Neurofilament Light Chain Levels, Skeletal Muscle Loss, and Nutritional Decline: Key Prognostic Factors in Amyotrophic Lateral Sclerosis

ABSTRACT

Introduction/Aims

Hypermetabolism and weight loss are established negative prognostic factors in amyotrophic lateral sclerosis (ALS). However, the role of individualized body composition parameters in predicting ALS progression has been underexplored. This study aimed to investigate the correlation between nutritional parameters, neurofilament light chain (NfL) levels, and disease progression in ALS patients.

Methods

The Global Leadership Initiative on Malnutrition criteria were used to define malnutrition in this study. Nutritional status was assessed using body mass index and bioelectrical impedance analysis. The rate of disease progression was defined by the change in the Revised ALS Functional Rating Scale score (ΔFRS). NfL was quantified using single molecule array technology. Spearman's analyses were used to assess correlations.

Results

Sixty of 110 ALS patients were classified as malnourished. There was a strong positive correlation between NfL and ΔFRS (r = 0.71), and a moderate negative correlation with disease duration (r = −0.55). The correlations between NfL and body composition parameters were statistically significant, although weak. NfL levels were significantly higher in fast progressors (p < 0.0001 compared to slow progressors) and in malnourished patients (p = 0.0001). Of the 34 fast progressor patients, 28 (82%) exhibited some degree of malnutrition.

Discussion

Our findings indicate that poor nutritional status, particularly reduced skeletal muscle mass—both independently and in combination with fat mass loss—is associated with elevated NfL levels and faster ALS progression. NfL, combined with nutritional parameters, could serve as a valuable biomarker for disease severity. Further research is warranted to clarify the role of skeletal muscle abnormalities in ALS progression.

Abbreviations

ALS

amyotrophic lateral sclerosis

ALSFRS‐R

amyotrophic lateral sclerosis functional rating scale revised

ASMM

appendicular skeletal muscle mass

ASMMI

appendicular skeletal muscle mass index

BIA

bioelectrical impedance analysis

BMI

body mass index

CVs

coefficients of variation

FFM

fat‐free mass

FFMI

fat‐free mass index

FM

fat mass

FMI

fat mass index

FVC

forced vital capacity

GLIM

Global Leadership Initiative on Malnutrition

H

height

MN

motor neuron

NfL

neurofilament light chain

PA

phase angle

SMM

skeletal muscle mass

SMMI

skeletal muscle mass index

WL

weight loss

ΔFRS

Delta Functional Rating Scale (rate of disease progression)

1. Introduction

Protein–energy malnutrition as well as weight loss are widely recognized as poor prognostic factors in amyotrophic lateral sclerosis (ALS) [1, 2, 3, 4]. However, the role of individualized body composition parameters and indexes in predicting ALS progression has not been studied in detail. Hypermetabolism has been identified as a major cause of nutritional decline influencing the prognosis of the disease [5]. Among ALS patients, the incidence of malnutrition ranges from 16% to 55%, depending on the assessment method, its cut‐off, and the time point of evaluation [6]. Changes in anthropometric parameters, such as body weight and body mass index (BMI) have been hypothesized to play a role in ALS progression. In particular, a decrease in both fat‐free mass (FFM) and fat mass (FM), evaluated through bioelectrical impedance analysis (BIA) has been associated with poor disease prognosis [7].

Additionally, molecular biomarkers could be helpful in guiding clinical diagnosis and/or prognosis. In particular, neurofilament light chain (NfL) has gained attention as a biomarker of disease severity and rate of progression in ALS [8]. However, its relationship with nutritional status or with the metabolic alterations in ALS, which are considered major prognostic factors, has not yet been explored.

The aim of the present study was, therefore, to examine whether nutritional parameters correlate with plasma NfL levels and disease progression in ALS patients.

2. Methods

2.1. Study Design

This was a cross‐sectional study conducted at the IRCCS Istituti Clinici Scientifici Maugeri, Department of Neurorehabilitation of Milan Institute, Italy. Patients were consecutively enrolled based on the following criteria: age over 18 years; diagnosis of ALS according to the revised El Escorial criteria for possible, probable, or definite ALS [9]; forced vital capacity (FVC) greater than 65% of the predicted value. Patients with diagnosis of other neurodegenerative diseases, poorly controlled systemic diseases, or other significant comorbidities, or undergoing mechanical ventilation were excluded. Written informed consent was obtained from each participant. The study was conducted in accordance with the Declaration of Helsinki and was approved by the Ethical Committee of IRCCS Istituti Clinici Scientifici Maugeri, Pavia (approval number: 2773 CE).

For each patient, demographic data and clinical characteristics, including ALS Functional Rating Scale revised (ALSFRS‐R) [10], date and region of symptom onset, anthropometric measurements, and bioelectrical impedance analysis (BIA), were recorded during the same outpatient visit. All data were collected at the baseline visit. The rate of disease progression was calculated using the change in the ALSFRS‐R (ΔFRS) according to the following formula: (48 − score at time of blood sampling)/(months between disease onset and time of blood sampling). ALS patients with ΔFRS < 0.4 points/month were considered slow progressors, those with 0.4 ≤ ΔFRS < 1 points/month were considered moderate progressors, while those with ΔFRS ≥ 1 points/month were classified as fast progressors.

2.2. Nutritional Parameters

Body weight was measured to the nearest 0.1 kg, using a standard balance with individuals wearing light clothes and no shoes. For those patients unable to maintain a standing position, a chair scale was used. Height was measured with a stadiometer if the patient could stand; otherwise, height was estimated from ulna length measurement [11].

BMI was calculated as the ratio between weight (kg) and height squared (meters). Mid‐upper arm circumference was measured on the non‐dominant upper arm at the midpoint between the tip of the shoulder and the tip of the olecranon process. Calf circumference was measured at the maximum circumference of the non‐dominant calf, with the patient seated with the leg hanging loosely after checking for the presence of edema before measurement.

Skinfold thickness of biceps, triceps, subscapular, and suprailiac (both sides of the body) was obtained with a Harpenden caliper. Each measure of skinfold thickness was performed three times and then averaged.

BIA was performed using a phase‐sensitive segmental bioelectrical analyzer (BIA 101 BIVA PRO, Akern, Florence, Italy) with a current of 250 μA at a single frequency of 50 kHz. The BIA was performed with the patients lying supine and with the limbs slightly away from their body, after an overnight fast and bladder voiding. Active electrodes (BIATRODES, Akern Srl, Florence, Italy) were placed on conventional metacarpal and metatarsal lines of the wrist and ankle, respectively. Starting from resistance (Rz) and reactance (Xc) values, we obtained the amount of FM (kg), FFM (kg), skeletal muscle mass (SMM) (kg), appendicular skeletal muscle mass (ASSM) (kg), and phase angle (PA) (°), through the manufacturer equations. The fat mass index (FMI), the free fat mass index (FFMI), the skeletal muscle mass index (SMMI), and the appendicular skeletal muscle mass index (ASMMI) were obtained by dividing the FM, FFM, SMM, and ASMM, respectively, by the subject's height in meters squared.

The percentage weight loss (%WL) was calculated as: (usual weight [considered before first symptom] − actual weight [consultation in nutrition unit])/usual weight. The Global Leadership Initiative on Malnutrition (GLIM) criteria were used to define malnutrition status [12]. According to the GLIM, at least one phenotypic criterion and one etiologic criterion should be present to diagnose malnutrition.

The phenotypic criteria we used were (1) percentage of unintentional weight loss, (2) a BMI less than 20 or 22 kg/m² for those younger or older than 70 years, respectively, and (3) low muscle mass identified as a FFMI less than 17 kg/m² in men and 15 kg/m² in women or an ASMMI less than 7 kg/m² in men and 5.5 kg/m² in women. Etiological criteria included (1) reduced food intake or assimilation and (2) presence of chronic progressive disease usually associated with inflammation. This last etiological criterion was satisfied a priori given that ALS is a chronic, progressive neurodegenerative disease usually characterized by a chronic inflammatory status [13].

2.3. Quantification of Plasma NfL

Blood samples for NfL measurements were collected during the same outpatient visit. Quantitative analysis of NfL in plasma samples was performed by Single Molecule Array (SiMoA) technology on the SR‐X analyzer from Quanterix (Billerica, MA, United States). NfL levels were determined using the commercially available Simoa NFLIGHT v2 Advantage kit (Item 104073, Quanterix), according to the manufacturer's instructions. Briefly, samples, calibrators, and two quality controls of known concentrations were run in duplicate. Plasma samples were run with a 4‐fold dilution, and results were compensated for this dilution. The mean value of the two NfL measurements (pg/mL) was used for statistical analysis. The limit of quantification was 2.56 pg/mL, and the limit of detection was 0.141 pg/mL. A single batch of reagents was used for all samples, and samples with coefficients of variation (CVs) higher than 20% were re‐run. Intra‐ and inter‐assay CVs for all measurements were < 10% and < 16%, respectively.

2.4. Statistical Analysis

Statistical analyses were performed using the GraphPad Prism 8.0.2 software (GraphPad Software LLC, San Diego, California, USA). Descriptive statistics were used to report the demographic, clinical, and nutritional characteristics of the participants. Means with standard deviations (SD) or medians with interquartile ranges (IQR) were used for continuous variables, while frequencies and percentages were used for categorical variables. Normality of continuous variables was assessed using the Kolmogorov–Smirnov test. Detailed information regarding sample sizes and the parametric or non‐parametric tests used are provided in the figure legends. The Spearman correlation test was used to assess correlations between clinical data, body composition parameters, and plasma NfL concentrations. We included correlation analysis instead of multiple regression analysis due to the relatively small sample size. Strong or very strong correlations were defined as those with coefficients of 0.70–1.00, moderate correlations as 0.40–0.69, and weak correlations as 0.10–0.39 [14]. A p value ≤ 0.05 was deemed statistically significant.

3. Results

A total of 115 ALS patients were consecutively enrolled and, after excluding five patients with missing data on key variables (i.e., NfL concentration) 110 were included in the analysis. Participant characteristics are presented in Table 1. According to the GLIM criteria, 54.5% of patients were classified as malnourished. Of these, 33 (30%) had mild to moderate malnutrition, while 27 (24.5%) had severe malnutrition. Additionally, 44 patients (40%) were found to have low muscle mass, according to the ASMMI cut‐off points of the GLIM criteria, with 31 (28.2%) also demonstrating low FFMI, based on the FFMI cut‐off value.

TABLE 1.

Clinical characteristics, anthropometrics and body composition data of the ALS cohort.

Patients (%)	Total	No malnutrition	Malnutrition	p
	110 (100)	50 (45.5)	60 (54.5)
Demografic and clinical data
Sex, female (%)	55 (50)	23 (46)	32 (53.3)	0.4437
Mean age (years)	63.5 (± 12.1)	61.4 (± 11.3)	65.4 (± 12.5)	0.0823
Mean age at onset (years)	59.4 (± 12.2)	57.4 (± 10.19)	60.9 (± 13.5)	0.1225
Spinal onset (%)	88 (80)	43 (86)	45 (75)	0.1510
Bulbar onset (%)	21 (19.1)	7 (14)	14 (23.3)	0.2149
Respiratory onset (%)	1 (0.9)	0 (0)	1 (1.7)	0.3591
ALSFRS‐R	36 [25–41]	39.5 [36–42.5]	28 [22–36]	< 0.0001
Δ FRS (points/month)	0.5 [0.3–1.1]	0.4 [0.2–0.6]	0.9 [0.4–1.3]	< 0.0001
Disease duration (months)	23 [12–41]	22 [12–46]	23 [12–39]	0.7729
Body composition
BMI (kg/m2)	23.35 [20.87–26.15]	25.2 [23.8–29.4]	21.15 [18.2–23.3]	< 0.0001
% WL	2.8 [0.0–11.03]	0.0 [0.0–2.65]	10.44 [1.66–16.79]	< 0.0001
FM (kg)	17.2 [10.4–23.1]	20.25 [13.08–25.75]	15 [9.15–21.15]	0.0035
FMI (kg/m2)	5.9 [3.5–8.4]	6.5 [4.67–9.15]	5.2 [3.2–7.4]	0.0073
FFM (kg)	47.2 [40.5–59.6]	57.05 [45.85–62.45]	41.8 [37.50–51.30]	< 0.0001
FFMI (kg/m2)	17.5 [14.8–19.8]	19.25 [17.8–21]	15.0 [14.15–17.7]	< 0.0001
SMM (kg)	23.8 [17.6–31.1]	29.3 [20.23–32.1]	20.3 [16.25–27.20]	< 0.0001
SMMI (kg/m2)	8.4 [6.8–10.2]	9.65 [7.97–10.78]	7.0 [5.95–8.95]	< 0.0001
ASMM (kg)	17.3 [13.9–23.3]	21.9 [16.68–24.30]	14.4 [12.35–19.55]	< 0.0001
ASMMI (kg/m2)	6.37 [5.31–7.61]	7.33 [6.4–7.87]	5.33 [4.61–6.65]	< 0.0001
PA (°)	4.2 [3.4–5.0]	4.8 [4.28–5.43]	3.7 [2.9–4.3]	< 0.0001
NfL (pg/mL)	49.98 [30.71–78.58]	36.77 [22.48–58.40]	59.27 [36.01–99.69]	0.0001

Note: Normally distributed data are presented as mean (±SD) while median [interquartile range, IQR] is used for variables with non‐normal distribution. Bold values are statistically significant between No malnutrition vs Malnutrition groups.

Abbreviations: ALSFRS‐R, amyotrophic lateral sclerosis functional rating scale revised; ASMM, appendicular skeletal muscle mass; ASMMI, appendicular skeletal muscle mass index; BMI, body mass index; FFM, fat‐free mass; FFMI, fat‐free mass index; FM, fat mass; FMI, fat mass index; NfL, Neurofilament light chain; SMM, skeletal muscle mass; SMMI, skeletal muscle mass index; ΔFRS, rate of change in ALSFRS‐R (rate of disease progression).

Spearman's coefficients exploring the correlates of clinical scores, anthropometrics, body composition parameters and plasma NfL concentrations are presented in Table 2. There was a strong positive correlation between NfL and the rate of disease progression ΔFRS, as well as a moderate negative correlation with disease duration. Moderate correlations were also observed between ΔFRS and both disease duration and percentage of weight loss. Notably, the ALSFRS‐R functional scale displayed a strong correlation with the PA and moderate correlations with fat‐free mass, skeletal muscle mass and appendicular skeletal muscle mass indexes. Analysis using Spearman's R revealed statistically significant correlations between NfL and all body composition parameters analyzed, except for the PA, though the correlations were weak. A sub‐group analysis of ALS patients showed significantly higher amounts of NfL in faster progressor patients and in moderate progressors compared to slow progressors (Figure 1A). Plasma NfL levels were significantly higher in fast progressors than in moderate progressors. When stratifying ALS patients according to nutritional status, we found that both the rate of disease progression and plasma NfL concentrations were significantly higher in individuals with any degree of malnutrition compared to patients without signs of malnutrition (Table 1, Figure 1B). Accordingly, 28 (82%) of the 34 fast progressor patients (ΔFRS ≥ 1) demonstrated some degree of malnutrition, as represented in a scatterplot showing the relationship between ΔFRS and NfL concentration (Figure 2).

TABLE 2.

Spearman's correlation matrix including the clinical features, anthropometrics, body composition parameters, and plasma NfL levels of the patients.

	NfL	Age	Age at onset	ALSFRS‐R	ΔFRS	Disease duration	BMI	%WL	FM	FMI	FFM	FFMI	SMM	SMMI	ASMM	ASMMI	PA
NfL	1.00
Age	0.13 ns	1.00
Age at onset	0.26**	0.96**	1.00
ALSFRS‐R	−0.20*	−0.26**	−0.21**	1.00
ΔFRS	0.71**	0.14 ns	0.28**	−0.45**	1.00
Disease duration	−0.55**	0.08 ns	−0.10 ns	−0.37**	−0.61**	1.00
BMI	−0.30**	−0.10 ns	−0.08 ns	0.34**	−0.26**	−0.01 ns	1.00
%WL	0.39**	0.13 ns	0.21*	−0.34**	0.49**	−0.22*	−0.47**	1.00
FM	−0.28**	−0.11 ns	−0.13 ns	0.03 ns	−0.21*	0.17 ns	0.76**	−0.35**	1.00
FMI	−0.26**	−0.03 ns	−0.07 ns	−0.02 ns	−0.23*	0.24*	0.73**	−0.38**	0.97**	1.00
FFM	−0.26**	−0.24*	−0.17 ns	0.53**	−0.18 ns	−0.27**	0.62**	−0.24*	0.22*	0.07 ns	1.00
FFMI	−0.22*	−0.09 ns	−0.03 ns	0.56**	−0.26**	−0.20*	0.75**	−0.36**	0.20*	0.14 ns	0.87**	1.00
SMM	−0.23*	−0.27**	−0.20*	0.42**	−0.13 ns	−0.23*	0.49**	−0.12 ns	0.10 ns	−0.06 ns	0.96**	0.79**	1.00
SMMI	−0.19*	−0.17 ns	−0.10 ns	0.43**	−0.15 ns	−0.20*	0.58**	−0.19 ns	0.09 ns	−0.02 ns	0.92**	0.90**	0.95**	1.00
ASMM	−0.26**	−0.31**	−0.24*	0.48**	−0.17 ns	−0.25**	0.61**	−0.20*	0.25**	0.08 ns	0.98**	0.82**	0.97**	0.92**	1.00
ASMMI	−0.26**	−0.22*	−0.16 ns	0.52**	−0.24*	−0.19*	0.74**	−0.31**	0.27**	0.16 ns	0.94**	0.95**	0.90**	0.95**	0.94**	1.00
PA	−0.16 ns	−0.25**	−0.19*	0.74**	−0.28**	−0.34**	0.41**	−0.32**	−0.06 ns	−0.13 ns	0.72**	0.78**	0.61**	0.65**	0.67**	0.74**	1.00

Note: ns, not statistically significant; *p ≤ 0.05; **p ≤ 0.01.

FIGURE 1.

(A) ALS patients were stratified according to the rate of disease progression as explained in the Section 2. Median values and interquartile ranges are shown in the violin plots. The nonparametric Kruskal–Wallis test was performed to compare NfL data. (B) ALS patients were stratified based on nutritional status according to GLIM criteria: No malnutrition vs. any degree of malnutrition. Median values and interquartile ranges are shown in the violin plots. The nonparametric Mann–Whitney test was used to compare NfL data.

FIGURE 2.

A scatter graph shows the positive association between the rate of disease progression (ΔFRS) and the levels of NfL. Red squares represent the patients with malnutrition while green circles represent the not malnourished patients. A discontinuous line intercepting y‐axis at y = 1 marks the fast progressors patients (ΔFRS ≥ 1 points/month).

4. Discussion

Our study highlighted an increased prevalence of malnutrition and higher NfL levels among ALS fast‐progressing patients compared to slow and moderate progressors. We also found a strong and significant correlation between the ΔFRS and NfL levels and weak but significant correlations among most of the nutritional parameters. Our results are partially consistent with other studies that have found a moderate correlation between circulating levels of NfL and the ΔFRS in ALS patients [15, 16]. The strong correlation between NfL and the ΔFRS observed in our study could be attributed to the thorough patient characterization, as the ALSFRS‐R was administered exclusively by certified professionals. Interestingly, NfL has recently been associated with sarcopenia determinants, such as muscle mass and strength in older adults, supporting its use as a potential biomarker for the negative impact of sarcopenia [17, 18]. Given the significant impact of nutritional status on disease progression, early identification and treatment of malnutrition are of paramount importance and could potentially improve patient outcomes. Unfortunately, as evidenced by recent studies, current National Institute for Health and Care Excellence (NICE), European, and American ALS nutritional guidelines are not evidence‐based. In fact, these guidelines are only focused on dysphagia and gastrostomy management [3, 19].

The implementation of the BIA method in our study enabled us to identify a percentage of malnourished ALS patients with poor prognosis who would have otherwise gone undetected if nutritional assessment had been based solely on BMI. While BMI is frequently used to evaluate body composition in routine clinical practice due to its ease of implementation, it has several limitations as it considers all body compartments together (one‐compartment model) [20]. BIA analysis, on the other hand, offers a cost‐effective method in ALS patients to evaluate body composition by analyzing separately two body compartments, although BIA equations can be influenced by the hydration status of the patient [4, 21].

In our study, both muscle and fat mass were weakly and negatively correlated with NF‐L and delta FRS in ALS patients. This is in line with previous studies and may be reflective of the extent of weight loss (including both fat and muscle losses) experienced by ALS individuals. In fact, ALS patients often experience significant weight loss because of reduced caloric intake, increased resting energy expenditure, and loss of both FM and FFM. Body mass decrease in ALS may depend on both FM loss and muscle atrophy, as reported in our study. This can be partly due to the increased metabolic demand (i.e., hypermetabolism), a condition widely reported in ALS patients [1]. In fact, when energy and protein intake are not adequate to meet individual demands, body fat and then muscle are catabolized to provide energy, especially in the presence of highly catabolic conditions [22]. Tandan et al. [23], beyond reduced fat mass and fat mass loss, also reported the FFM was associated with poor survival. Indeed, this might be the case (as observed in our study) of a wasting syndrome (loss of muscle with or without loss of fat mass) partly mediated by disease‐related hypercatabolic conditions. In another study, Roubeau et al. [7] showed that weight loss greater than 5% and low PA were poor prognostic factors at the time of diagnosis, while a reduction of both PA and FFM was associated with shorter survival during the follow‐up period.

In our sample, we found a high prevalence of low ASMMI, confirming the significant involvement of limb muscle mass reduction in the ALS population. Mounting evidence suggests that skeletal muscle plays an active role in ALS pathogenesis and is also affected independently of MN influence or denervation [24]. Recently, Dorst and colleagues [25] demonstrated that alterations in body composition and metabolic changes are early events that precede motor neuron degeneration in presymptomatic ALS mutation carriers. Interestingly, these findings support the “dying‐back” hypothesis, which proposes that ALS originates in the peripheral tissues, including skeletal muscle, and a retrograde signaling cascade leads the way to MN death [25].

Our study presents some limitations. First, the relatively small sample size combined with a substantial number of nutritional variables may have compromised the power of the multiple linear regression analysis, making it difficult to detect statistically significant associations between nutritional parameters and NfL. Second, the cross‐sectional nature of our study may limit the generalizability of the results and restrict causal inference. Lastly, the single center experience does not allow us to exclude that specific peculiarities of our setting might influence our findings.

In conclusion, our study advances the understanding of ALS by highlighting the potential role of nutritional status, and in particular of skeletal muscle loss both as a stand‐alone component and in combination with fat mass loss as a contributor to disease severity and progression. The NfL levels, in association with muscle and body composition parameters in future studies on ALS patients, could serve as biomarkers for disease severity and prognosis, providing valuable insights into disease progression. Future larger studies exploring other biomarkers and body composition assessment methods are needed to better elucidate the role of skeletal muscle abnormalities in ALS. This research could pave the way for new therapeutic strategies aimed at improving nutritional status and potentially slowing disease progression in ALS patients.

Author Contributions

Aida Zulueta: conceptualization, investigation, writing – original draft, formal analysis, writing – review and editing, validation, data curation, methodology, visualization, project administration. Rachele Piras: conceptualization, investigation, writing – original draft, writing – review and editing, data curation, validation, methodology, visualization. Domenico Azzolino: conceptualization, investigation, writing – original draft, writing – review and editing, validation, data curation, methodology, visualization. Paola Mariani: data curation, validation, investigation, methodology. Riccardo Sideri: data curation, investigation, validation. Camilla Garrè: data curation, investigation, validation. Giuliana Federico: data curation, validation, investigation. Tiziano Lucchi: writing – review and editing. Paolo Magni: writing – review and editing, supervision. Eugenio Agostino Parati: writing – review and editing, supervision. Christian Lunetta: visualization, writing – review and editing, funding acquisition, supervision, conceptualization, data curation, investigation, project administration, validation, methodology.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.