Skip to main content
NCBI home page
HHS Author Manuscripts logoLink to HHS Author Manuscripts
. Author manuscript; available in PMC: 2025 Sep 23.
Published in final edited form as: Biomark Med. 2025 Sep 16;19(17):813–823. doi: 10.1080/17520363.2025.2562546

Trends in clinical studies evaluating neurofilament light chain as a biomarker

Sydney Stern a,*, Kristin Jamenis a,*, Sreedharan Sabarinath a, Shawna L Weis b, Robert Schuck a, Michael Pacanowski a
PMCID: PMC12453014  NIHMSID: NIHMS2111643  PMID: 40955802

Abstract

Introduction:

The use of biomarkers as surrogate endpoints, supported by strong mechanistic, epidemiologic, and/or clinical data, provides drug development programs with endpoints that predict clinical benefit and may be more sensitive to drug effects than clinical endpoints. Neurofilament light chain (NfL) is a marker of neuroaxonal injury that has emerged as a promising biomarker for neurodegenerative diseases.

Methods:

We identified Investigational New Drug programs submitted to the U.S. Food and Drug Administration between 2005–2024 that proposed to use NfL as a pharmacodynamic biomarker, biomarker for patient selection or stratification, and/or surrogate endpoint for accelerated approval.

Results:

A total of 50 programs were identified with most from the last five years. Of the 50 programs, 94% (n = 47) proposed NfL as a pharmacodynamic biomarker, 8% (n = 4) for patient selection, 52% (n = 26) for patient stratification, and 20% (n = 10) as a surrogate endpoint. Of the programs evaluating NfL as a pharmacodynamic biomarker with available data on NfL levels (n = 21), approximately 50% reported NfL changes that correlated with drug exposure.

Conclusion:

This analysis highlights the important role that NfL plays in clinical trials and identifies future areas of research and study design considerations to strengthen the support of NfL as a biomarker.

Keywords: Biomarker, neurofilament light chain, pharmacodynamic, surrogate endpoint, clinical trials, drug development

1. Introduction

Biomarkers are increasingly used to improve disease diagnosis, prognostication, and monitoring of therapeutic response. In drug development, particularly for rare diseases, biomarkers are often used in clinical trials to evaluate treatment response. Such biomarkers may be used to predict clinical benefit, guide dosing, or detect safety issues [1]. Response biomarkers may be subcategorized as pharmacodynamic (PD) biomarkers or surrogate endpoint biomarkers [1]. A PD biomarker is a response biomarker that indicates biological activity of a medical product without drawing conclusions about efficacy or disease outcome or linking biological activity to a mechanism of action [1]. If preliminary evidence demonstrates that a biomarker is altered by drug exposure, the evidence can then be used to inform dose selection and potentially serve as confirmatory evidence when supported by mechanistic evidence. Surrogate endpoint biomarkers are “a substitute for a direct measure of how a patient feels, functions and survives” and can be used to more quickly determine whether a drug is effective, thereby accelerating product development following regulatory determination that the surrogate biomarker is sufficiently validated to capture drug efficacy [1]. Validation of surrogate endpoints generally requires clear evidence that the surrogate predicts clinical benefit. Surrogate endpoints should be specific to their context of use and, while a particular therapy may improve a surrogate endpoint in one setting, improvement of that same surrogate endpoint in a different setting may be misleading [1,2].

First described as a measurable biomarker in 1996, neurofilament light chain (NfL) has emerged as a promising biomarker of neuroaxonal damage in a wide variety of neurological conditions, ranging from neurodegenerative diseases to inflammatory and traumatic conditions [3]. Neurofilaments are neuronal scaffolding proteins of the cytoskeleton composed of four subunits including NfL (68 kDa), neurofilament medium chain (150 kDa), neurofilament heavy chain (190–210 kDa), and either α-internexin in the central nervous system (CNS) or peripherin in the peripheral nervous system [3–5]. Importantly, NfL is the most abundant and soluble subunit, allowing for broader distribution and easier measurement of NfL compared to other subunits [6]. Under normal physiological conditions, and in response to aging, NfL is released from neurons into the cerebrospinal fluid (CSF) [7–9]. Furthermore, NfL is released continuously and at low levels in healthy individuals under normal conditions. However, following axonal injury secondary to neurodegeneration or other neurological conditions, NfL is released into the extracellular space and eventually into the blood, where it remains elevated (Figure 1), and it can be sampled either in the blood or CSF. The exact mechanism by which NfL is released into the CSF and blood is unknown. The mechanistic associations between NfL and pathogenic neurodegenerative processes make it a good candidate for assessing drug-related reductions in neuronal injury. Although the clinical applications are not established, elevated NfL levels have been used to detect neurological conditions in patients prior to symptom onset [10–12], which for many rare diseases, offers an opportunity to identify patients at risk and intervene prior to disease manifestation. Given the irreversible nature of axonal damage, monitoring relevant axonal biomarkers such as NfL may also provide insight into disease status and progression, enabling physicians to potentially prevent or slow further neurological injury [13,14].

Figure 1.

Figure 1.

Open in a new tab

NfL release from brain neurons into CSF and blood. (a) This figure depicts a schematic of NfL release from neurons of a healthy, age-matched brain. Under normal physiological conditions, such as in response to aging, NfL is released from neurons into the CSF in the central nervous system and into the extracellular space and blood. (b) Upon neuroaxonal injury secondary to a multitude of potential neurological conditions, NfL release into the CSF and blood is increased. Created in Biorender.com.

The U.S. Food and Drug Administration (FDA) relied on NfL as a surrogate endpoint that is reasonably likely to predict clinical benefit in the accelerated approval of tofersen for the treatment of amyotrophic lateral sclerosis (ALS) in adults who have a mutation in the superoxide dismutase 1 (SOD1) gene (SOD1-ALS) [15]. Approved in 2023, tofersen is an intrathecally administered antisense oligonucleotide that elicits degradation of SOD1 mRNA and subsequently reduces the synthesis of wild-type and pathogenic SOD1 protein [15–17]. In the pivotal trial, a statistically significant reduction in plasma NfL concentrations was observed at Week 28 in the tofersen group compared to the placebo group (p < 0.0001) [17]. A statistically significant reduction in CSF SOD1, further verifying the mechanism of action of tofersen, was also observed at Week 28 in the tofersen group compared to the placebo group (p < 0.0001) [18]. The following factors were considered in the evaluation for accelerated approval: 1) mechanistic evidence that tofersen reduced SOD1 protein, the intended target of the drug and a known contributor to the pathophysiology of disease, 2) scientific evidence demonstrating the prognostic value of plasma NfL in predicting disease progression and survival in ALS, and 3) the observed correlation between the reduction in NfL and the reduction of decline in a clinical outcome. These findings led to the determination that the reduction in plasma NfL concentrations could serve as a surrogate endpoint that is reasonably likely to predict clinical benefit in adult patients with SOD1-ALS [18].

While tofersen is the first drug approved based on changes in plasma NfL concentrations, NfL has also been implicated in other diseases such as Alzheimer’s disease, multiple sclerosis, and Parkinson’s disease [19–21]. However, several properties of NfL have limited broader use to inform treatment and drug development decisions. In particular, because NfL is a nonspecific biomarker of neuroaxonal damage, concentrations of NfL vary as a function of multiple patient characteristics (e.g., age, immunoglobulin levels), and the relationship between the pathophysiology of some neurological diseases and CNS NfL concentrations is unclear [8,9,22]. The magnitude of change in NfL that translates into a clinically meaningful outcome is unknown, and whether this magnitude of change, if it can be defined, varies across neurodegenerative diseases is also unknown.

To better understand how NfL is being used in drug development, we sought to better understand trends in how drug developers design clinical studies to evaluate NfL as a biomarker. This was accomplished by surveying clinical protocols derived from Investigational New Drug (IND) programs that proposed to measure NfL levels. We then sought to elucidate the role of NfL in the clinical trials and examine the respective study design characteristics, with a focus on specific elements of interest including 1) study population, 2) route of administration and product type, 3) study design and objectives, 4) matrix in which NfL was measured, 5) PD assays, 6) and the overall role of NfL as a PD biomarker, biomarker for patient selection or stratification, and/or surrogate endpoint.

2. Methods

We reviewed IND programs for investigational therapies submitted to the Center for Drug Evaluation and Research (CDER) at the U.S. FDA between June 2005 and June 2024 (Figure 2). The first protocol submission that we identified as proposing to use NfL as a response biomarker was from 2010. The IND programs were identified with an internal FDA query platform using the following search terms and Boolean operators: “neurofilament light chain OR NfL AND (biomarker OR surrogate endpoint OR reasonably likely OR pharmacodynamic).” We excluded IND programs that were in the pre-submission stage, on clinical hold, terminated, inactive, or withdrawn at the time of data collection. Additionally, we included only INDs that were intended for commercial drug development. The survey did not include therapeutics submitted only to other regulatory agencies outside of the U.S. FDA.

Figure 2.

Figure 2.

Open in a new tab

Flowchart of investigational New drug (IND) programs included in this analysis. Using an internal FDA platform, terms “neurofilament light chain” or “NfL” and (biomarker or surrogate endpoint or reasonably likely or pharmacodynamic) were used to identify relevant programs. IND programs were captured between June 2005 and June 2024. Eighty-six IND programs were identified, and the exclusion criteria was applied (n = 36). Fifty IND programs were eligible for assessment for the use of NfL as a biomarker for patient selection or stratification, pharmacodynamic biomarker, and/or surrogate endpoint.

We considered NfL a biomarker for patient selection or patient stratification (including subgroup analyses) if it was used to enroll or group based on identification of patients that 1) may be presymptomatic and expected to have disease conversion or 2) have a higher likelihood of neurodegenerative progression. The authors did not make a determination about the adequacy of NfL as a PD biomarker, biomarker for patient selection or stratification, or surrogate endpoint for the purposes of this assessment. Such determinations, if made, were under the purview of the appropriate review divisions and outside the scope of this review. Therapeutic classifications of IND programs were determined by etiology of disease (e.g., inborn error of metabolism) and target indication area (i.e., assigned FDA review division).

Our investigation focused on study design elements of clinical trials in which NfL concentrations were proposed to be assessed, regardless of the disease indication. Information was extracted from clinical study protocols, meeting packages, and other background information available to the FDA. Multiple features of the phase 1–3 studies were collected, including whether NfL was to be assessed as a primary, secondary, or exploratory objective; trial design, including study population and inclusion criteria related to NfL concentrations at enrollment; route of administration, dosage form, and product type; dosing regimen; and biomarker assessment plan, such as sample collection timepoints and what other biomarkers were assessed and in which matrices.

Dosing regimens for each clinical trial protocol were classified as follows: 1) single ascending dose (SAD) if the subjects received a single administration of an investigational drug in ascending dose cohorts without regard to randomization or blinding; 2) multiple ascending dose (MAD) if subjects received multiple administrations of an investigational drug and ascending dose cohorts without regard to randomization or blinding; 3) dose escalation or titration and parallel group designs if the study subjects received increasing doses followed by a continuation period for the remaining duration of the treatment or received one of two fixed doses in parallel cohorts; 4) randomized, double-blind, placebo-controlled (RCTs) if subjects were randomly assigned to receive an investigational drug or a placebo, and neither the investigators nor patients knew whether they received the investigational drug or the placebo; 5) seamless design with a dose-finding study (e.g., SAD/MAD) and a RCT evaluating clinical efficacy in the same trial to evaluate pharmacokinetics, optimize dose, and assess clinical efficacy; and 6) other trial design approaches (e.g., maximum tolerable dose or single arm, open-label studies).

3. Results

A total of 86 IND programs, for distinct investigational agents, that proposed to measure NfL were submitted to the FDA between June 2005 and June 2024. Of these, 50 (58.1%) used NfL as a PD biomarker and/or biomarker for patient selection or stratification, had evaluable protocol information, and met inclusion criteria for this study. Although IND programs dating from 2005 were assessed, investigation of NfL as a treatment response biomarker was not noted in a clinical trial protocol until 2010. The 50 IND programs included 106 relevant clinical trial protocols dating from May 2010 to April 2024. Of the 106 clinical trial protocols that planned to assess NfL, the majority were from the last five years (57.5%, n = 61). In contrast, 34.9% (n = 37) of protocols dated from 2015 to 2019, and only 7.5% (n = 8) predated 2015 (Figure 3).

Figure 3.

Figure 3.

Open in a new tab

IND clinical trial protocols as a function of time. Bar chart shows IND clinical trial protocols for which NfL was assessed (n = 106), divided into four, five-year increments based on the date of original protocol submission to the FDA.

3.1. Indications

Out of 50 IND programs that proposed to use NfL as a PD biomarker and/or biomarker for patient selection/stratification, 84.0% (n = 42) were pursuing development for treatment of neurological conditions, 10.0% (n = 5) for treatment of inborn errors of metabolism, and 6.0% (n = 3) for other conditions (Figure 4(a)). Twelve percent of IND programs (n = 6) included pediatric patients, and the majority of IND programs (78.0%, n = 39) explored a rare disease indication. Among the 42 IND programs pursuing treatment of neurological conditions, the three most common diseases were ALS (33.3%, n = 14), Huntington’s disease (11.9%, n = 5), and progressive supranuclear palsy (9.5%, n = 4) (Figure 4(b)).

Figure 4.

Figure 4.

Open in a new tab

Characterization of IND programs. Pie charts illustrate the percentage of (a) IND programs (n = 50) by therapeutic classification, (b) IND programs (n = 42 IND programs) by indication, and investigational therapies (n = 50) by (c) route of administration and (d) product type. ALS, amyotrophic lateral sclerosis; HD, Huntington’s disease; MS, multiple sclerosis; PSP, progressive supranuclear palsy; IV, intravenous; IT, intrathecal; SC, subcutaneous.

3.2. Route of administration and product type

Further assessment of the 50 IND programs revealed that the two most commonly used routes of administration for investigational therapies were oral (40.0%, n = 20) and intravenous (IV; 34.0%, n = 17) routes, followed by intrathecal (IT; 14.0%, n = 7), subcutaneous (SC; 8.0%, n = 4), and other (4.0%, n = 2) routes (Figure 4(c)).

The most common product types were small molecules (52.0%, n = 26), followed by biologics (34%, n = 17) and RNA-based therapeutics (14.0%, n = 7) (Figure 4(d)). Of the RNA-based therapeutics, the majority were antisense oligonucleotides intended to degrade a target pathogenic mRNA.

3.3. NfL measurements

The 106 clinical trial protocols included in the 50 INDs detailed that NfL was to be collected in plasma (39.6%, n = 42), serum (34.9%, n = 37), or CSF (53.8%, n = 57), and 9.4% (n = 10) were unspecified in blood. Sixteen percent (n = 17) of protocols planned to collect only a single matrix, most often plasma (47.0%, n = 8). Bioanalytical assay information was not provided for the majority of IND programs.

3.4. Trial characteristics

Out of 106 total clinical trial protocols that proposed to evaluate NfL, 17.0% (n = 18) were phase 1 trials, 36.8% (n = 39) were phase 2 trials, 29.2% (n = 31) were phase 3 trials, and the remaining 17.0% (n = 18) were combined phases (e.g., seamless phase 1/2 or 2/3 trials; n = 10). Furthermore, 5.7% (n = 6) of the clinical trial protocols were SAD, 7.5% (n = 8) were MAD, 2.8% (n = 3) were combined SAD and MAD, 9.4% (n = 10) were dose escalation studies with continuation arms or parallel groups, 34.0% (n = 36) were RCTs with a primary objective of clinical efficacy, and 31.1% (n = 33) were other trial designs. Approximately nine percent (n = 10) of the clinical trial designs were seamless designs, and of these 10 seamless trials, 80% (n = 8) were pursuing development for treatment of rare diseases.

Among 50 IND programs, 14% (n = 7) included an assessment of NfL concentrations as a primary objective in at least one clinical trial protocol. Forty-six percent (n = 23) and 74.0% (n = 37) of programs included an assessment of NfL concentrations as a secondary or exploratory objective in at least one protocol, respectively. Out of 106 clinical trial protocols, only 3.8% (n = 4) of protocols classified NfL as a primary objective. In contrast, 27.4% (n = 29) of protocols classified NfL as a secondary objective, and 55.7% (n = 59) classified NfL as an exploratory objective. The remaining protocols (13.2%, n = 14) classified NfL under multiple objective categories (e.g., secondary, exploratory, primary and secondary, etc.). All (n = 4) of the protocols for which NfL was included as a primary objective were phase 2 trials (Figure 5(a)). Phase 2 (37.9%, n = 11) and phase 3 (31.0%, n = 9) trials were the two most common trial phases for the 29 protocols that classified NfL as a secondary objective. Further examination of the 59 protocols classifying NfL as exploratory revealed that 25.4% (n = 15) were designated as phase 1 trials, 37.3% (n = 22) were phase 2 trials, 25.4% (n = 15) were phase 3 trials, and 11.9% (n = 7) were seamless trials.

Figure 5.

Figure 5.

Open in a new tab

Proposed evaluations of NfL as a biomarker in clinical trials. (a) Bar chart shows the number of IND clinical trial protocols (n = 92) for which NfL was evaluated as primary, secondary, or exploratory objective based on clinical trial phase. The others (n = 14) listed NfL under multiple objectives in the same protocol (e.g., both primary and secondary objectives). Pie charts illustrate the percentage of IND programs (n = 50) for which NfL was proposed as a (b) patient selection, (c) patient stratification, or (d) surrogate endpoint biomarker.

3.5. NfL for patient selection or stratification

Eight percent of programs (n = 4) proposed use of NfL as a biomarker for patient selection in at least one study (Figure 5(b)). For instance, several programs noted that, to be eligible for enrollment into a clinical trial, baseline NfL concentrations had to be elevated over a pre-specified threshold.

We assessed whether NfL was evaluated as a biomarker for patient stratification (including subgroup analyses) across protocols in IND programs. Out of 50 programs, 52% (n = 26) proposed leveraging NfL concentrations to identify patients: 1) who have a higher likelihood of neurodegenerative progression, supported by literature evidence, mechanistic data, or clinical trial data; or 2) who are more likely to experience a clinical response as an exploratory endpoint (Figure 5(c)).

3.6. NfL as a treatment response biomarker

Evaluation of NfL as a PD biomarker was proposed in at least one clinical trial protocol for 47 of 50 (94%) IND programs. Of the programs evaluating NfL as a PD biomarker with data available (n = 21 out of 47), NfL exhibited changes in response to treatment in 52.4% (n = 11) of the programs, while the rest of the programs did not have available results to make a conclusion. The change in NfL was drug developer-reported; the magnitude of the change in NfL and clinical meaningfulness of treatment-mediated changes in NfL were not evaluated.

Assessment of the relationship between drug exposure and change in NfL levels from baseline to a prespecified timepoint was proposed in at least one clinical trial protocol for 30% (n = 15) of the 50 IND programs. In addition, almost half of programs (46%, n = 23) included at least one clinical trial that proposed to associate changes in NfL concentration with a clinical outcome measure. Findings in these clinical trials were largely unknown either because the clinical trial is ongoing or the results have not been submitted.

3.7. NfL as a surrogate endpoint

Throughout the 50 IND programs, NfL was proposed as a surrogate endpoint in protocols by drug developers, meeting background packages, and IND general plans in 20% of programs (n = 10) (Figure 5(d)). The FDA review division’s assessment of the adequacy of NfL as a surrogate endpoint, or the concurrence from the FDA regarding the use of NfL as a surrogate, was not included in this analysis.

3.8. NfL relationship with other biomarkers

Assessments of several other biomarkers related to neurodegeneration, as well as their relationship with NfL, were proposed. Eighty-eight percent (n = 44) of the programs included assessments of other biomarkers. Of the 44 programs, 36.4% (n = 16) correlated changes in those biomarkers with changes in NfL. Other PD biomarkers included tau (22.0%, n = 11), neurogranin (8.0%, n = 4), glial fibrillary acidic protein (GFAP; 24.0%, n = 12), amyloid proteins (8.0%, n = 4), phospho-neurofilament heavy chain (20.0%, n = 10), and ubiquitin carboxy-terminal hydrolase L1 (UCH-L1; 6.0%, n = 3). Biomarkers were generally not disease specific.

4. Discussion

Increasing evidence from epidemiological studies and clinical trials has highlighted a potential role of NfL as a biomarker of neuroaxonal injury in drug development programs across many neurological diseases. This investigation sought to better understand how NfL may support development of investigational therapies. We reviewed 106 clinical trial protocols from 50 IND programs submitted to the FDA between June 2005 and June 2024 and found that NfL was proposed by the drug developers as a PD biomarker uniformly across 94% of programs, and changes in NfL concentrations following exposure to an investigational therapy were observed in approximately half of programs with data available. Of note, no phase 3 trial protocols proposed changes in NfL as the primary objective, but 37.5% of phase 3 trials proposed changes in NfL as a secondary objective. The magnitude of the change, directionality, relevance to disease pathophysiology, and clinical meaningfulness of NfL were not included in this analysis, as they are context- and disease-dependent.

We found that 8% of IND programs proposed to utilize NfL as a biomarker for patient selection. Use of a biomarker in this manner may increase study efficiency or feasibility, or potentially enhance the benefit-risk relationship for patients in that subset compared to the overall population [23]. For instance, the use of plasma NfL levels to inform patient selection and predict 10-year prognosis in patients with early Huntington’s disease in clinical trials has been reported in the literature [24]. The clinical value of NfL as a prognostic biomarker when measured early in the disease course has also been proposed for other diseases including multiple sclerosis, Parkinson’s disease, and ALS [24–27].

In this assessment, we observed that 52% of IND programs proposed using NfL as a biomarker for patient stratification or subgroup analyses. NfL could potentially enable earlier intervention by identifying individuals at high risk of developing a disease and reducing delay in diagnosis to prevent further axonal damage and delay morbidity. The use of biomarkers for patient stratification and subgroup analyses may inform investigations of heterogenous outcomes by accounting for variability in the natural history of diseases commonly attributed to symptom onset, age of onset, and other factors.

In addition to patient selection, biomarker data may inform dose selection and optimization, particularly when identification of dosages based on functional measures in neurological diseases is challenging. Functional measures may not always be sensitive enough to detect dose-dependent differences in response to treatment, especially in smaller early-phase clinical trials. As such, inclusion of NfL as a biomarker throughout drug development may support dose selection. However, it remains unclear whether NfL concentrations are sufficiently sensitive to change in a dose-dependent manner. In this analysis, our assessment of exposure-response relationships between investigational drug exposure and NfL concentrations was limited because most trials were still ongoing, and data had not yet been submitted. Overall, our assessments did not provide clarity on whether NfL concentrations are sensitive to changes in drug exposure or how informative NfL concentrations are for dose optimization. Changes in PD biomarkers are often rapid compared to clinical outcomes, but neurofilament turnover under physiological conditions is a slow process in neurons [28], and several studies report half-lives for NfL that range from several weeks to months [29–32]. Therefore, NfL may display slower changes over time in response to therapy. Additional research may be required to optimize collection timepoints for NfL when used in different populations and disease states, with consideration of intrinsic/extrinsic patient characteristic and contexts of use (e.g., dose optimization).

Although there are challenges associated with using NfL to inform dose selection, evidence suggests that NfL concentrations are stable within a given patient, such that NfL displays low intra-subject variability in the absence of clinically relevant events [33,34]. For example, Brum and colleagues calculated that the mean coefficient of variation of NfL concentrations between repeated measures was around 7.4% in healthy individuals [34]. Similarly, Hviid and colleagues demonstrated that there were no significant differences in NfL concentrations across healthy individuals following blood collection every three hours for a 12-hour period [33]. Given the low intra-subject variability associated with NfL, modulation of NfL concentrations may be indicative of a treatment effect when the assay is sensitive enough to detect changes. While it is unclear if disease populations would have the same low intra-subject variability, this highlights the potential utility of NfL concentration as a PD biomarker with more research warranted.

When planning to apply NfL as a PD biomarker, the impact of age and body mass indices on NfL should be carefully considered, as these are key covariates of NfL levels, even in healthy individuals. Several studies report non-linear, age-dependent increases in blood NfL concentrations of about 1–2% per year in early to middle adulthood with further increases of about 4–5% per year in late adulthood [8,9]. Contrary to the positive associations observed between NfL concentrations and aging in adults, serum NfL concentrations have been found to decrease by approximately 6.8% per year in healthy children under 10 years old. As such, these age-dependent changes in NfL concentrations may account for the different reference ranges observed in studies of pediatric neurological diseases compared to studies in adults [35]. Furthermore, blood volume and body mass index have been shown to be inversely associated with blood NfL concentrations, which introduces an additional confounding factor that may require consideration when measuring NfL concentrations in children and adolescents compared to adults [36].

Use of NfL as a surrogate endpoint or as a biomarker to select or stratify patients was proposed for several IND programs in protocol submissions or other submitted materials. Surrogate endpoints may be used, particularly in the context of rare, severe, debilitating, and life-threatening diseases, when it is infeasible to measure a clinical outcome in a typical study duration or when the disease is mechanistically well-characterized and improvements in a surrogate are anchored to measurable clinical outcomes. Use and acceptability of surrogate endpoints is context-dependent, relying in part on factors such as disease, studied population, therapeutic mechanism of action, and availability of current treatments. In the case of tofersen, the reduction in plasma NfL observed with tofersen treatment aligned with the mechanistic understanding of the pathophysiology of disease and another biomarker for target engagement (i.e., SOD1) [37]. The relevance of NfL as a surrogate in other diseases or for other drug mechanisms remains unclear, presenting a key regulatory challenge. More evidence is needed to demonstrate the exact magnitude of change in NfL concentrations that is associated with meaningful clinical improvements. This evidence should be disease-specific and should also be supported by a clear understanding of the drug’s mechanism of action and disease pathophysiology. Of note, our analysis included investigational drug programs for which the final results were often unavailable and clinical trials were ongoing. Therefore, assessments of the clinical meaningfulness of changes in NfL concentrations were not possible.

Several disease states are associated with elevated CSF and plasma NfL concentrations at baseline, and correlations between NfL concentrations and disease severity or progression have been reported for many diseases, including multiple sclerosis [38,39], ALS [37], Alzheimer’s disease [40], dementia [41], spinal muscular atrophy (SMA) [42,43], progressive supranuclear palsy [44,45], and others [46,47]. A recent review of the connections between NfL levels and MS pathology described that progressive stages of multiple sclerosis are characterized by chronic, slowly increasing serum NfL concentrations, suggesting that axonal loss is sustained, irrespective of inflammatory involvement [38]. Similarly, in a study of 162 patients with progressive supranuclear palsy, baseline plasma NfL concentrations were found to predict neuropsychological progression and brain volume loss over a one-year period [45]. This evidence demonstrating elevation of NfL in a range of neurodegenerative diseases and correlations of NfL with disease severity and progression supports the use of NfL as a potentially useful prognostic biomarker to stratify patients. Importantly, prognostic biomarkers may inform the likely course of a disease, but these biomarkers may still not be reliable indicators of who is actually responding to a given treatment.

Characterizing whether a treatment is changing the course of the disease is a critical need in many neurodegenerative diseases, particularly in the context of early drug development. As such, understanding how modulation of NfL corresponds to changes in treatment outcomes is an important aspect of developing NfL as a treatment response biomarker, potentially as a surrogate endpoint. Longitudinal data evaluating associations between NfL and clinical outcomes with and without treatment reveal inconsistencies in the findings for some diseases. The approval of the antisense oligonucleotide nusinersen for the treatment of SMA was based on significant improvements in motor milestone responses in pediatric and adult patients [48], and additional studies have demonstrated nusinersen significantly improves event-free survival and overall survival in patients with SMA [49,50]. A recently published retrospective analysis evaluating NfL concentrations throughout long-term nusinersen treatment in 13 adult patients with SMA showed that baseline NfL concentrations were elevated compared to controls and that NfL concentrations decreased significantly with each nusinersen treatment interval compared to the loading phase [43]. However, over 50% of patients without measurable clinical improvement still had decreased NfL concentrations [43]. While NfL appears to be a sensitive pharmacodynamic biomarker, it is difficult to discern if NfL is more sensitive than the outcomes measured and what magnitude of change may be clinically meaningful [43]. Contrary to the abovementioned retrospective analysis, a prior analysis of 12 children with SMA treated with nusinersen demonstrated that NfL reduction was correlated with improved motor function [51]. Of note, NfL may have limitations that must be kept in mind. For example, treatment with gene therapy has transiently increased NfL concentrations within 30 days, potentially reflecting an immunologic reaction [52]. Overall, evidence shows an unclear relationship between changes in NfL concentrations and clinical outcomes in SMA following various treatments, making it difficult to interpret the biomarker’s significance as a treatment response biomarker across the range of therapeutic modalities.

The regulatory acceptability of surrogate endpoints for use in a particular drug or biologic development program is determined by the Agency on a case-by-case basis [53]. Because of the complexity of the role of a particular biomarker in the context of a given patient population, drug developers are encouraged to seek early input from the review division about the adequacy of their selected biomarker, particularly for the purposes of trial design, dose optimization, or establishing evidence of efficacy, particularly when using the biomarker as a surrogate endpoint.

In addition to NfL, other biomarkers were commonly proposed as exploratory objectives in clinical trial protocols. Biomarkers such as GFAP, neurogranin, and UCH-L1 are not disease-specific but may still be indicative of a pharmacodynamic effect or treatment response. For example, blood GFAP has emerged as a promising biomarker for brain and spinal cord disorders and has been found to correlate with severity of the condition and pathology [54,55]. To date, the FDA has authorized marketing of several blood tests that measure both GFAP and UCH-L1 in serum and/or plasma for clinical use in patients with traumatic brain injury [56–58]. A composite panel of nonspecific and disease-specific biomarkers may provide valuable data to strengthen the totality of evidence that a drug elicits a pharmacodynamic response and may aid in analyzing exposure-response relationships across multiple biomarkers. Disease-specific biomarkers within the causal pathway and that are linked to a drug’s mechanism of action may further support NfL’s contribution, such as in the case of SOD1 in SOD1-ALS.

Regardless of how NfL is being used in drug development, whether as a PD biomarker, biomarker for patient selection or stratification, or as a surrogate endpoint, assay methods should be analytically validated, such that the data collected can be relied upon to make drug development and regulatory decisions. We found that clinical trial protocols rarely specified the bioanalytical assay intended to quantify NfL concentrations in a particular matrix (e.g., plasma, serum, or CSF). Specifying and validating the bioanalytical method early in development can inform the accuracy and precision of the biomarker and whether it is appropriate for its intended use.

5. Conclusion

Overall, our assessment of IND programs for which drug developers proposed to evaluate NfL as a biomarker highlights the important role that NfL can play in clinical trials supported by mechanistic justification and comprehensive assessment. Currently, it remains unknown if NfL demonstrates dose-dependent changes or if doses can be optimized based on NfL concentrations. This is in part because there remains considerable uncertainty about the magnitude of change in NfL concentrations that would confer clinical meaningfulness. Given that NfL is a nonspecific marker of neurodegeneration with significant heterogeneity across age and BMI strata, the specific NfL reduction threshold that correlates with clinical meaningfulness may need to be established separately for each disease and patient population. The relationship between NfL in a specific matrix and clinical outcomes has also not been fully elucidated for many neurological conditions. Additionally, the duration of time required to detect a PD effect with NfL and to determine if the effect is dependent on the mechanism of the drug or disease is unclear. We identified patterns in the design of clinical trials evaluating NfL that may inform future drug development programs and expedite regulatory assessment of therapies for a multitude of neurological indications. In alignment with our findings from this analysis, we provide the following considerations in the design of clinical trials aiming to evaluate NfL [1]: collect pharmacokinetic and PD samples (including baseline samples) to allow for exposure-response analyses early in development and potentially identify if NfL is modulated in a dose-dependent manner [2]; discuss with the appropriate review division, a comprehensive biomarker assessment plan to facilitate early PD studies with NfL that may be anchored to clinical outcomes in the future for each disease [3]; consider collection of NfL samples from CSF as well as blood, if feasible, and collect samples at consistent timepoints across clinical trials to better characterize the role of NfL [4]; establish the role of neuroaxonal degeneration in the specific disease pathogenesis under investigation [5]; seek early input from the review division about the mechanistic hypothesis, the use of the biomarker(s) of interest to inform the study design, dose optimization, and adequacy of the data to predict clinical meaningfulness; and [6] assess the relationship between NfL and other putative disease-related biomarkers on clinical outcomes derived from clinical studies or natural history studies to create a complete, high-quality source of data to substantiate the use of NfL in specific disease states and at different stages of disease (e.g., presymptomatic).

Article highlights.

Introduction

  • Neurofilament light chain (NfL) is a marker of neuroaxonal injury that has emerged as a promising biomarker for neurodegenerative diseases.

Methods

  • Clinical protocols derived from Investigational New Drug (IND) programs that proposed to measure NfL levels were surveyed.

  • The role of NfL in clinical trials and the respective study design characteristics were assessed.

Results

  • Of the 50 IND programs identified between 2005 and 2024, 94% proposed NfL as a pharmacodynamic biomarker, 52% for patient stratification, and 8% for patient selection.

  • Of the programs with available data (n = 21), approximately 50% reported that NfL changes following drug exposure.

Discussion

  • Whether NfL demonstrates dose-dependent changes, or if doses can be optimized based on NfL concentrations, remains unknown due to the uncertainties regarding the magnitude of change in NfL required to confer clinical benefit.

  • NfL is commonly referred to prognostic because it is elevated at baseline and correlates with disease severity and progression in multiple diseases.

Conclusion

  • NfL plays an important role as a biomarker in clinical trials.

  • Additional areas of research and study design considerations to strengthen the support of NfL as a biomarker are highlighted.

Acknowledgments

The authors thank the scientists and patients who have made contributions to the field and treatment advances in this space.

We would like to thank Teresa Buracchio and Hao Zhu for their critical feedback.

Disclosure statement

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Footnotes

These findings facilitate future development of NfL and further elucidate the role that NfL may play in clinical trials, supported by mechanistic justification and comprehensive assessment. Our conclusions may bridge knowledge gaps in NfL research to support utilization of a biomarker that is fit for many contexts of use and accelerate drug development for patients with neurological diseases.

Financial Disclosure

No funding received.

Writing disclosure

No writing assistance was utilized in the production of this manuscript.

Data availability statement

The datasets generated and/or analyzed during the current study are not publicly available due to the confidential nature of data submitted to the FDA by third parties.

References

  • 1.Group F-NBW. Best (biomarkers, EndpointS, and other tools) resource. 2021.
  • 2.U.S. Food and Drug Administration. Fda facts: biomarkers and surrogate endpoints. 2017. [Available from: https://www.fda.gov/about-fda/innovation-fda/fda-facts-biomarkers-and-surrogate-endpoints
  • 3.Rosengren LE, Karlsson JE, Karlsson JO, et al. Patients with amyotrophic lateral sclerosis and other neurodegenerative diseases have increased levels of neurofilament protein in CSF. J NeuroChem. 1996;67(5):2013–2018. doi: 10.1046/j.1471-4159.1996.67052013.x [DOI] [PubMed] [Google Scholar]
  • 4.Yuan A, Rao MV, Sasaki T, et al. Alpha-internexin is structurally and functionally associated with the neurofilament triplet proteins in the mature CNS. J Neurosci. 2006;26(39):10006–10019. doi: 10.1523/JNEUROSCI.2580-06.2006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Yan Y, Jensen K, Brown A. The polypeptide composition of moving and stationary neurofilaments in cultured sympathetic neurons. Cell Motil Cytoskeleton. 2007;64(4):299–309. doi: 10.1002/cm.20184 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Narayanan S, Shanker A, Khera T, et al. Neurofilament light: a narrative review on biomarker utility. Fac Rev. 2021;10:46. doi: 10.12703/r/10-46 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Khalil M, Teunissen CE, Otto M, et al. Neurofilaments as biomarkers in neurological disorders. Nat Rev Neurol. 2018;14(10):577–589. doi: 10.1038/s41582-018-0058-z [DOI] [PubMed] [Google Scholar]
  • 8.Khalil M, Pirpamer L, Hofer E, et al. Serum neurofilament light levels in normal aging and their association with morphologic brain changes. Nat Commun. 2020;11(1):812. doi: 10.1038/s41467-020-14612-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Barro C, Benkert P, Disanto G, et al. Serum neurofilament as a predictor of disease worsening and brain and spinal cord atrophy in multiple sclerosis. Brain. 2018;141(8):2382–2391. doi: 10.1093/brain/awy154 [DOI] [PubMed] [Google Scholar]
  • 10.Tortelli R, Ruggieri M, Cortese R, et al. Elevated cerebrospinal fluid neurofilament light levels in patients with amyotrophic lateral sclerosis: a possible marker of disease severity and progression. Eur J Neurol. 2012;19(12):1561–1567. doi: 10.1111/j.1468-1331.2012.03777.x [DOI] [PubMed] [Google Scholar]
  • 11.Preische O, Schultz SA, Apel A, et al. Serum neurofilament dynamics predicts neurodegeneration and clinical progression in presymptomatic Alzheimer’s disease. Nat Med. 2019;25(2):277–283. doi: 10.1038/s41591-018-0304-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Norgren N, Rosengren L, Stigbrand T. Elevated neurofilament levels in neurological diseases. Brain Res. 2003;987(1):25–31. doi: 10.1016/s0006-8993(03)03219-0 [DOI] [PubMed] [Google Scholar]
  • 13.De Stefano N, Matthews PM, Fu L, et al. Axonal damage correlates with disability in patients with relapsing-remitting multiple sclerosis. Results of a longitudinal magnetic resonance spectroscopy study. Brain. 1998;121(Pt 8):1469–1477. doi: 10.1093/brain/121.8.1469 [DOI] [PubMed] [Google Scholar]
  • 14.Filippi M, Bozzali M, Rovaris M, et al. Evidence for widespread axonal damage at the earliest clinical stage of multiple sclerosis. Brain. 2003;126(Pt 2):433–437. doi: 10.1093/brain/awg038 [DOI] [PubMed] [Google Scholar]
  • 15.U.S. Food and Drug Administration. Drug approval package: Qalsody. FDA; 2023. https://www.accessdata.fda.gov/drugsatfda_docs/nda/2023/215887Orig1s000IntegratedR.pdf [Google Scholar]
  • 16.Meyer T, Schumann P, Weydt P, et al. Neurofilament light-chain response during therapy with antisense oligonucleotide tofersen in SOD1-related ALS: treatment experience in clinical practice. Muscle Nerve. 2023;67(6):515–521. doi: 10.1002/mus.27818 [DOI] [PubMed] [Google Scholar]
  • 17.Miller TM, Cudkowicz ME, Genge A, et al. Trial of antisense oligonucleotide tofersen for SOD1 ALS. N Engl J Med. 2022;387 (12):1099–1110. doi: 10.1056/NEJMoa2204705 [DOI] [PubMed] [Google Scholar]
  • 18.U.S. Food and Drug Administration. Integrated review: NDA215887. FDA; 2022. https://www.accessdata.fda.gov/drugsatfda_docs/nda/2023/215887Orig1s000IntegratedR.pdf [Google Scholar]
  • 19.Tropea TF, Chen-Plotkin AS. Unlocking the mystery of biomarkers: a brief introduction, challenges and opportunities in Parkinson disease. Parkinsonism Relat Disord. 2018;46(Suppl 1):S15–S8. doi: 10.1016/j.parkreldis.2017.07.021 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Olsson B, Lautner R, Andreasson U, et al. Csf and blood biomarkers for the diagnosis of Alzheimer’s disease: a systematic review and meta-analysis. Lancet Neurol. 2016;15(7):673–684. doi: 10.1016/S1474-4422(16)00070-3 [DOI] [PubMed] [Google Scholar]
  • 21.Dietmann AS, Kruse N, Stork L, et al. Neurofilament light chains in serum as biomarkers of axonal damage in early MS lesions: a histological-serological correlative study. J Neurol. 2023;270 (3):1416–1429. doi: 10.1007/s00415-022-11468-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Thebault S, Booth RA, Rush CA, et al. Serum neurofilament light chain measurement in MS: hurdles to clinical translation. Front Neurosci. 2021;15:654942. doi: 10.3389/fnins.2021.654942 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Food and Drug Administration. Enrichment strategies for clinical trials to support approval of human drugs and biological products guidance for industry. FDA; 2019. https://www.fda.gov/media/121320/download [Google Scholar]
  • 24.Parkin GM, Thomas EA, Corey-Bloom J. Plasma NfL as a prognostic biomarker for enriching HD-ISS stage 1 categorisation: a cross-sectional study. EBioMedicine. 2023;93:104646. doi: 10.1016/j.ebiom.2023.104646 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Bar-Or A, Nicholas J, Feng J, et al. Exploring the clinical utility of neurofilament light chain assays in multiple sclerosis management. Neurol Neuroimmunol Neuroinflamm. 2025;12(4):e200427. doi: 10.1212/NXI.0000000000200427 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Pedersen CC, Ushakova A, Alves G, et al. Serum neurofilament light at diagnosis: a prognostic indicator for accelerated disease progression in Parkinson’s disease. NPJ Parkinsons Dis. 2024;10(1):162. doi: 10.1038/s41531-024-00768-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Benatar M, Macklin EA, Malaspina A, et al. Prognostic clinical and biological markers for amyotrophic lateral sclerosis disease progression: validation and implications for clinical trial design and analysis. EBioMedicine. 2024;108:105323. doi: 10.1016/j.ebiom.2024.105323 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Leckey Tag CA, Coulton JB, He Y, et al. Paterson stable isotope labeling kinetics of neurofilament light in vitro and in vivo. MedRxiv. 2025. doi: 10.1101/2025.01.10.24319636 [DOI] [Google Scholar]
  • 29.Barry DM, Millecamps S, Julien JP, et al. New movements in neurofilament transport, turnover and disease. Exp Cell Res. 2007;313 (10):2110–2120. doi: 10.1016/j.yexcr.2007.03.011 [DOI] [PubMed] [Google Scholar]
  • 30.Yuan A, Nixon RA. Neurofilament proteins as biomarkers to monitor neurological diseases and the efficacy of therapies. Front Neurosci. 2021;15:689938. doi: 10.3389/fnins.2021.689938 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Zetterberg H, Blennow K. From cerebrospinal fluid to blood: the third wave of fluid biomarkers for Alzheimer’s disease. J Alzheimers Dis. 2018;64(s1):S271–S9. doi: 10.3233/JAD-179926 [DOI] [PubMed] [Google Scholar]
  • 32.Bergman J, Dring A, Zetterberg H, et al. Neurofilament light in CSF and serum is a sensitive marker for axonal white matter injury in MS. Neurol Neuroimmunol Neuroinflamm. 2016;3(5):e271. doi: 10.1212/NXI.0000000000000271 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Hviid CVB, Madsen AT, Winther-Larsen A. Biological variation of serum neurofilament light chain. Clin Chem Lab Med. 2022;60 (4):569–575. doi: 10.1515/cclm-2020-1276 [DOI] [PubMed] [Google Scholar]
  • 34.Brum WS, Ashton NJ, Simren J, et al. Biological variation estimates of Alzheimer’s disease plasma biomarkers in healthy individuals. Alzheimers & Dementia. 2024;20(2):1284–1297. doi: 10.1002/alz.13518 [DOI] [Google Scholar]
  • 35.Abdelhak A, Petermeier F, Benkert P, et al. Serum neurofilament light chain reference database for individual application in paediatric care: a retrospective modelling and validation study. Lancet Neurol. 2023;22(9):826–833. doi: 10.1016/S1474-4422(23)00210-7 [DOI] [PubMed] [Google Scholar]
  • 36.Simren J, Andreasson U, Gobom J, et al. Establishment of reference values for plasma neurofilament light based on healthy individuals aged 5–90 years. Brain Commun. 2022;4(4):fcac174. doi: 10.1093/braincomms/fcac174 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37.Lu CH, Macdonald-Wallis C, Gray E, et al. Neurofilament light chain: a prognostic biomarker in amyotrophic lateral sclerosis. Neurology. 2015;84(22):2247–2257. doi: 10.1212/WNL.0000000000001642 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Aburashed R, Eghzawi A, Long D, et al. Neurofilament light chain and multiple sclerosis: building a neurofoundational model of biomarkers and diagnosis. Neurol Int. 2025;17(4):56. doi: 10.3390/neurolint17040056 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Abdelhak A, Benkert P, Schaedelin S, et al. Neurofilament light chain elevation and disability progression in multiple sclerosis. JAMA Neurol. 2023;80(12):1317–1325. doi: 10.1001/jamaneurol.2023.3997 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Giacomucci G, Mazzeo S, Bagnoli S, et al. Plasma neurofilament light chain as a biomarker of Alzheimer’s disease in subjective cognitive decline and mild cognitive impairment. J Neurol. 2022;269(8):4270–4280. doi: 10.1007/s00415-022-11055-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Lee EH, Kwon HS, Koh SH, et al. Serum neurofilament light chain level as a predictor of cognitive stage transition. Alzheimers Res Ther. 2022;14(1):6. doi: 10.1186/s13195-021-00953-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Nitz E, Smitka M, Schallner J, et al. Serum neurofilament light chain in pediatric spinal muscular atrophy patients and healthy children. Ann Clin Transl Neurol. 2021;8(10):2013–2024. doi: 10.1002/acn3.51449 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Cordts I, Fuetterer C, Wachinger A, et al. Long-term dynamics of CSF and serum neurofilament light chain in adult patients with 5q spinal muscular atrophy treated with Nusinersen. Neurology. 2025;104(5):e213371. doi: 10.1212/WNL.0000000000213371 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 44.Donker Kaat L, Meeter LH, Chiu WZ, et al. Serum neurofilament light chain in progressive supranuclear palsy. Parkinsonism Relat Disord. 2018;56:98–101. doi: 10.1016/j.parkreldis.2018.06.018 [DOI] [PubMed] [Google Scholar]
  • 45.Rojas JC, Karydas A, Bang J, et al. Plasma neurofilament light chain predicts progression in progressive supranuclear palsy. Ann Clin Transl Neurol. 2016;3(3):216–225. doi: 10.1002/acn3.290 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Gaetani L, Blennow K, Calabresi P, et al. Neurofilament light chain as a biomarker in neurological disorders. J Neurol Neurosurg Psychiatry. 2019;90(8):870–881. doi: 10.1136/jnnp-2018-320106 [DOI] [PubMed] [Google Scholar]
  • 47.Freedman MS, Gnanapavan S, Booth RA, et al. Guidance for use of neurofilament light chain as a cerebrospinal fluid and blood biomarker in multiple sclerosis management. EBioMedicine. 2024;101:104970. doi: 10.1016/j.ebiom.2024.104970 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 48.SPINRAZA (nusinerse) prescribing information. 2016.
  • 49.Cho J, Lee J, Kim J, et al. Nusinersen demonstrates effectiveness in treating spinal muscular atrophy: findings from a three-year nation-wide study in Korea. Front Neurol. 2023;14:1294028. doi: 10.3389/fneur.2023.1294028 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 50.Hoy SM. Nusinersen: a review in 5q spinal muscular atrophy. CNS Drugs. 2021;35(12):1317–1328. doi: 10.1007/s40263-021-00878-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 51.Olsson B, Alberg L, Cullen NC, et al. Nfl is a marker of treatment response in children with SMA treated with nusinersen. J Neurol. 2019;266(9):2129–2136. doi: 10.1007/s00415-019-09389-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 52.Flotats-Bastardas M, Bitzan L, Grell C, et al. Paradoxical increase of neurofilaments in SMA patients treated with onasemnogene abeparvovec-xioi. Front Neurol. 2023;14:1269406. doi: 10.3389/fneur.2023.1269406 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 53.U.S. Food and Drug Administration. Table of surrogate endpoints that were the basis of drug approval or licensure. FDA. https://www.fda.gov/drugs/development-resources/table-surrogate-endpoints-were-basis-drug-approval-or-licensure [Google Scholar]
  • 54.Abdelhak A, Foschi M, Abu-Rumeileh S, et al. Blood GFAP as an emerging biomarker in brain and spinal cord disorders. Nat Rev Neurol. 2022;18(3):158–172. doi: 10.1038/s41582-021-00616-3 [DOI] [PubMed] [Google Scholar]
  • 55.Gogishvili D, Honey MIJ, Verberk IMW, et al. The GFAP proteoform puzzle: how to advance GFAP as a fluid biomarker in neurological diseases. J NeuroChem. 2025;169(1):e16226. doi: 10.1111/jnc.16226 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 56.U.S. Food and Drug Administration. 510(k) substantial equivalence determination decision summary K240279. FDA; 2024. https://www.accessdata.fda.gov/cdrh_docs/reviews/K240279.pdf [Google Scholar]
  • 57.U.S. Food and Drug Administration. 510(k) substantial equivalence determination decision summary K234143. FDA; 2023. https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpmn/pmn.cfm?ID=K234143 [Google Scholar]
  • 58.U.S. Food and Drug Administration. Bioanalytical method validation guidance for industry. FDA; 2018. https://www.fda.gov/files/drugs/published/Bioanalytical-Method-Validation-Guidance-for-Industry.pdf [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The datasets generated and/or analyzed during the current study are not publicly available due to the confidential nature of data submitted to the FDA by third parties.

RESOURCES