Background A small proportion of cases seen in neurodegenerative conditions such as amyotrophic lateral sclerosis (ALS), Parkinson's disease and Alzheimer disease are familial. These familial cases are usually clinically indistinguishable from sporadic cases. Identifying familial cases is important both in terms of clinical guidance for family members and for gene discovery.
Method Surveys assessing the definition of familial amyotrophic lateral sclerosis (FALS) were completed by clinicians with an interest in ALS.
Results 95 surveys were completed by respondents from 15 countries. A third of total respondents stated that they thought that neurologists were using the same definition for FALS (33.3%, 30). No consensus was achieved among clinicians when provided with five different definitions for FALS. However, the preferred definition was ‘a patient with ALS with either a first or second degree relative also with ALS’ (37.8%, 31).
Conclusion There is no consensus on a standard definition for FALS among clinicians. It is likely that similar inconsistencies apply to other conditions, such as Parkinson's disease and Alzheimer disease, in which both familial and sporadic diseases occur. Inconsistent classification could hinder gene discovery.
Statistics from Altmetric.com
Amyotrophic lateral sclerosis (ALS) is a neuro-degenerative disease affecting 1 in 400 people in their lifetime.1 A recent meta-analysis reported that the rate of familial amyotrophic lateral sclerosis (FALS) was 5%.2 However, only 6% (2/33) of studies included in the meta-analysis provided a definition for FALS. Similarly, only 2 of 13 (15%) of papers reporting the rate of SOD1 mutations in specific cohorts define the term FALS, and the two definitions differ.2 These findings posit the question as to the precise definition of FALS, and as to the presence of a uniformly applied classification among clinicians.
We designed and administered an online questionnaire, which was distributed to clinicians involved in the diagnosis and management of ALS. The survey had three aims: to determine whether a consensus exists among clinicians regarding a standard definition for FALS, to examine clinical practice in the diagnosis of FALS and to seek opinion as to the need for a consensus meeting on the definition of FALS.
Based on a series of qualitative interviews with key opinion leaders in ALS, a 26-item questionnaire was devised (supplementary appendix e-1). The study was classified and approved as an audit by the local IRB. The questionnaire sought information in six domains:
demographic information (A)
is there a standard definition for FALS in use by neurologists? (B)
the definition for FALS used by the respondent in their clinical practice (C)
existing practice among responding clinicians when confronted with a possible FALS pedigree (D)
desire for a consensus meeting to define FALS (E).
The questionnaire was distributed using SurveyMonkey® to ALS mailing lists in Europe (ENCALS), North America (NEALS), Australia and India.
Demographic information (A)
There were 95 respondents from 15 countries (see supplementary table 1). In all, 61 (64.9%) of the respondents were male subjects. Sixty-eight (75.6%) of the respondents were neurologists with the remainder being trainee neurologists and clinical geneticists (21.1% and 4.4%, respectively). Eighty (85.1%) declared that they had a special interest in ALS.
Is there a standard definition for FALS in use among neurologists? (B)
Respondents were asked if they thought that there was a standard definition among neurologists for FALS. A third of total respondents stated that neurologists were using the same definition for FALS (33.3%, 30). There was a statistically significant difference when subgroup analysis based on country of practice was carried out: over half of respondents from North America (51.4%, 18) stated that there was a standard definition for FALS in use among neurologists in comparison with less than a quarter of respondents from Europe (22.0%, 9) (p=0.015).
To explore the definition for FALS used by the respondent in their clinical practice (C)
Respondents were provided with five possible definitions for FALS. They were asked to select the definition that in their opinion is most applied in publications relating to FALS. There was no consensus among the respondents (see Q1 results, table 1).
Respondents were then provided with the same five definitions and asked to select the one that most matched their own clinical practice (see Q2 results, table 1). Again, there was no consensus, but the preferred definition was ‘a patient with ALS with either a first or second degree relative also with ALS’ (37.8%, 31).
In subgroup analysis, the preferred response for the definition of FALS among respondents who stated that they had a special interest in genetics and FALS was ‘a patient with ALS with any relative with ALS, no matter how distant’ (34.8%, 8).
If the respondents did not use any of the given definitions in their clinical practice they were provided with an option to add their own definition.
To evaluate practice among responding clinicians when given a kindred scenario (D)
Respondents were then provided with eight pedigrees and asked whether they would diagnose the proband with FALS (figure 1).
In cases (A) and (B), the level of consensus was high. However, in over half of the pedigrees the majority of respondents chose the ‘Maybe’ option. In all, 51 (58.6%) respondents gave answers that differed from the definition that they had previously stated that they used in clinical practice.
Fifty-nine (67.0%) of respondents stated that they carried out genetic testing on patients who had a family history of ALS. Only 9 (10.3%) respondents stated that they routinely carry out genetic testing on ALS patients with no family history.
In all, 52.3% (45) of respondents stated that they refer patients with FALS for genetic counselling.
To assess the need for a consensus meeting among clinicians (E)
Sixty-five (74.7%) respondents agreed that a consensus meeting would be helpful in defining FALS.
This study demonstrates that there is no standard definition for FALS. Although the majority of respondents agreed that having one other affected family member with ALS was sufficient to constitute a diagnosis of FALS, opinions differed as to whether this relative should be at the very least a first-degree relative, a second-degree relative or any relative, no matter how distant. Notwithstanding, over a third of respondents believe that the term FALS carries a standard definition among specialists. Despite this, in a series of eight kindred scenarios the majority of respondents chose the ‘maybe’ option in four of the scenarios, indicating a lack of certainty as to what constitutes the familial form of the disease. The greatest degree of uncertainly was observed in scenarios where a distant relative was affected. Moreover, even when clinicians had clearly stated their preferred definition of FALS, over half of respondents were inconsistent in their definition when confronted with specific scenarios.
As is the case for all conditions in which there is both a familial and sporadic form, a clear and uniformly applied definition is important. We have shown previously that there are large differences in the reported rates of FALS across different countries.2 While it is plausible that these differences are due to a founder effect with any given population, it is also conceivable that these differences arise from differing definitions for the term FALS. The same is true for reported rates of SOD1, which vary widely among different research cohorts.3 Again, this disparity may be due to a true difference in the rate of SOD1. However, it is also possible that these differences are a function of inconsistencies in classification.
A consensus definition of what is familial disease is also important when selecting kindreds for linkage and exome sequencing. Among respondents with an interest in ALS genetics and familial ALS, the most common definition for FALS was ‘any relative with ALS, no matter how distant.’ We have shown previously that chance plays a role in a high proportion of FALS kindreds where only two members are affected with ALS and that true role of genetics in a kindred where only two members are affected may be considerably lower that than where three or more members are affected.4 Therefore, research groups that study high proportions of kindreds with only two affected members are likely to observe a rate of SOD1 mutations that is lower than groups that draw from kindreds with larger affected number (see supplementary figure 1). A recent paper by Al-Chalabi and Lewis demonstrated that family size and gene penetrance directly contribute to observed inheritance patterns within a kindred, and that poorly penetrant genes in small families may appear to cause sporadic disease.5
A further factor to consider is that poorly ascertained family histories may mask familial disease. It is important to take a thorough family history from a reliable source. The clinician must acknowledge the many reasons that a patient may not report a family history of ALS or frontotemporal dementia (FTD) when in fact one does exist, including lack of information about past generations, misdiagnosis, relatives dying of another condition before they develop ALS and denial.4 Moreover, the recent identification of an expanded hexanucleotide repeat in families with both ALS and FTD6 7 further underlines the importance of obtaining as extensive a family history as possible in apparently sporadic cases, and neurologists in training should be educated to this effect. It should also be noted that familial disease does not always imply a genetic origin, although in the case of ALS, no definitive environmental exposure has been identified to date.
In all, 75% of respondents agreed that there should be a consensus meeting on the definition of FALS. Such a meeting should include both neurologists and geneticists. Outcomes from such a consensus should be an agreed definition that is clear, concise and clinically applicable. An additional research requirement could be that clinicians record the total number of people within the kindred and the degree of relatedness of the other affected family member. Certainty as to the definition of FALS will render epidemiological and genetic studies more transparent and will in turn permit more accurate comparison between geographic regions.
It is likely that similar inconsistencies apply to other conditions, such as Parkinson's disease and Alzheimer's disease, in which both familial and sporadic diseases occur. A consensus initiative to provide precise definitions of familial and sporadic disease is urgently required.
Statistical analysis was completed by Susan Byrne using SPSS V.16.0.
Funding This work was funded by the Health Research Board, Ireland. Prof Hardiman is an HRB Clinician Scientist. Her group has received unrestricted research grants from Merck Serono, Biogen Idec and Bayer Schering. She has received honoraria for providing expert advice to Merck Serono, Biogen Idec, Janssen Cilag, Allergan, Ono Pharmaceuticals and CytRx.
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement Statistical and data tables are available from the corresponding author ( ).
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.