Comparative Performances of 4 Serum NfL Assays, pTau181, and GFAP in Patients With Amyotrophic Lateral Sclerosis

Abstract

Background and objectives: Selecting the most appropriate blood tests is crucial for the management of patients with amyotrophic lateral sclerosis (ALS). This study evaluates the diagnostic and prognostic performance of neurofilament light chain (NfL), glial fibrillary acidic protein (GFAP), and phosphorylated tau 181 (pTau181) biomarkers in ALS to establish their clinical relevance and cutoff values.

Methods: In a cohort of patients from the ALS center in Montpellier, we conducted a head-to-head comparison of 4 different technologies and 3 distinct serum analytes: NfL was tested using the ultrasensitive Simoa and the microfluidic Ella platforms, along with 2 assays recently set up on clinical-grade platforms: Lumipulse and Elecsys. We also used Elecsys to assess serum GFAP and pTau181.

Results: Our cohort included 139 patients with ALS and 70 non-ALS patients, with a mean age of 66.1 ± 11.4 years and 47.4% of women. The mean follow-up was 42 ± 26.3 months for patients with ALS and 141.6 ± 106.3 months for non-ALS patients, with a mortality rate of 85.5% vs 7.7%. There was a high correlation between all methods tested for serum NfL quantification (R² = 0.939 to 0.963). The area under the curve (AUC) for ALS diagnosis was 0.889 (0.827-0.932) for NfL Simoa, 0.906 (0.847-0.944) for Ella, 0.912 (0.853-0.948) for Lumipulse, and 0.910 (0.851-0.946) for Elecsys. Serum pTau181 and GFAP showed poor diagnostic performance with AUCs of 0.565 (0.472-0.649) and 0.546 (0.461-0.636), respectively. Kaplan-Meier survival analysis revealed significant hazard ratios (4.4-5.4) for blood NfL. Patients with ALS had a 40%-50% chance of surviving 50 weeks below the prognostic cutoff values while survival rates dropped to near zero above. NfL and GFAP levels were associated with age and body mass index, considered confounding factors. pTau181 levels varied significantly in patients with ALS depending on the site of onset.

Discussion: This study demonstrates the consistent performance of 4 immunoassays for serum NfL quantification in ALS. NfL showed high diagnostic and prognostic accuracy, making it suitable for individual assessment, unlike GFAP or pTau181. We propose diagnostic and prognostic cutoff values for serum NfL, providing a basis for wider implementation, especially with the clinically accredited Lumipulse and Elecsys platforms, which are becoming standard practice.

Classification of evidence: This study provides Class II evidence that serum NfL levels are useful in identifying over 80% of patients with ALS and predicting survival in patients with ALS compared with pTau181 and GFAP levels.

Figure 1. Simoa, Ella, Elecsys, and Lumipulse NfL Blood Tests Are Highly Correlated

Serum NfL measured with Simoa, Ella, and Lumipulse plotted against Elecsys concentration. Lines of linear correlation with the formula are indicated on the graph along with correlation factor r. NfL = neurofilament light chain.

Figure 2. NfL Blood Tests Diagnose ALS With High Specificity and Sensitivity

(A) ROC and (B) AUC with 95% confidence intervals for ALS diagnosis of serum NfL measured with Simoa, Ella, Lumipulse, and Elecsys; GFAP; and pTau181. ALS = amyotrophic lateral sclerosis; AUC = area under the curve; GFAP = glial fibrillary acidic protein; NfL = neurofilament light chain; pTau181 = phosphorylated tau 181; ROC = receiver operating curve.

Figure 3. NfL Blood Cutoff Values Predict Survival

Kaplan-Meier survival analysis in ALS using the cutoff values in Table 2 for (A) serum Elecsys NfL, (B) Ella NfL, (C) Lumipulse NfL, (D) Simoa NfL, (E) pTau181, (F) GFAP, (G) ALSFRS score, and (H) FVC. Log-rank proportional HR with 95% CI is indicated for each biomarker. ALS = amyotrophic lateral sclerosis; ALSFRS = ALS Functional Rating Scale; FVC = forced vital capacity; GFAP = glial fibrillary acidic protein; HR = hazard ratio; NfL = neurofilament light chain; pTau181 = phosphorylated tau 181.

Figure 4. Bulbar-Onset ALS Has Poorer Prognosis and the Relation Between Onset Sites and Biomarkers

(A) Kaplan-Meier survival analysis in ALS for different sites of onset. Bulbar onset HR compared with upper limb onset is 1.53 (95% CI 0.94–2.49) and with lower limb onset is 1.65 (95% CI 1.04–2.62). Biomarker concentration (pg/mL) by site of onset for serum Elecsys NfL (B), pTau181 (C), and GFAP (D). Only pTau181 showed a significant difference between sites of onset (p values of the Student t test are indicated). ALS = amyotrophic lateral sclerosis; GFAP = glial acidic protein; HR = hazard ratio; NfL = neurofilament light chain; pTau181 = phosphorylated tau 181.

Figure 5. Relationship Between Biomarkers and Variables in All Populations

Forest plots of associations using linear regression between serum Elecsys NfL (A), pTau181 (B), or GFAP (C) and age, sex, BMI, site of onset, ALFRS score, and ALS using linear regression. Unadjusted (red) and age-adjusted means and 95% CIs are provided. ALS = amyotrophic lateral sclerosis; ALSFRS = ALS Functional Rating Scale; BMI = body mass index; GFAP = glial fibrillary acidic protein; NfL = neurofilament light chain; pTau181 = phosphorylated tau 181.