Electrical Impedance Myography (EIM) in a Natural History Study of C9ORF72 Mutation Carriers.

Abstract

Objective:

Electrical Impedance Myography (EIM) was used to evaluate disease progression in subjects with C9ORF72 expansion mutations and to assess correlations with Medical Research Council (MRC) Scale and revised ALS Functional Rating Scale (ALSFRS-R) measurements. Four types of clinical presentations were assessed; Amyotrophic Lateral Sclerosis (ALS), Frontotemporal dementia (FTD) or other dementia, ALS-FTD, and asymptomatic (ASYMP).

Methods:

Subjects were divided into an ALS Group (ALS/ALS-FTD) and non-ALS Group (FTD/ASYMP) based on initial visit and evaluated at 0, 6, 18, and 30 months with EIM of 4 arm and 4 leg muscles, ALSFRS-R, and MRC scales. The change in EIM from baseline and correlation with the functional scale and strength testing were analyzed.

Results:

EIM 50kHz phase values significantly declined over time in the ALS group (n=31) compared to the non-ALS group (FTD/ASYMP) (n=19). In the ALS group , the decline in EIM was correlated with decline in the ALSFRS-R and MRC scores using within-subject correlations.

Conclusion:

In clinical trials with small populations of genetically associated ALS such as C9ORF-related ALS, EIM may be a useful quantitative biomarker. We did not detect decline in asymptomatic subjects, but longer term studies may detect early changes in this group.

Introduction:

One of the most common genetic defects associated with ALS in the United States is the C9ORF72 expansion mutation, accounting for up to 40% familial ALS and 5-10% of sporadic ALS.^1-3 The mutation consists of a hexanucleotide repeat expansion located in the C9ORF72’s first intron.⁴ The spectrum of clinical manifestations of C9ORF72 gene mutations includes ALS, frontotemporal dementia or other dementia (FTD/dementia), combined ALS-FTD, and asymptomatic carriers (ASYMP).^{5, 6} The development of symptoms is age-dependent, with a wide range of ALS symptom onset from mid-20s to nearly 90 years old and median onset at age 58. Males who develop ALS tend to be younger, while FTD patients and females with ALS develop symptoms at slightly older ages.^{7, 8} In designing clinical trials targeted at ALS patients with the C9ORF72 gene mutation, there is a critical need for biomarkers that are sensitive to change in small population groups.⁹

Electrical impedance myography (EIM) has been studied in ALS and other neuromuscular disorders as a promising biomarker of disease progression. ^10-14 The technique employs a non-invasive high-frequency, low-intensity current to obtain the resistance and reactance values of muscle.¹⁵ The 50-kHz phase measurement is the most commonly used value for assessing disease progression as a function of changes in the muscle attributes.¹⁶ The phase is calculated from the resistance, a measure of the change in electrical current input and output, and reactance, measure of muscle fiber response to electrical charge^{9, 17}. EIM has the advantage that proximal and distal muscles can be quantitively evaluated in a quick and painless manner. Previous studies have shown that EIM phase declines with clinical progression of ALS.¹² EIM also correlated with the revised ALS Functional Rating Scale (ALSFRS-R), a commonly used outcome measure in clinical trials in a combined population of sporadic and familial forms of ALS.¹⁰

This is the first study on the use of EIM to longitudinally follow a population of ALS (ALS and ALS-FTD) and non-ALS (asymptomatic and FTD/dementia) C9ORF72 gene carriers. It provides a unique opportunity to observe EIM changes in response to each group’s different clinical trajectories. Furthermore, we provide evidence that the EIM findings in the ALS group are comparable to the ALSFRS-R and the Medical Research Council (MRC) scale as measures of functionality and strength.

Methods

Overall Study Design

EIM was carried out as part of the clinical protocol, “Natural history and biomarkers of ALS and FTD caused by the C9ORF72 gene mutation” (NCT01925196). The protocol was authorized by the institutional review board at the National Institutes of Health. Informed consent was obtained on all subjects or their legal representative. Participants were eligible to enroll if over age 18 and had a confirmed C9ORF72 gene mutation. They were recruited mailings to ALS and memory clinics and through the National ALS registry (https://www.cdc.gov/als) . The subjects were scheduled for follow-up visits at 6, 18, and 30 months. All subjects had clinical neurologic exam, EMG, MRI-brain, and neuropsychological testing at baseline and at each visit. EIM, MRC ratings, and ALSFRS-R were included in this testing.

Subjects

All subjects had a repeat expansion mutation in C9ORF72 confirmed in a CLIA-certified laboratory. They were evaluated at 0 (baseline), 6, 18, and 30 months by an experienced neurologist and underwent electromyography and cognitive testing to determine their clinical diagnosis. The 2015 revised El Escorial criteria^{18, 19} was used to diagnose ALS and the Rascovsky criteria²⁰ for diagnosis of FTD. These individuals were then grouped into 4 presentations based on their clinical presentation at the initial visit: ALS, ALS-FTD, FTD/Dementia, and ASYMP. All subjects with possible, probable, or definite ALS by El Escorial Criteria, including ALS and ALS-FTD subjects were designated the ALS Group. The non-ALS group did not have evidence of ALS or other motor neuron disease and included the ASYMP, FTD or dementia (non-characteristic for FTD) subjects.

EIM Measurements

EIM was performed on eight muscles on both sides of each subject. For this study, only the 8 muscles on the right side were included in keeping with prior ALS studies and after it was determined that the combination of sides did not significantly affect results. EIM was performed on standardized positions on eight muscles: deltoid (4 cm distal from acromion), triceps (10 cm proximal from olecranon), biceps (4 cm proximal from elbow crease), wrist flexor (6 cm distal from wrist crease), quadriceps (10 cm proximal from patella) , hamstrings (10 cm proximal from knee crease), tibialis anterior (6 cm distal from tibial head), and medial gastrocnemius (6 cm distal from tibial head). All subjects were evaluated with the P/N EIM-014-009 “adult” sensor, Myolex, Inc (Figure 1). Patients were all tested in a supine position with muscle relaxed. Sensor placement and type of sensor were recorded at each visit. The EIM measurements were performed by trained personnel, including an experienced technician, and all under the supervision of one author (TJL). The 8 muscle EIM average (EIM-8MU), the primary outcome measure, was the mean of the 50-kHz phase from all 8 muscles. The 4 Upper Extremity average (EIM-4UE) was the mean of the 50-kHz phase of the 4 arm muscles; deltoid, triceps, biceps, and wrist flexor. The 4 Lower Extremity Average (EIM-4LE) was the mean of the 50-kHz phase of the 4 leg muscles; quadriceps, hamstrings, medial gastrocnemius, and tibialis anterior.

Figure 1:

Electrical Impedance Myography sensor. A. P/N EIM-014-009 “adult” sensor, Myolex, Inc, used for testing all subjects. B. Application of the EIM sensor to the deltoid muscle.

MRC Scale

The MRC Scale^{21, 22} was used to grade the strength of each of the 8 muscles assessed. The upper extremity muscles evaluated were the deltoid, biceps, triceps and wrist flexors and the lower extremity muscles were the quadriceps, hamstrings, foot flexors and extensors. The MRC-8MU was the total MRC scale of all 8 muscles observed. The MRC-4UE was the total MRC score of the 4 arm muscles and the MRC-4LE was the total of the 4 leg muscles.

ALSFRS-R

The ALSFRS-R was used as a clinical measure to evaluate bulbar, fine motor, gross motor, and respiratory function.¹⁹ The ALSFRS-R total was the score from the 12-item questionnaire for a maximal normal score of 48 points. The ALSFRS-R Fine motor subscore was the score of the 3 items evaluating fine motor function—handwriting, dressing and hygiene, and cutting with utensils. The ALSFRS-R Gross motor subscore was the score of the 3 items evaluating gross motor function—turning in bed, walking, and climbing stairs. The ALSFRS-R fine motor and gross motor subscores were used as surrogates for upper extremity (fine-motor) and lower extremity (gross motor) functions.

Statistics

For baseline data, two-sample t-test was used to assess differences between ALS and non-ALS group for EIM measurements and age. A Wilcoxon two-sample t-test was applied to MRC and ALSFRS-R measurements. Fisher’s exact test was performed to assess the difference in sex proportions between ALS and non-ALS groups. Significance was defined as p< 0.05.

For the longitudinal data, the relationship between EIM and MRC/ALSFRS-R was examined using correlation between subjects and within subjects²³ for each EIM measurement (EIM-8MU, EIM-4UE, EIM-4LE) separately. For the correlation between subjects, Spearman correlation coefficient was calculated using baseline data. Analysis of covariance (ANCOVA) ²³ was performed to estimate the correlation coefficient within subjects. A random coefficient model for longitudinal data was applied to evaluate the change of EIM measurements over 30 months, where the subjects with only baseline data were excluded. The model involved a random intercept and slope for each subject. The random coefficient model contained group, time effect and the interaction effect between group and time. The significant interaction indicated that the slope in ALS group was different from that in non-ALS group. Then, the null hypothesis of β (slope)=0 was tested for each group. Age and gender could not be used as covariates because of the disparity between the two groups.

The statistical analysis was performed using SAS 9.4. For MRC, and ALSFRS-R measurements, Box-Cox transformed data were used in ANCOVA.

Results

Demographics

Fifty subjects with genetically confirmed C9ORF72 were evaluated at baseline. There were 31 subjects in the ALS group (24 ALS, 7 ALS-FTD) and 19 subjects in the non-ALS group (4 FTD/dementia and 15 ASYMP) (Table 1). The non-ALS group had one subject with dementia, atypical for frontotemporal dementia. The ASYMP subjects in the non-ALS group were significantly younger than the ALS group, whereas the FTD/dementia patients were similar in age to the ALS group. The predominant initial presentation was bulbar (54%) with a median duration of symptoms of 18.5 months (range 4.4-120.0 months). The ALS group had significantly more men (65%) than the non-ALS group (32% male). Five subjects did not have a baseline EIM evaluation, three in the ALS group and two in the non-ALS group. In the ALS group, there was a drop-off in follow-up visits due to disease progression. Seven ALS patients had 3 visits, 14 had 2 visits and 10 ALS patients had only a baseline visit. In the non-ALS group, all subjects had at least 2 visits with 63% having 3 or more visits. No subject switched from the non-ALS group to the ALS group during the study period.

Table 1.

Baseline Demographics and Functional Measure Results.

Variable		ALS Group (N=31)	Non-ALS Group (N=19)	p-value
Age	Mean±SD	56.8 ± 8.6	47.6± 12.4	0.0033 a
	(Range)	(36-71)	(29-73)
Gender	F:M	11:20	13:6	0.0404b
EIM-8MU	Mean±SD	6.7 ± 2.2	6.9 ± 2.3	0.7716a
	Range	(3.0-12.4)	(3.8-12.2)
ALSFRS-R Total	Median/IQR	39/13	48/1	<0.0001c
	Range	12-46	43-48
MRC 8MU Total	Median/IQR	39/8	40 / 0
	Range	13-40	40-40	0.0002c

SD-standard deviation, F-female, M-Male, interquartile range (IQR)

^a: two-sample t-test

^b: Fisher’s exact test

^c: Wilcoxon two-sample t-test

Baseline EIM, MRC Scale and ALSFRS-R

At baseline, there were no significant differences between the ALS and non-ALS groups for EIM-8MU, EIM-4UE or EIM-4LE (Table 1). Since sex was significantly associated with diagnosis group (more males in the ALS group), and EIM measure (lower value for females), subgroup analyses based on sex also showed that there was no significant difference in the EIM-8EU mean between ALS and non-ALS groups. Age had non-significant effect on EIM measure (Pearson r=0.019 for non-ALS, r=-0.292 for ALS, p>0.1) and was not considered a covariate.

The ALSFRS-R scale differed between the two groups for ALSFRS-R Total score, and ALSFRS-R Fine motor and ALSFRS-R Gross motor subscores. The non-ALS group had 5 subjects with ALSFRS-R scale slightly below 48; 3 subjects had low scores related to dementia and 2 ASYMP had scores of 47 related to minor physical disabilities. The MRC scale also showed significant differences between the two groups in the MRC-8 MU, MRC-4UE and MRC-4LE.

In the baseline testing of the ALS group, there was a significant correlation between EIM-4LE and MRC-4LE (Spearman’s r=0.557, p=0.0021, n=28), and between EIM-4LE and ALSFRS-R Gross motor subscore (Spearman’s r=0.510, p=0.0056, n=28). The subjects with low values of EIM also tended to have low values of MRC and ALSFRS-R though correlations were not significant for EIM-8MU and EIM-4UE.

Change in EIM over time

Comparison between the ALS and non-ALS groups using the random regression coefficient model analysis found that the slope in ALS group was different from that in non-ALS group, based on the test of interaction between group and time, for EIM-8MU (p=0.0142) and EIM-4UE (p = 0.0118), and EIM-4LE (p = 0.04). The EIM measures declined over time in the ALS group but not in the non-ALS group. The negative slope for the ALS group was significantly different from zero for EIM-8MU (p=0.0029, slope = −0.074, 95%CI[−0.0119, −0.028]), EIM-4UE (p= 0.0024, slope = −0.086, 95%CI[−0.138,−0.034]), and EIM-4LE (p=0.0125, slope = −0.023, 95% CI[−0.215,−0.070]) (Figure 2). Five of the bulbar onset ALS subjects did not decrease in EIM-8MU over the course limited to two visits but all of the lower and upper extremity onset ALS subject did decrease. The slope in non-ALS group was not significantly different from zero for these same three measures. Residual analysis identified two outliers (asymptomatic subjects from the non-ALS group) had EIM-MU increasing over time. By excluding the two outliers, little change was shown in the slopes. Of the four FTD/dementia subjects in the non-ALS group, two subjects had a slight decline in EIM values but no clinical evidence nor electrodiagnostic evidence of ALS. Exploratory analysis for individual muscles showed that the deltoid, wrist extensor, and hamstring muscles had significant differences in slope between ALS and non-ALS group based on p-value of interaction between group and time (deltoid p= 0.015, wrist flexors p= 0.0009, hamstrings p = 0.04)

Figure 2:

Comparison EIM slope for ALS Group and Non-ALS Group using random regression coefficient model analysis. Negative slope was significantly different from zero in the ALS group; A. EIM-8MU (p=0.0029), B. EIM-4UE (p= 0.0024), and C. EIM-4LE (p=0.0125) , but not in the non-ALS group.

Relationship between EIM and Standard Functional Measures over Time

Using ANCOVA analysis of all available time-points (Table 2), ALS group showed the decrease in the EIM-8MU over the 30 months had a significant association with decreases in MRC-8MU (r=0.379, p=0.0391) and ALSFRS-R total score (r=0.528, p=0.0027) within subjects. Also, a decrease in the EIM-4UE was associated with a decrease in MRC-4UE (r=0.488, p=0.0063); a decrease in the EIM-4LE was associated with a decrease in ALSFRS-R Gross motor subscore (r=0.392, p=0.0323).

Table 2.

Within subject correlations between EIM and Functional Measures

		ALS group		Non-ALS group
EIM		coefficient	p-value	coefficient	p-value
8MU	MRC-8MU	0.379	0.0391	0
	ALSFRS-R Total	0.528	0.0027	0.164	0.3194
4UE	MRC-4UE	0.488	0.0063	0
	ALSFRS-R Fine motor	0.313	0.0926	0.248	0.1282
4LE	MRC-4LE	0.055	0.7730	0
	ALSFRS-R Gross motor	0.392	0.0323	0.005	0.9743

ALS-amyotrophic lateral sclerosis, EIM – electrical impedance myography, MRC -medical research council, MU-muscle, UE upper extremity, LE-lower extremity, ALSFRS-R- Amyotrophic lateral sclerosis Functional Rating Scale-Revised

Discussion

In this study, we had the unique ability to compare the ALS group with carriers of the C9ORF72 gene mutation that were not currently exhibiting signs of ALS and to appreciate the divergence of findings between the ALS and non-ALS group. EIM showed a significant change from baseline during the study period in the ALS group compared to the non-ALS group in the EIM 8MU, EIM-4UE, and EIM- 4LE. Bulbar onset ALS subjects may be less likely to show a decline in the EIM, possibly related to their symptoms sparing the limbs to a greater extent that ALS beginning in the arms or legs. The decline in EIM-8MU over time was correlated with the ALSFRS-R, a functional scale, and with strength, graded using the MRC scale, thus validating the EIM as a biomarker of progression in C9ORF72 ALS. When the EIM for upper and lower extremity were separately assessed, the EIM-4UE correlated with changes in the MRC-UE score but not the ALSFRS-Fine motor subscore, whereas the EIM-4LE correlated with changes in the ALSFRS-R Gross motor subscore but not the MRC scale. One explanation for the lack of correlation between the UE-EIM and the ALSFRS-R Fine motor subscore may be due to the ALSFRS-R subscore predominantly evaluates hand dexterity and EIM was measured from more proximal upper extremity muscles. In the non-ALS group, two FTD/Dementia had a decline in their EIM with no clinical or electrodiagnostic evidence of ALS, but we don’t know if this would have been predictive of ALS developing in the future or reflect sarcopenia in a compromised individual. In the asymptomatic subjects, we did not appreciate a change in the EIM that would predict a conversion from non-ALS to ALS though a longer follow-up may have been needed. The asymptomatic subjects tended to be younger than the other subjects and in overall good health which would made the development of ALS in the near future less likely.

There are several limitations of the study. The relatively small sample of patients compounded with the high number of dropouts limited the analyses, particularly after the first 6 months. Dropouts were largely due to disease progression or death in C9ORF72 carriers with ALS, ALS-FTD, and FTD, and thus it is likely that the cohort with the longer follow-up reflects patients with slower progression. Age and sex differences were significant in the two diagnostic groups, with more men and older women in the ALS group. Since age with little effect on EIM-8MU, only sex was considered as a covariate. Subgroup analysis based on sex did not show significant differences in the EIM-8MU mean between the ALS and non-ALS groups. Factors such as muscle bulk and subcutaneous fat affect EIM measurements, with generally higher EIM phase values in men compared to women,²⁴ while declining muscle mass with age or disease decreases EIM phase values. ²⁵ The similarity in the baseline EIM of both groups may be reflective of the ALS group disease severity at baseline even though it was male predominant with an expectant higher EIM. Lastly, although all subjects were evaluated with the same EIM sensor model and measures were repeated on the same muscles of the same limb to limit variability, slight variations in sensor placement and pressure on sensor could have affected the results.

This paper has parallel findings to earlier ALS studies with EIM, but it is the first study to show that EIM has the potential to be used as a biomarker to track the disease progression in a small patient population of ALS patients with the C9ORF72 gene mutation. The use of 8MU-EIM had more robust correlations with the 8-MU MRC and full ALSFRS-R scale than the UE-EIM and LE-EIM, suggesting that it may be the preferential parameter to use in clinical trials. These correlations also support the validity of EIM compared to a standard ALS functional scale and clinical strength testing. Though no changes in EIM were observed in the non-ALS group, it has the potential to be used as an early marker for onset of ALS in people with familial ALS if evaluated over a longer time period.

Abbreviations:

4 LE

Four muscle lower extremity average

4 UE

Four muscle upper extremity average

8 MU

Total eight muscle average

ALS

Amyotrophic lateral sclerosis

ALSFRS-R

Amyotrophic lateral sclerosis Functional Rating Scale-Revised

ASYMP

Asymptomatic

EIM

Electrical Impedance Myography

FTD

Frontotemporal Dementia

MRC

Medical Research Council