Lack of consensus in ALS genetic testing practices and divergent views between ALS clinicians and patients

Abstract

Recent advances in ALS gene discovery have both empowered and challenged clinicians providing evaluation and care for persons with ALS, many of whom seek an answer as to the cause of their condition. In order to study clinician practices and attitudes towards genetic testing, we surveyed members of the Northeast ALS Consortium, an international group of specialist ALS clinicians; responses were received from 80/255 (response rate = 31.4%). While 92.3% indicated they offered genetic testing to patients with familial ALS, 57.0% offered testing to patients with ALS and a family history of dementia, and 36.9% offered testing to patients with sporadic ALS, revealing a lack of consensus with respect to the approach to the typical ALS patient encountered in clinical practice. In addition, comparison of clinician and patient attitudes towards genetic testing revealed that clinicians valued the scientific potential of testing, but were less likely to say they would have testing themselves, or to see the value in testing for family members. People with ALS were more likely to see value of testing for themselves and for family members, and less likely to strongly value the scientific potential of testing.

Introduction

Ten years ago, commercial ALS genetic testing was available for only one major ALS gene, SOD1, which causes up to 20% of fALS in some populations. Current genetic testing options include assays for the C9orf72 expansion, multigene panels consisting of >25 genes, and whole exome sequencing. Although a genetic etiology may now be established in ~70% of fALS and ~15% of sALS (1), the offer of genetic testing is not yet ‘standard of care’ in practice, particularly for individuals with sALS (2, 3, 4). In 2012, the European Federation of Neurological Societies (EFNS) directed that genetic testing should not be performed in ‘cases with sporadic ALS with a typical classical ALS phenotype’ (5), while US guidelines do not address the issue (6).

Several recent studies have investigated genetic testing practices among ALS clinicians. Byrne et al. (2) surveyed an international group of neurologists and reported that 67.0% offered genetic testing to individuals with fALS, while 10.3% offered genetic testing to sALS patients. Arthur et al. (3) surveyed the Northeast ALS Consortium (NEALS, an international group of clinicians who specialise in ALS management), and reported that 93.0% offered testing to individuals with fALS, 30.2% offered to those with sALS, and 97.7% provided genetic counselling. Most recently, Vajda et al. (4) surveyed an international group of neurologists and found that 90.2% offered genetic testing to fALS patients and 49.4% offered to sALS patients. Taken together, this research demonstrates a growing consensus that individuals with fALS are offered genetic testing, but the offer of testing to persons with sALS remains inconsistent and perhaps represents the most acute unresolved issue affecting patient care in this area. The underlying reasons for practice variability and clinician attitudes towards the value of genetic information have not been extensively studied.

With the goal of characterising clinician practices, perceived challenges, and attitudes towards ALS genetic testing, we surveyed NEALS Consortium. The survey gathered information regarding genetic testing and counselling practices, as well as clinician attitudes and other factors that may be associated with the offer of genetic testing. In addition, we compared clinician attitudes towards genetic testing to attitudes of persons with ALS, which we have previously reported (7). The findings of this study may inform the development of consensus practice guidelines for ALS genetic testing and counselling, which may in turn enable more consistent patient access to genetic information, risk assessment, and, in the future, targeted treatment options for persons with ALS.

Methods

The Institutional Review Board (IRB) at The Ohio State University, Columbus, Ohio, approved this survey. Clinicians with a specialty in motor neuron disorders were identified from the NEALS directory. Physical or occupational therapists, study coordinators, or others deemed unlikely to order genetic testing were excluded, as were ten members who opted out of receiving research announcements. The remaining members (n=255) were eligible to receive the survey and were sent an email in August 2017 containing a link to the survey webpage. Ten participants completed the survey at the 2017 NEALS Annual Conference via tablet computers. All responses were stored anonymously using the survey engine SurveyMonkey.

The clinician survey (see supplementary material) consisted of 46 multiple-choice questions, with skip-logic. Questions covered respondent practice type (academic, private, hospital), training history, and volume of ALS patients seen. The bulk of the survey addressed practices surrounding ALS genetic testing and counselling, including family history documentation, criteria for the offer of genetic testing, and result interpretation and disclosure. The final question consisted of a 9-item Likert scale series documenting respondent attitudes towards genetic testing and its utility for patients, clinicians, and society. The responses ranged from ‘strongly agree’ to ‘strongly disagree’, with the option of ‘neutral’ when applicable. For some analyses, the responses were grouped into three categories: ‘agree’, ‘neutral’, and ‘disagree’.

Statistical methods

Nominal responses were summarised using proportions. Associations between nominal responses were studied using contingency tables and Pearson chi-square tests. Continuous responses were summarised using means and standard deviations. Comparisons of mean Likert scores for individual questions and grand means between two groups were carried out using 2-sample t-tests assuming unequal variances. A two-tailed p-value of 0.05 or less was considered statistically significant. All statistical analyses were performed using JMP Version 12 Pro software (SAS Institute, Cary, NC).

Results

The survey was emailed to 255 members of the NEALS; 80 responded (participant response rate = 31.4%). Approximately half (n=59/121, 48.8%) of NEALS sites are represented by at least one completed survey. The majority of respondents identified as neurologists (96.2%); the remainder identified as ‘Other’ (n=2, 2.5%) or genetic counsellor (n=1, 1.3%). Most indicated that they were employed in an academic practice setting (68/76, 89.47%), while 20/76 (26.3%) indicated hospital, 9/76 (11.8%) indicated private practice, and 1/76 (1.3%) indicated other; multiple selections were permitted. Approximately half (35/76, 46.1%) of respondents reported seeing patients in an ALS tertiary centre. Data gathered on patient volume indicated that the majority of respondents (69.7%, 53/76) saw 40 or more ALS patients per year, while 10.5% (8/76) saw between 30 and 40, 9.2% (7/76) saw between 20 and 30, 7.9% (6/76) saw between 10 and 20, and 2.6% (2/76) saw fewer than 10. Most respondents were board-certified in neurology (94.7%, 72/76) and neuromuscular medicine (61.5%, 47/76). Respondents were asked to report which disciplines provide care in their respective ALS clinics; this included neurologist (98.7%, 74/75), MDA/ALSA representative (93.3%, 70/75), and physical/ Occupation therapist (, 93.3%, 70/75), Social worker (81.3%, 61/75), respiratory therapist (78.7%, 59/75), nurse practitioner (48.0%, 36.8), and genetic counsellor (32.0%, 24/75).

Genetic testing and counselling practices

Questions were asked about clinician practices regarding genetic testing and counselling. Most respondents (94.7%, 71/75) reported that family history information is documented for ‘all ALS patients’ in their clinics, while 5.3% (4/75) document for ‘some’ patients. Methods of family history documentation included the narrative note (90.7%, 68/75), the family history tool of the electronic medical record (54.7%, 41/75), a graphic pedigree (18.7%, 14/75), or ‘other’ method (5.3%, 4/75). Most respondents (70.3%, 52/74,) indicated that the genetics of ALS was discussed with ‘all patients’, while half specifically indicated ‘patients with a family history of ALS’ (51.4%, 38/75), ‘patients who ask’ (50.0%, 37/74), those who report a ‘family history of dementia or other neurodegenerative condition’ (44.6%, 33/74), or those who have a ‘young age of onset’ (36.5%, 27/74); multiple responses were permitted.

The majority of respondents (93.2%, 69/74) reported that ALS genetic testing was offered in their practices. A few (5.4%, 4/74,) indicated that patients are referred to genetics or other services for testing and one respondent indicated that patients are not offered nor referred for testing. Most respondents (63.8%, 44/69) indicated that clinical genetic testing and research-based testing are offered, while 31.9% (22/69) offer clinical only, and 4.3% (3/69) offer research only. The majority of respondents (73.8%, 48/65) indicated they would be more likely to offer ALS genetic testing in their practice if there were consensus guidelines to direct the offer of testing.

Clinicians were asked to indicate which family history indication typically leads to the offer of genetic testing in their practice. The majority of respondents (92.3%, 60/65) indicated a family history of ALS (fALS); 57.0% (37/65) indicated a family history of dementia or other neurodegenerative condition (dALS), and 36.9% (24/65) indicated that testing is offered to patients with sporadic ALS (sALS) (see Table 1). In addition, 53.9% (35/65) indicated that testing is offered to patients who have a young age of onset, and 50.8% (33/65) offer testing on patient request. Only 15.4% (10/65) of respondents reported that their clinics have a policy in place to guide which patients are offered genetic testing.

Table 1.

Offer of genetic testing to patients with ALS, by family history indication

Family History Category	fALS*	dALS**	sALS***
Genetic Testing offered?	92.3% (60.65)	52.3% (34/65)	36.9% (24/65)

^*family history of ALS

^**family history of dementia

^***no family history of ALS or dementia

Data gathered on test result disclosure revealed that most respondents provide results to the patient and/or family members in a face-to-face encounter (98.4%, 62/63), followed by telephone (33.3%, 21/63), and/or by letter (15.9%, 10/63). Clinicians were asked to indicate how results yielding only variant(s) of uncertain significance (VUS) are handled in their practice; the majority (83.6%, 51/61) indicated that patients are told the result is inconclusive, 15.0% (15/61) refer such cases to genetics or ALS tertiary centre, 16.4% (10/61) indicated ‘other’ actions, 3.3% (2/61) indicated ‘not sure’ and1.6% (one) indicated the patient is told the test is negative. Clinicians were also asked what discussion is provided to patients regarding the implications of positive test results for family members of a proband. The majority of respondents (95.1%, 58/61) reported that the risk to family members is always discussed, while 4.9% (3/61) indicated that this is discussed only if the patient or family asks about this.

Attitudes towards genetic testing

Data from a 9-item Likert scale assessing clinician attitudes towards ALS genetic testing found that respondents overall had positive perceptions of genetic testing and its value for ALS patients, families, research, and clinicians (Table 2). Most respondents agreed that genetic testing should be offered to ‘patients with ALS’ (80.3%; n=53) and to ‘patients with ALS who have a family history of ALS’ (97.01%; n=65). The majority of clinicians also agreed that the testing provides useful information to patients (92.54%; n=62). However, responses to the statement ‘Genetic testing should be offered to family members of persons with ALS’ were more variable (agree=37.3%, neutral=29.9%, disagree=32.5%). A mean attitude score was calculated for each respondent; the average attitude score of all respondents was 2.23 (with a possible range of 0 to 10, with 0 being the most positive and 5 representing a neutral attitude).

Table 2.

Clinician attitudes towards genetic testing

Attitude Item	n	% Agree	% Neutral	% Disagree	Mean Attitude Score*
Genetic testing should be offered to patients with ALS	66	80.30	12.12	7.58	2.67
Genetic testing should be offered to patients with ALS who have a family history of ALS	67	97.01	2.99	0.00	1.39
Genetic testing can provide useful information to patients with ALS	67	92.54	7.46	0.00	1.93
Genetic testing should be offered to family members of persons with ALS	67	37.31	29.85	32.84	1.93
Genetic testing should be offered to family members of persons with ALS who have a gene mutation	66	68.18	18.18	13.64	3.15
Genetic testing can provide useful information to family members of person with ALS	65	72.31	20.00	7.69	3.15
If I had ALS and were offered genetic testing, I would have testing	67	62.69	22.39	14.93	3.24
Genetic testing can help researchers/clinicians understand ALS	67	97.01	1.49	1.49	1.42
In the future, genetic testing may help researchers/clinicians	66	98.48	1.52	0.00	1.30
					Average:2.23

^*Attitude score was calculated for each respondent by assigning a numeric value to each Likert response, with a possible range of 0 to 10, with 0 being the most positive and 5 representing a neutral attitude.

In order to investigate possible factors involved in clinician practice variability, respondent characteristics and attitudes were examined for association with genetic testing practice. Excluding five respondents who never offer genetic testing, clinicians who indicated that sALS patients should be offered genetic testing had more positive test attitude scores (p=0.0001). Respondents who saw more than 40 ALS patients per year were somewhat more likely to offer genetic testing to patients with sALS, but this did not reach statistical significance (p=0.09). Clinicians who work in a clinic also staffed by a genetic counsellor were no more likely to offer genetic testing to patients with sALS (p=0.49), to patients with early onset ALS (p=0.71), or to patients with ALS with a history of dementia (p=0.99). However, clinicians who work with a genetic counsellor had more favourable mean attitude scores (p=0.03).

Clinician responses to the genetic testing attitude scale were compared to patient responses to the same scale in a survey of ALS patients administered by the authors (7; Table 3). Both the clinician and patient groups had similar, positive overall scores (mean clinician score = 2.23, mean patient score 2.11, p=0.39). However, comparison of individual item scores revealed differences: patients were much more likely than clinicians to agree that ‘Genetic testing should be offered to family members of persons with ALS who have a gene mutation’ (p=0.00001), ‘Genetic testing provides useful information to family members of persons with ALS’ (p=0.0006), and that ‘If I were offered genetic testing for ALS, I would have testing’ (p=0.004). On the other hand, clinicians were more likely than patients to agree that ‘Genetic testing can help researchers/clinicians understand ALS’ (p=0.004) and ‘In the future, genetic testing can help researchers/clinicians’ (p=0.00001)‥

Table 3.

Comparison of patient and clinician attitudes towards genetic testing

n	Patient Survey (2015)	Mean Attitude Score*	p	Mean Attitude Score*	Clinician Survey (2017)	n
449	Genetic testing should be offered to patients with ALS	2.17	0.0400	2.67	Genetic testing should be offered to patients with ALS	66
449	Genetic testing should be offered to patients with ALS who have a family history of ALS	1.54	0.2291	1.39	Genetic testing should be offered to patients with ALS who have a family history of ALS	67
448	Genetic testing can provide useful information to patients with ALS	2.26	0.0541	1.93	Genetic testing can provide useful information to patients with ALS	67
449	Genetic testing should be offered to family members of persons with ALS	2.22	0.0909	1.93	Genetic testing should be offered to family members of persons with ALS	67
448	Genetic testing should be offered to family members of persons with ALS who have a gene mutation	1.90	<0.0001***	3.15	Genetic testing should be offered to family members of persons with ALS who have a gene mutation	66
447	Genetic testing can provide useful information to family members of persons with ALS	2.20	0.0006***	3.15	Genetic testing can provide useful information to family members of person with ALS	65
448	If I were offered genetic testing for ALS, I would have genetic testing*	2.30	0.0036***	3.24	If I had ALS and were offered genetic testing, I would have testing*	67
447	Genetic testing can help doctors understand my condition*	2.43	<0.0001***	1.42	Genetic testing can help researchers/clinicians understand ALS*	67
448	In the future, genetic testing may help doctors treat ALS*	2.01	<0.0001***	1.30	In the future, genetic testing may help researchers/clinicians*	66
		Avg: 2.11	Avg p-value: 0.3905	Avg: 2.23

^*Attitude score calculated for each respondent and averaged (with a possible range of 0 to 10, with 0 being the most positive and 5 representing a neutral attitude)

^**Wording of question tailored to respondent context

^***p-value indicates significance (<0.05)

Discussion

Results from this survey support findings of previous studies indicating that a growing proportion of clinicians are offering ALS genetic testing in their practices. The offer of testing to ALS patients who do not have a family history of the condition remains variable, however (see Table 4). We sought to identify clinician or practice characteristics associated with the offer of genetic testing to patients with sALS. Vadja et al. (4) showed that clinicians with higher volumes of ALS patients were more likely to offer genetic testing to sALS and fALS patients; our results suggested a similar trend that did not reach statistical significance, possibly due to smaller sample size. Clinicians who worked in a clinic also staffed by a genetic counsellor were not more likely to offer testing to patients with sALS. The only clinician characteristic we identified that was positively associated with the offer of testing to sALS patients was the genetic testing attitude score.

Table 4.

Percentage of clinicians offering ALS genetic testing, by family history indication, in practice surveys

Study	n	fALS*	sALS***
Byrne et al., 2012	95	67.0%	10.3%
Arthur et al., 2016	43	93.0%	30.2%
Vajda et al., 2017	167	90.2%	49.4%
Current study, 2017	65	92.9%	36.9% (sALS)*** 57.0% (dALS)**

Percentage of clinicians offering genetic testing for ALS patients with *fALS, family history of ALS; **dALS, family history of dementia; ***sALS, no family history of ALS or dementia

We measured clinician attitudes towards genetic testing with the same scale used in the Wagner et al. (7) study, permitting direct comparison of clinician and ALS patient attitudes. Both clinicians and patients had mean scores in the positive range, but responses to individual items revealed significant differences in their perceptions. Clinicians valued the scientific and research potential of genetic testing, but were less likely to say they would have testing themselves, or to see the value in testing for family members. People with ALS were more likely to see the value of genetic testing for themselves and for family members, and less likely to strongly value the scientific potential of testing. Clinicians should be aware that their own attitudes and motivations might be different than those of their patients, as this can influence the way in which genetic testing is introduced and discussed.

In addition to the complexities surrounding the offer of testing, challenges also exist with clinician interpretation of ALS genetic test results. Evidence suggests that many neurologists have a need for support in understanding genetic testing results (8, 9). The high VUS rate on ALS multigene panel tests is a particularly acute challenge (10); although rare variant burden may play a role in the etiology of ALS (11), caution must be used in the approach to VUS in the clinical setting. VUS investigation is particularly difficult in ALS because affected family members are often not available for segregation analysis (12). Incomplete knowledge of genotype-phenotype correlations, penetrance, and technical challenges and lack of sensitivity of commonly used assays for the C9orf72 hexanucleotide repeat expansion test (C9orf72HRE) have posed additional challenges to the incorporation of the offer of genetic testing into standard of care management for ALS.

Although the majority of clinicians in this survey reported that they did offer genetic testing in at least some circumstances, 73.9% (48/65) stated they would be more likely to offer testing if there were consensus guidelines to direct the offer of testing. At this time, there is no consistent approach to the offer of genetic testing to patients with sALS, which represent the majority of ALS patients. As other authors have pointed out, difficulties in obtaining family history information, in addition to inconsistent definitions of fALS, further complicate the situation (13, 2, 4). The universal offer of the C9orf72HRE to all patients with ALS would obviate some of these difficulties (12). ALS genetic testing and counselling guidelines, addressing test indication, interpretation, and counselling may assist clinicians in navigating the challenges of this technology and enable equitable patient access to genetic information.

Limitations

This study has several important limitations. The participant response rate was 31.4% (n=80/255), limiting statistical analysis and generalizability. Respondents with an interest in genetic testing may been more likely to complete the survey, possibly skewing results towards greater clinician participation in genetic testing than actually exists in this group. However, approximately half (48.8%, n=59/121) of NEALS sites were represented by at least one respondent. In addition, results from similar studies (see Table 4) were generally congruent with our findings. Perhaps the most important limitation is that the target clinician group, NEALS, is not representative of the range of clinicians who provide care to ALS patients worldwide. Had a larger and more diverse group of clinicians been surveyed, it is likely that genetic testing practices would be even more variable.

Conclusion

Most academic ALS clinicians offer genetic testing in their clinics; however, there are ongoing inconsistencies in practice that have significant implications for the care of the ‘typical’ ALS patient. Both ALS clinicians and persons with ALS perceive genetic testing positively, although clinicians were less likely to indicate they would have testing themselves, and more likely to strongly value the scientific potential of testing. The findings of this study highlight the critical need for current consensus guidelines for genetic testing and counselling in ALS, particularly as genetic diagnosis becomes more integral to the evaluation and treatment of persons with ALS.