Serum Neurofilament Light Promising in ALS as Prognostic Indicator of Death

Eric Thouvenot, MD, PhD, of the department of neurology, Centre Hospitalier Regional Universitaire de Nimes

Eric Thouvenot, MD, PhD

According to new study results measuring serum neurofilament (sNfL) levels with a single molecule array in almost 200 patients with amyotrophic lateral sclerosis (ALS), the biomarker may hold strong and independent prognostic value for death in ALS as early as the time of diagnosis.

Led by Eric Thouvenot, MD, PhD, of the department of neurology, Centre Hospitalier Régional Universitaire de Nîmes, the work revealed that along with 6 other factors—older age, bulbar onset, higher ALS Functional Rating Scale&#8208;Revised (ALSFRS) score, greater weight loss, lower maximal inspiratory pressure (MIP), forced vital capacity (FVC)&mdash;higher sNFL levels were statistically predictive of mortality (P <.0001).

As well, weight loss (P = .040) and site at onset, either bulbar or spinal (P = .048), were independent predictive factors of death. A decrease of 1 kg in weight resulted in a 4.9% increased risk of death. Bulbar onset increased the risk of death by 68% compared to spinal onset.

“Several candidate biomarkers have been proposed in ALS, but their clinical relevance remains uncertain. A sensitive and reliable biomarker would be useful for physicians to inform the patient as well as for better evaluating clinical trial outcomes,” Thouvenot and colleagues wrote. “Our results confirm and expand on previous studies showing that sNfL can be reliably measured in serum using the Simoa technology, even at very low concentrations.”

In addition, sNfL levels allowed for significant differentiation of those with ALS from those without, with significantly higher levels in those with disease (71.2 ±4.21 pg/mL) compared to controls (7.22 ±061 pg/mL; P <.0001). Ultimately, sNfL levels above 15 pg/mL were discriminatory of those with and without disease with a sensitivity of 94% and specificity of 100%.

The total study cohort consisted of 207 patients with ALS, of which 198 were analyzed (mean age, 65.4 years), and 21 age-matched controls (mean age, 64.5 years). Patients were followed up out to 185 months (median, 27), during which time 39.4% (n = 78) patients died.

The analysis revealed that an increase of 1 pg/mL in sNfL levels during follow-up corresponded with a 0.74% increase in the risk of death. Thouvenot and colleagues noted that using Kaplan Meier representation of survival suggested that sNfL levels ≥71.2 pg/mL came along with an increased risk of death (hazard ratio [HR], 4.7; 95% CI, 3.0—7.4; P <.0001).

“In line with previous works showing that a late first visit to an ALS center was associated with slower progression, we observed that the later the first visit at the referral center, the lower the sNfL levels,” the authors wrote. “Taken together, these findings indicate that sNfL is a reliable biological marker to predict ALS progression and survival as soon as the first visit in a referral center.”

Similarly, ALSFR scores lower than 42 (HR, 2.63; 95% CI, 1.69—4.01; P = .014) and an age above 66 years (HR, 1.74; 95% CI, 1.13—2.72; P = .014) were predictive of mortality. Along with those, the aforementioned bulbar onset (HR, 2.00; 95% CI, 1.19—3.33; P = .0019), weight loss greater than 2 kg at diagnosis (HR, 2.22; 95% CI, 1.42—3.48; P = .0005), FCV above 85% (HR, 2.17; 95% CI, 1.35—3.51; P = .0015), and MIP under 62 cmH₂0 (HR, 3.57; 95% CI, 2.16—5.89; P <.0001), were also significant predictors of death.

Subgroup analysis of 142 patients for which ALSFRS score was available during follow-up showed that sNfL levels were positively correlated with ALSFRS rate of decline (r = .571; P < 10^-12).

Thouvenot and his co-authors noted that no patients had been exposed to riluzole prior to the study, but all received the treatment diagnosis confirmation. “It would be interesting to determine whether the treatment modifies sNfL levels as this drug is expected to have neuroprotective effects,” they detailed.

The group did identify some limitations to their work, most notably that the study was monocentric, and thus multicentric results would help confirm the findings. Although, they noted that the prospective collection of samples ensured good reproducibility. There was also no inclusion of comorbidity data, despite their known and negative role in ALS prognosis. As well, there was no kinetic determination of sNfL levels as the disease progressed, though literature suggests that sNfL concentrations may evolve over time, at least early on, the group explained.

“It would be interesting to further study the role of such comorbidities on NfL concentrations as the convergent studies on sNfL in ALS prognosis and diagnosis encourages us to use this measure in individual patients,” Thouvenot and company concluded.

REFERENCE

Thouvenot E, Dematteir C, Lehmann S, et al. Serum neurofilament light chain at time of diagnosis is an independent prognostic factor of survival in amyotrophic lateral sclerosis. Eur J Neurol. Published online August 22, 2019. doi: 10.1111/ene.14063.