing for myotonic dystrophy has now been available for eight years, we are not aware of studies so far on the counselling approach for this disorder in relation to presymptomatic testing. This lack of knowledge has prompted the present study, in conjunction with experience of direct molecular testing as a service in our centre, reported separately.5
Patients and methods
Since the identification of the specific gene mutation in 1992, up to June 2000, out of a total of 287 molecular analyses for the disorder in subjects living in Wales (population 2.9 million), there were 78 presymptomatic tests for myotonic dystrophy; by comparison 205 diagnostic tests on symptomatic subjects were performed and four prenatal tests. The laboratory methods used and the overall composition and outcomes of the series are described in the accompanying paper.5 For the purpose of the present study, we only included the 72 subjects seen for presymptomatic testing by 11 different clinical geneticists providing the clinical genetics service for Wales over this eight year period, based in the Institute of Medical Genetics, Cardiff, the remaining six samples having been received from neurologists and paediatricians. Data about the counselling approach were ascertained through the clinical genetics service notes and correspondence, supplemented by further discussion with the relevant clinical geneticist. We did not recontact any patients.
No single or specific counselling approach to presymptomatic testing in myotonic dystrophy has been advocated in Wales over this period, though one of us (PSH) has had a long standing clinical and research interest in the disorder. In general, subjects requesting presymptomatic testing were seen in their local medical genetics clinic by the clinical geneticist and genetics nurse specialist with designated responsibility for the particular district, not in a specialist clinic for the disorder.
When grouped according to the perspective of the subject tested, three different presymptomatic test situations could be distinguished (table 1). The largest group of subjects (A, n=58) represented the classical presymptomatic test situation with persons at 50% risk. Ten proved to have an abnormal result. Two of them showed minimal clinical abnormalities at the time of testing, but did not regard their symptoms as abnormal. Patients with an ongoing pregnancy formed a separate subgroup (B, n=5). All five (four females, one male) were healthy subjects at 50% risk asking for a presymptomatic test during an ongoing pregnancy. Three of them wished for a prenatal diagnosis in case of an abnormal presymptomatic test result. Results proved normal in all of them. The third category (C) concerned patients with no known family history of myotonic dystrophy at the time of the testing. For all of them, the issue of presymptomatic testing arose during the genetic counselling session although their original request had concerned another issue. We included nine cases in this group, six of them parents (five mothers, one father) of newly diagnosed affected children. All were referred as part of making the diagnosis in their children. They felt healthy, although five showed minor clinical features of myotonic dystrophy. Molecular testing was abnormal in all of them. The other three (one couple and a mother) had lost a fetus with unexplained arthrogryposis and/or neuromuscular abnormalities and wanted to know the risk for a further pregnancy. All three were completely healthy at clinical examination and the test result was normal in all of them.
As summarised in table 2 and discussed below, three different counselling approaches to presymptomatic testing for myotonic dystrophy could be identified. The most common practice (n=46) was to take the blood sample at the end of the initial genetic counselling session. The factors which decided the choice of the approach are listed in table3.
There is a general consensus that presymptomatic testing for late onset, mendelian, genetic disorders should not be considered as a purely laboratory procedure, but that it should be linked to appropriate genetic counselling in order to achieve a proper foundation of information, preparation, and support.6 How this can be ensured will vary according to the disorder and a universal model cannot be applied. There is, however, a range of issues common to most genetic disorders of late onset. These issues include the implications of an abnormal result for reproductive decisions, the genetic implications to offspring and other relatives, consequences for future employment and insurance, strategies for coping with an abnormal result, and the availability of support from family, friends, and professionals. Most reported experience so far has come from Huntington's disease, which represents an extreme situation regarding its severity and current lack of therapy.7 A two step approach has generally been recommended, both to allow the complex issues to be discussed fully and to give space to think about the testing process.4
For myotonic dystrophy, however, no guidelines have been produced so far, and the approaches used have been based on individual patterns of clinical practice. One clear difference between the two disorders relates to the penetrance of the gene, since whereas in Huntington's disease a considerable risk of serious disease exists for healthy subjects at risk who have passed 50 years of age, for myotonic dystrophy the great majority of offspring of an affected person will show definite clinical features of the disorder by early adult life, while most late onset cases are clinically mild. Brunneret al 8 showed in a total of 139 clinically normal offspring of myotonic dystrophy patients that the residual risk of carrying the myotonic dystrophy gene mutation is approximately 8% between the ages 20 and 39 and a comparable risk (8.6%) was shown in a linkage based series.9 In the present series, the proportion of abnormal presymptomatic test results in clinically normal subjects (seven out of 69, 10.1%) was comparable. However, in only two such cases (males aged 19 and 37 years) was the abnormal expansion in the range likely to be associated with significant neuromuscular symptoms (>100 repeats). Thus, likelihood of a clinically normal myotonic dystrophy relative carrying a mutation significant for their own health is small, especially after the age of 40 years.
We can now consider in turn the different approaches to genetic counselling identified in this series (table 2). The factors listed in table 3 are among those particularly relevant to the approach that may be most appropriate.
The two stage approach, corresponding to that used for Huntington's disease and comparable late onset neurodegenerative disorders, was used in only seven of the 72 cases, being offered but rejected in two others. Factors favouring this approach include unfamiliarity with the disorder and its consequences, complexities of family relationships and dynamics, and young age of the person to be tested, giving a greater likelihood that an abnormal result might indicate relatively severe disease. In general, we would advocate it in any situation where the person is uncertain about a decision to be tested and where the amount or complexity of new information to be considered is too great to be given satisfactorily in a single session or where a period of reflection is needed.
In contrast to Huntington's disease, where this complex situation is almost invariably present, this is not always the case for myotonic dystrophy, so we see no need for a two step approach to be recommended in all presymptomatic testing situations for myotonic dystrophy; equally, where it is suggested but declined, as occurred in two cases in our series, we see no reason why it should be insisted on.
In the great majority of presymptomatic tests in the present series (46 out of 72), a blood sample for analysis was taken at the end of the initial genetic counselling session. This approach seems particularly appropriate where the subject is familiar with the disorder, where there are no family complexities, or where their age (over 40 years) makes the likelihood of an abnormal result small and where even an abnormal result is unlikely to have major clinical significance. This last factor contrasts strongly with presymptomatic testing for Huntington's disease, where the chance of an abnormal result remains high at a relatively advanced age and where the clinical consequences of the disorder are serious regardless of age at onset. In myotonic dystrophy, older subjects detected as having clinically insignificant expansions should perhaps be considered gene carriers rather than affected patients.
In eight instances in this series, an intermediate approach was used (2A and B in table 2), where opportunity was given for reflection following the interview by either storing the blood sample until the person had confirmed that they wished testing to proceed (five cases) or by arranging for the blood sample to be taken after an interval by the family doctor or genetics nurse specialist (three cases). This would appear to provide a useful option when the case does not clearly fall into either the two stage or single stage situations.
In five cases, the request for presymptomatic testing was made in the context of an ongoing pregnancy, with prenatal diagnosis being wished for in three of these. There was thus a situation of time pressure, making a full two step approach difficult even if considered desirable, a problem not infrequently met also in presymptomatic testing for Huntington's disease. In all three cases potentially requiring prenatal diagnosis, blood was taken at the end of the first interview; all had a normal result so that no further action was required. In the two cases where prenatal diagnosis was not wished for, an intermediate approach as outlined above was used.
In nine cases in this series, blood was taken at a home visit by the genetics nurse specialist before the genetic counselling clinic attendance, this only being arranged if the result was abnormal. This was considered appropriate at the time in cases where no special complexities were expected and it is likely that some unanticipated difficulties would have been detected at the home visit, allowing the sampling to be postponed. However, we would not now recommend this approach for a number of reasons. First, it gives no opportunity for clinical assessment before testing is undertaken, a factor of considerable importance in view of the high frequency of asymptomatic subjects at risk who show clinical abnormalities. Second, while some of the relevant issues would have been able to be discussed at the home visit, it is unlikely that a full picture of the potential consequences of testing would have been obtained in this way. Thirdly, by seeing only those subjects with an abnormal result in clinic, no opportunity was given to those with a normal outcome to discuss the disorder and its consequences for affected family members, something that might be relevant to them as a relative, even though not affecting themselves.
Two areas which require particular discussion are the role of clinical examination in the presymptomatic testing process and the importance of the subject's perception as to whether testing was presymptomatic or diagnostic. Previous family studies have clearly shown that a considerable proportion of asymptomatic family members show definite clinical abnormalities on examination (17.6% in one early study).8 In the present study, some clinical abnormalities were present in nine of the 16 cases with an abnormal presymptomatic test result, in the total series of 78 cases (see accompanying paper5 for details). By contrast, the likelihood of an abnormal genetic test result in a carefully examined and clinically normal adult is low (around 8% in the studies discussed above). The figure of seven out of 78 cases in the present series is comparable. It can thus be argued that clinical assessment is likely to make a greater contribution to outcome than laboratory analysis and should always be undertaken before genetic testing. On the other hand, since some people request testing primarily for reproductive reasons and may not recognise the possibility of being themselves affected, this possibility and the reasons for clinical assessment will need careful explanation before testing is performed, with the timing of the assessment depending on the individual situation.
A related important issue is the possible difference in perception of the person being tested and the clinician involved as to whether the testing is “presymptomatic” or “diagnostic”. We have considered as presymptomatic in our series all those cases who had no complaints, even though abnormal or suspicious clinical features might have been present at the time of testing. When the referring clinician has recognised these features it is possible that the test situation may be handled as a diagnostic confirmation. A particular issue in myotonic dystrophy exists for the mothers of children suspected of having congenital myotonic dystrophy, who may be tested as part of their child's diagnosis, whereas for the mother, the situation is likely to be presymptomatic. Example 1 illustrates this.
A 3 year old boy, accompanied by his mother, was referred by a paediatrician for evaluation of a possible genetic cause for his unexplained motor difficulties. The mother wanted to know the diagnosis in her son and the risk for further children. The family history was unremarkable. Clinical examination of the son showed reduced muscle tone and slight developmental delay. The mother had grip myotonia on examination, but had not previously regarded it as abnormal. She stated that she and her son both slept without fully closing their eyes. For the clinical geneticist involved, the diagnosis of myotonic dystrophy was clear. A possible clinical connection between the symptoms of both were discussed and blood was taken for molecular testing for myotonic dystrophy from the mother only. At the second appointment, the test result, which confirmed the diagnosis in the mother, was discussed. The mother was very upset about the result and a pause for reflection was necessary before testing the son. At the third appointment, all medical and reproductive implications were discussed in detail and blood was taken from the son for molecular analysis. The result showed an abnormal expansion.
Example 1 illustrates the difficulty in disclosing the diagnosis in minimally affected subjects with no subjective complaints and who are not familiar with the disorder. The reaction of the mother clearly shows that she was caught in a dilemma and did not have enough space to understand what was going on or to think about her own health. The diagnosis became shifted towards herself although the clinical request had concerned her son. For the clinician, the mother's situation appeared diagnostic. The view of the authors, however, is that as such people consider themselves healthy, a presymptomatic rather than diagnostic approach seems more appropriate. This means that there is a need for information and counselling before molecular testing. It seems appropriate to answer first the question of the diagnosis in the child and postpone the blood sampling in the mother until she is informed, prepared, and supported.
No instance of presymptomatic testing of a healthy young child was recorded in our series, the youngest being aged 16 years. In general, the policy recommended for late onset disorders was followed6 with full discussion, clinical assessment where required, but with postponing of genetic testing until an age was reached where the implications of testing could be understood and consent given.
In general, this policy was understood and supported by parents, but in a small number of cases a strong and persistent wish for testing of a healthy child was expressed, as outlined in example 2.
Example 2 shows how perceived barriers can interfere with full appreciation of the issues involved in a difficult situation. One possible explanation of the outcome might be that removal of the perceived barrier allowed a fuller understanding of the potential problems, even though these had been repeatedly explained on previous occasions.
Presymptomatic testing was requested for a 1 year old child by the mother on account of a family history of myotonic dystrophy in the child's father. At the time of the consultation, the parents had separated and it was not possible to determine whether or not the father was clinically affected, though the diagnosis of myotonic dystrophy in other family members was clear. The child was healthy and developing normally; clinical examination was offered and was entirely normal. It was explained that in view of this, it would be wise for presymptomatic testing to be postponed until the child was of an age to consent and annual review was offered. Despite this, the mother remained determined that the child should be tested, although recognising the issues of the child being unable to consent and that there were no specific medical or other benefits from childhood testing. The mother and child were seen on three further occasions, with the mother's demand unchanged despite normal clinical assessments. Following discussion between the professionals involved, it was considered that the issue of testing was itself becoming a significant problem that might outweigh any disadvantages. Four years after the initial consultation, testing of the child was offered, again in the context of full explanation of the potential consequences, and the mother was requested to bring the child at a convenient time for a blood sample to be taken. She did not return and has not requested testing of the child over the subsequent two years.
A final counselling issue requiring mention relates to the potential need for disclosure of genetic test results in relation to insurance, a topic of considerable debate in the UK and elsewhere. Although most of the period covered by our retrospective study predated the emergence of this issue, it is now a topic that requires discussion with people before testing. However, the small number of subjects in the series (two out of 78) who were clinically normal, yet had an abnormal test result of clinical significance to themselves, suggests that there is little need for insurers to have access to such results, and that a normal clinical assessment is a more relevant factor.
In conclusion, the counselling approach in relation to presymptomatic testing for myotonic dystrophy needs to be flexible, so as to respond to the variable and complex issues that may arise. Adherence to a two stage process does not seem required for the majority of situations, but each instance requires careful consideration and at least one full interview giving opportunity for detailed discussion of the issues would seem essential. While the retrospective nature of the present study makes it unwise to draw definitive conclusions, the issues raised should provide the starting point for further, more systematic study, as well as giving a general framework that may be useful for those involved with presymptomatic testing for this important and exceptionally variable condition.
We thank all the clinical staff of the Institute of Medical Genetics for permission to report information on their patients and, in particular, Professor Angus Clarke and Dr Helen Hughes for helpful discussions. Siv Fokstuen was supported by a grant from the Swiss Academy of Medical Sciences.