Установление на демографической основе нормативных значений для световой цепи нейрофильтрации и глиального фибрового кислотного белка в образцах сыворотки и плазмы.

Introduction

Neurofilament light chain (NfL) and glial fibrillary acidic protein (GFAP) are increasingly recognized blood-based biomarkers in multiple sclerosis (MS) and other autoimmune neurologic diseases. NfL, a neuronal cytoskeletal protein, reflects neuroaxonal injury, while GFAP, an astrocytic cytoskeletal protein, indicates glial injury or microglial activation. Although cerebrospinal fluid (CSF) sampling provides direct measurement, blood testing is more practical and has been studied across dementias, stroke, traumatic brain injury, neuropathies, and neuroimmunologic disorders. 1 – 3 In MS, blood NfL predicts relapse risk and disability progression, while GFAP correlates with disease activity. 4 – 8 Despite growing use, reference ranges in healthy individuals remain insufficiently defined.

Clinical application is limited by several challenges. First, these biomarkers are mechanism-specific but not disease-specific, with elevations across neurologic conditions. Levels vary with demographic and physiologic factors such as age, body mass index (BMI), renal function, and hemoglobin A 1 C. 9 – 11 Second, healthy-control reference ranges remain incomplete. Prior studies often relied on broad age ranges, convenience samples, or incomplete screening, limiting generalizability. Third, the inconsistent use of serum versus plasma further complicates comparisons, as most studies of NfL rely on serum. 4 , 6 , 9 , 12 – 14 While strong intraindividual correlations exist between serum and plasma NfL and GFAP, 14 – 16 absolute values differ, preventing interchangeability. Some studies report median differences, 15 – 18 and only one has proposed a serum–plasma conversion for NfL in older adults,
19
with none for GFAP. This prevents comparison between studies in which different preparations of blood samples are used.

To address these gaps, we prospectively enrolled adolescents through older adults at community events and a senior's clinic. Our objectives were to measure paired serum and plasma NfL and GFAP in screened healthy controls, quantify effects of age, sex, and BMI, establish reference estimates, and characterize serum–plasma relationships to support cross-study interpretation.

Methods

Healthy controls recruitment

We enrolled individuals at various community events and a senior's clinic to obtain a representative sample of the Colorado population based on the 2020 US census ( Figure 1 ).
20
Events included local college campuses, a 5 K race, outside a community of independent eateries, the Rocky Mountain Multiple Sclerosis Center Education Summits and kickoff events, the Anschutz Medical Campus Block Party, and the University's recreational center. Individuals aged 65 years and older were additionally recruited through the Seniors Clinic at the University of Colorado Anschutz Medical Campus ( Table 1 ). We created a screening tool ( Appendix A in the supplementary materials ) to determine the presence or absence of self-reported comorbidities including other major neurologic or autoimmune disease which were reviewed by a neurologist ( Figure 2 ).

Comparison of (a) racial and (b) ethnic composition of the study cohort versus the 2020 Colorado census.

*Other includes self-reported race as other or multiple.

Table 1. Distribution of the study cohort. Age Number of females Number of males % Females Total 16–19 a 12 1 92.3 13 20–29 26 19 57.8 45 30–39 13 11 54.2 24 40–49 14 3 82.3 17 50–59 10 8 55.6 18 60–69 9 4 69.2 13 70–79 4 3 57.1 7 80–85 3 3 50.0 6 Total 91 52 63.6 143
a
There were only two pediatric patients, one male and one female between 16 and 18 years.

Distribution of the study cohort.

There were only two pediatric patients, one male and one female between 16 and 18 years.

Sample selection.

Inclusion criteria were:

Male and female participants from the ages of 13 up to 85.

Willing and able to complete the medical history screening questionnaire.

Able to get blood drawn.

Able to provide consent or have a parent or guardian who can provide consent.

The exclusion criteria were:

Pregnant or nursing at the time of sample collection.

First-degree relative (parent, sibling, or child) of a person diagnosed with MS.

Prior or concurrent diagnosis of Type 1 and Type 2 diabetes mellitus.

Prior or concurrent diagnosis of a malignant neoplasm in the last 5 years (skin tumors other than melanoma were not exclusionary (e.g., basal cell or squamous cell).

Prior or concurrent diagnosis of epilepsy or a seizure disorder.

Prior or concurrent chemotherapy in the last 2 years.

Brain or spinal surgery or trauma in the last 2 years.

History of stroke at any time.

History of severe concussion or concussion with loss of consciousness in the last 2 years.

History of having undergone cranial radiation.

Prior known or diagnosed COVID-19 infection requiring hospitalization.

Prior diagnosis of autoimmune disease, multiple sclerosis, or dementia.

Community data collection

At community events, we operated a staffed booth with privacy screens for specimen collection and research education. Interested individuals reviewed study information and provided e-consent on REDCap generated forms. After consent, participants completed a screening questionnaire; eligibility was confirmed on site. For eligible participants, height, weight, and seated blood pressure were obtained before venipuncture. A serum separator tube and one K2-ethylenediaminetetraacetic acid (EDTA) plasma tube were collected per participant, stored on wet ice (≤4 h), then transported to the University of Colorado Anschutz Medical Campus for processing and −80 °C storage. Participants received a $20 gift card. The goal was to recruit up to 20 participants per decade of age (13–85 years).

Measurement of NfL and GFAP

Details of assay setup are in Appendix B in the supplementary materials . Briefly, paired EDTA plasma and serum were analyzed on the Simoa SR-X platform (Quanterix, Billerica, MA, USA) using the Neuro 4-Plex B™ Advantage kit for plasma and the Neuro 2-Plex B™ Advantage kit for serum, following the manufacturer's two-step protocols. Assays were run with lot-specific calibrators and results were generated from extrapolated standard curves. 21 , 22

Statistical methods

Analyses were performed using complete-case data. NfL and GFAP concentrations in both serum and plasma were natural log-transformed to reduce right-skew and support log-linear modeling. Geometric means and 95% confidence intervals were summarized by sex within age decades beginning at age 13 years. For each biomarker and sample type, graphical diagnostics and spline-based assessments were used to evaluate the adequacy of a log-linear age effect and to inform covariate selection. The log-transformed biomarker values were regressed on age, BMI, and sex; observations were weighted by the number of replicate measurements. Piecewise log-linear models were used when departures from log-linearity were important, and sex-stratified models were also fit. Regression coefficients for effects of age, BMI, and sex were tested for statistical significance. Model results are reported as percent change per year of age, percent change per unit BMI, and male-to-female ratios, with simultaneous 95% Scheffé confidence bands for predicted geometric means. Based on these models, we developed algorithms to estimate conditional geometric means as a function of age, BMI, and sex with simultaneous 95% Scheffé confidence bands and 95% prediction intervals.

Serum and plasma NfL and GFAP concentrations were compared using Spearman correlations. To derive conversion formulas between samples we fit linear regression models of log(serum) on log(plasma) and of log(plasma) on log(serum) for each biomarker. All tests were two-sided with α  = 0.05 unless otherwise specified. Analyses were conducted in SAS 9.4, R, and Stata 17.

Results

Neurofilament light (NfL)

Geometric mean NfL and GFAP were measured in plasma and serum for all participants. Plasma NfL ranged from 2.48 pg/mL (95% CI: 1.87–3.29) at ages 16–19 to 19.55 pg/mL (95% CI: 9.56–39.99) at ages 80–85 ( Table 2 ). A log-linear age effect fit the data. Adjusted for BMI and sex, plasma NfL increased 2.96% per year (95% CI: 2.55–3.36, p < . 0001); results were similar by sex ( Figure 3 ). Plasma NfL decreased 1.70% per BMI unit (95% CI: 0.42–2.96, p = . 0095). The male:female ratio, adjusted for age and BMI, was 1.14 (95% CI: 0.99–1.32, p = . 0625) ( Table 3 ).

Table 2. Geometric mean NfL and GFAP results in plasma and serum back transformed from the logarithmic scale to original sale. Age range Sample size ( n ) Geometric mean NfL – plasma [95% CI] Geometric mean NfL – serum [95% CI] Geometric mean GFAP – plasma [95% CI] Geometric mean GFAP – serum [95% CI] 16–19 13 2.48 [1.87, 3.29] 3.36 [2.84, 3.97] 49.99 [41.74, 59.87] 54.11 [44.28, 66.11] 20–29 45 3.43 [2.98, 3.95] 3.84 [3.39, 4.34] 46.17 [40.94, 52.06] 46.83 [41.23, 53.19] 30–39 24 3.99 [3.35, 4.75] 4.30 [3.69, 5.00] 43.11 [35.01, 53.10] 42.36 [32.69, 54.88] 40–49 17 5.18 [4.36, 6.15] 5.80 [5.00, 6.72] 55.35 [44.43, 68.95] 56.81 [44.89, 71.88] 50–59 18 7.25 [6.10, 8.61] 8.08 [6.93, 9.43] 70.71 [59.14, 84.54] 77.95 [65.58, 92.65] 60–69 13 8.79 [7.27, 10.63] 9.83 [8.07, 11.98] 87.35 [69.97, 109.05] 91.33 [72.96, 114.34] 70–79 7 14.39 [9.37, 22.12] 16.36 [11.02, 24.28] 118.55 [92.52, 151.90] 137.41 [101.30, 186.40] 80–85 6 19.55 [9.56, 39.99] 21.85 [10.87, 43.94] 121.77 [91.07, 162.82] 154.47 [121.58, 196.26]

Geometric mean NfL and GFAP results in plasma and serum back transformed from the logarithmic scale to original sale.

Standard curve for log-transformed Nfl and GFAP values from 17 to 85 years of age in plasma.

Note: The bounds on the input ranges for age (years) and BMI were set to the first and last integers within the range of the data. Extrapolating the prediction model beyond the range of data used to fit the model is not valid.

Table 3. Effects on NfL and GFAP in plasma for age, sex and BMI. Sex Response Effect Estimate a 95% confidence interval p -value All [NfL] Age 2.96 2.55 3.36 <.0001 BMI −1.70 −2.96 −0.42 .0095 Sex 1.14 0.99 1.32 .0625 Age (<30) −2.28 −4.13 −0.39 .0185 [GFAP] Age (≥30) 2.34 1.86 2.82 <.0001 BMI −0.89 −2.09 0.32 .1488 Sex 0.91 0.80 1.04 .1808 Female [NfL] Age 3.00 2.53 3.47 <.0001 BMI −1.67 −2.99 −0.33 .0153 Age (<30) −2.23 −4.52 0.11 .0616 [GFAP] Age (≥30) 2.39 1.70 3.09 <.0001 BMI −0.62 −2.13 0.92 .4227 Male [NfL] Age 2.89 2.12 3.67 <.0001 BMI −1.89 −4.81 1.11 .2087 Age (<30) −2.53 −5.89 0.95 .1485 [GFAP] Age (≥30) 2.23 1.56 2.91 <.0001 BMI −1.76 −3.89 0.42 .1096
a
Estimates are percent change per year of age, percent change per unit of BMI, and ratio between males and females for sex that are back transformed from the logarithmic scale to original scale.

Effects on NfL and GFAP in plasma for age, sex and BMI.

Estimates are percent change per year of age, percent change per unit of BMI, and ratio between males and females for sex that are back transformed from the logarithmic scale to original scale.

Serum NfL ranged from 3.34 pg/mL (95% CI: 2.84–3.97) at ages 16–19 to 21.85 pg/mL (95% CI: 10.87–43.94) at ages 80–85 ( Table 2 ). Adjusted for BMI and sex, serum NfL increased 2.90% per year (95% CI: 2.57–3.24, p < . 0001); results were similar by sex ( Figure 4 ). Serum NfL decreased 1.88% per BMI unit (95% CI: 0.76–2.97, p = . 0011). The adjusted male:female ratio was 1.08 (95% CI: 0.96–1.22, p = . 2128) ( Table 4 ).

Standard curve for log-transformed Nfl and GFAP values from 17 to 85 years of age in serum.

Note: The bounds on the input ranges for age (years) and BMI were set to the first and last integers within the range of the data. Extrapolating the prediction model beyond the range of the data used to fit the model is not recommended and may be unreliable.

Table 4. Effects on NfL and GFAP in serum for age, sex and BMI. Sex Response Effect Estimate a 95% confidence interval p -value All [NfL] Age 2.90 2.57 3.24 <.0001 BMI −1.88 −2.97 −0.76 .0011 Sex 1.08 0.96 1.22 .2128 Age (<30) −2.71 −4.73 −0.64 .0108 [GFAP] Age (≥30) 2.69 2.19 3.19 <.0001 BMI −1.07 −2.38 −0.25 .1112 Sex 0.82 0.71 0.95 .0077 Female [NfL] Age 2.81 2.39 3.23 <.0001 BMI −1.26 −2.53 0.01 .0523 Age (<30) −2.17 −4.69 0.42 .0983 [GFAP] Age (≥30) 2.49 1.78 3.21 <.0001 BMI −0.84 −2.48 0.83 .3181 Male [NfL] Age 2.97 2.41 3.54 <.0001 BMI −3.51 −5.80 −1.17 .0043 Age (<30) −4.01 −7.63 −0.26 .0396 [GFAP] Age (≥30) 2.95 2.26 3.65 <.0001 BMI −1.48 −3.83 0.93 .2199
a
Estimates are percent change per year of age, percent change per unit of BMI, and ratio between males and females for sex that are back transformed from the logarithmic scale to original scale.

Effects on NfL and GFAP in serum for age, sex and BMI.

Estimates are percent change per year of age, percent change per unit of BMI, and ratio between males and females for sex that are back transformed from the logarithmic scale to original scale.

GFAP

Plasma GFAP ranged from 49.99 pg/mL (95% CI: 41.74–59.87) at ages 16–19 to 121.77 pg/mL (95% CI: 91.07–162.82) at ages 80–85 ( Table 2 ). Diagnostics supported a piecewise age model with a knot at 30 years: ≤30 years, GFAP decreased 2.28%/year (95% CI: 0.39–4.13, p = . 0185); ≥30 years, GFAP increased 2.34%/year (95% CI: 1.86–2.82, p < . 0001; difference in slopes p < . 0001) ( Figure 3 ). The BMI effect was not significant (−0.89% per unit; 95% CI: −2.09–0.32, p = . 1488). Adjusted plasma GFAP did not differ meaningfully by sex (male:female 0.91, 95% CI: 0.80–1.04, p = . 1808) ( Table 3 ).

Serum GFAP ranged from 54.11 pg/mL (95% CI: 42.28–66.11) at ages 16–19 to 154.47 pg/mL (95% CI: 121.58–196.26) at ages 80–85 ( Table 2 ). With the same piece-wise model, ≤30 years, GFAP decreased 2.71%/year (95% CI: 0.64–4.73, p = . 0108); ≥30 years, GFAP increased 2.69%/year (95% CI: 2.19–3.19, p < . 0001; difference in slopes p < . 0001) ( Figure 4 ). The BMI effect was not significant (−1.07% per unit; 95% CI: −0.25 to 2.38, p = . 1112). Adjusted serum GFAP was lower in males (male:female 0.82, 95% CI: 0.71–0.95, p = . 0077) ( Table 4 ).

Effects of age and other factors

Age was the dominant predictor for both biomarkers across sample types. NfL increased with age in plasma and serum (≈2.9–3.0% per year; Tables 3 and 4 ). GFAP demonstrated a non-linear relationship: levels declined modestly before age 30 and rose thereafter, consistent with the piecewise models used for inference ( Tables 3 and 4 ). Only two participants in the entire cohort endorsed regular tobacco use, which limits the ability to understand the impact of this covariate on GFAP and NfL.

Effect of BMI

Higher BMI was associated with lower NfL concentrations in the overall cohort, although sex-stratified analyses showed variability in the estimated strength and statistical significance of these associations. GFAP demonstrated a similar inverse directional trend, but BMI effects were not statistically significant ( Tables 3 and 4 ).

Plasma-Serum correlation

Figures 5 and 6 show that the association between plasma and serum levels for NfL and GFAP, on the logarithmic scale, is approximately linear. The fitted linear regression formulas are displayed below:

Predicted LN[NfL Serum (pg/mL)] = 0.35589 + 0.86145*LN[NfL Plasma (pg/mL)]

Predicted LN[GFAP Serum (pg/mL)] = 0.26540 + 0.94948*LN[GFAP Plasma (pg/mL)]

Predicted LN[NfL Plasma (pg/mL)] = −0.11519 + 0.98814*LN[NfL Serum (pg/mL)]

Predicted LN[GFAP Plasma (pg/mL)] = 0.88732 + 0.76813*LN[GFAP Serum (pg/mL)]

Correlation between plasma and serum Nfl levels.

Correlation between plasma and serum GFAP levels.

The prediction is the expected value of either LN[serum] or LN[plasma] conditional on a measurement of the other. Predicted LN[serum] or LN[plasma] can then be exponential (base e) to the original scale. Logarithms and exponentiations are one-to-one monotonic transformations. The back transformation to the original scale is built into the online calculator described below. Conversion predictions are limited to range of the observed data used and approximations are not advised outside of that range. See Table 2 for conversion-specific acceptable inputs.

Online tool

We provide an online calculator (age = 17–85 years; BMI = 17–44; sex female/male) that returns geometric mean estimates with simultaneous 95% Scheffé confidence bands, and 95% prediction intervals based on the fitted models: https://eric-gutierrez.shinyapps.io/HCdata_tool/ .

The tool enables researchers to generate sample-specific (serum/plasma) reference values for NfL and GFAP matched on age, BMI, and sex, and to compare expected values across plasma and serum. We also provide an algorithm for converting plasma concentrations into serum concentrations, and vice-versa, with accompanying simultaneous 95% Scheffé confidence bands for the estimates. 23 , 24

Discussion

Our principal findings were consistent across biomarkers and sample types. Age was the dominant driver: NfL increased by approximately 3% per year in both plasma and serum, whereas GFAP showed a non-linear relationship, decreasing before age 30 and increasing thereafter. Our principal findings were consistent across biomarkers and sample types. Age was the dominant driver: NfL increased by approximately 3% per year in both plasma and serum, whereas GFAP showed a non-linear relationship, decreasing before age 30 and increasing thereafter. Our findings regarding BMI should be interpreted with some nuance. Higher BMI was associated with lower NfL (and a similar direction for GFAP) concentrations in the overall cohort for both plasma and serum samples; however, sex-stratified analyses demonstrated variability in the estimated magnitude and statistical significance of these associations. Additional exploratory analysis found no statistically significant effect modification for age and BMI by sex (results not shown and can be provided if needed). Given the modest sample size within sex-stratified subgroups, these differences may reflect limited statistical power and wider confidence intervals rather than definitive biologic effect modification by sex. Nonetheless, the findings suggest that potential sex-specific relationships between BMI and circulating neurologic biomarkers merit further evaluation in larger, adequately powered cohorts .

When compared to prior research summarized in Table 5 , Fitzgerald et al. leveraged the U.S.-nationally representative NHANES dataset to model serum NfL and identify which demographic, clinical, and laboratory factors explain variability beyond age, helping contextualize NfL in a broad general-population setting. Khalil et al. analyzed serum NfL in the Austrian Stroke Prevention cohort and linked longitudinal NfL trajectories to subclinical MRI brain changes and cognitive outcomes over ∼6 years, positioning NfL as a marker of brain aging and cerebrovascular risk rather than a reference-range tool.
25
Vermunt et al. presented age- and disease-specific NfL reference values in an interactive support interface, addressing the need to contextualize NfL across healthy and diseased cohorts.
26
Benkert et al. derived age- and BMI-adjusted percentiles and z -scores for serum NfL and validated clinical relevance in MS cohorts, enabling standardized interpretation across individuals and therapies.
27
Our work complements these efforts by adding GFAP and by providing paired plasma and serum reference estimates derived from a screened, community-based cohort. Our work also highlights sample-specific differences whose considerations remain important for harmonizing the literature. While within-person correlations between serum and plasma are typically strong, systematic differences in absolute values and pre-analytic factors can influence measurements. 14 – 16 , 19 When possible, clinicians and researchers should interpret values using reference estimates derived for the same sample type and assay platform.

Table 5. Summary of relevant prior evidence. Study What it did Key ways it differs from the community healthy-control study Current community healthy-control study Community-based recruitment to derive age/BMI/sex-referenced values for Nfl and GFAP, with paired serum and plasma measures and an online calculator. Includes GFAP and paired serum/plasma; designed to better reflect local demography; focuses on a concise covariate set (age/BMI/sex) rather than extensive clinical covariate modeling. Fitzgerald et al., 2022 Used a U.S. nationally representative survey dataset to model serum NfL and quantify which demographic/clinical/lab factors explain variability beyond age. Serum Nfl only (no GFAP); emphasis on broad covariate discovery/adjustment (52 candidate predictors); “healthy” defined largely by exclusions in a general-population survey rather than targeted community sampling. Khalil et al., 2020 Measured serum Nfl in a population-based aging cohort and linked Nfl trajectories to subclinical MRI brain changes and cognition over ∼6 years. Not a reference-range paper per se; focuses on neuroimaging/longitudinal brain outcomes; serum Nfl only (no GFAP, no paired plasma); cohort enriched for older adults and cerebrovascular risk context. Vermunt et al., 2022 Built an online interface to contextualize Nfl against age-specific reference values and disease distributions. Primarily an interpretation/interface paper spanning healthy and disease cohorts; details of matrices and cohorts differ Benkert et al., 2022 Created a large serum Nfl reference database corrected for age and BMI and used Z-scores/percentiles to predict future MS activity and compare therapy effectiveness. MS-focused application and validation; serum Nfl only (no GFAP); reference database pooled from multiple cohorts and genetic-study controls (not community sampling); modeling explicitly incorporates BMI and emphasizes Z-scores/percentiles for clinical use.

Summary of relevant prior evidence.

Limitations

Several limitations should be considered. First, the data are cross-sectional; therefore, we report associations across age rather than within-person longitudinal change. Second, medical history and symptom screening were self-reported and may not fully exclude subclinical conditions that influence biomarker levels. Third, despite community outreach, the cohort may not fully represent all Colorado demographic groups; future studies could improve representativeness by partnering with community clinics, using bilingual materials, mobile or home phlebotomy, and targeted recruitment or oversampling of underrepresented groups. Fourth, sample sizes in the youngest and oldest strata were smaller, leading to wider confidence intervals at the extremes. Fifth, although participants with a history of diabetes were excluded, renal function was not systematically incorporated into the analyses. Because biomarkers such as NfL and GFAP may be influenced by kidney function, and renal function changes with age, the absence of adjustment for measures such as estimated glomerular filtration rate (eGFR) could have introduced residual confounding and may partially influence the observed age-related associations. Sixth, while we can estimate expected values of the biomarkers with decent precision, variability among individual patients renders the prediction intervals wide. Physicians and researchers who use our predictions should be aware of the heterogeneity of biomarker levels among individual patients.

Conclusions

In screened, community-based healthy controls, both NfL and GFAP increase with age in plasma and serum, with GFAP demonstrating a piece-wise relationship before and after age 30. Higher BMI was modestly associated with lower biomarker concentrations, particularly for NfL. Independent sex effects were generally limited after adjustment for age and BMI, although serum GFAP concentrations were significantly lower in males compared with females. Qualitatively, results were mostly similar by sex.

Strong serum-plasma correlations support concordant trends across matrices but do not imply interchangeability of absolute values. Together, these data provide practical, sample-specific reference estimates and an online tool to support interpretation of NfL and GFAP in research and clinical contexts.

Supplemental Material

sj-docx-1-mso-10.1177_20552173261466440 - Supplemental material for Establishing demographic-based normative values for neurofilament light chain and glial fibrillary acidic protein in serum and plasma samples

Supplemental material, sj-docx-1-mso-10.1177_20552173261466440 for Establishing demographic-based normative values for neurofilament light chain and glial fibrillary acidic protein in serum and plasma samples by Tyler L Borko, Alanna Ritchie, Sean Selva, Stefan Sillau, Eric Engebretson, Andrew B Wolf, Enrique Alvarez, Robert Gross, Eric Gutierrez, John Corboy, Diego Silva, Ryan Kammeyer, Amanda L Piquet, Timothy L Vollmer, Salim Chahin, Gregory Owens, Jeffrey L Bennett and Kavita V Nair in Multiple Sclerosis Journal – Experimental, Translational and Clinical

sj-docx-2-mso-10.1177_20552173261466440 - Supplemental material for Establishing demographic-based normative values for neurofilament light chain and glial fibrillary acidic protein in serum and plasma samples

Supplemental material, sj-docx-2-mso-10.1177_20552173261466440 for Establishing demographic-based normative values for neurofilament light chain and glial fibrillary acidic protein in serum and plasma samples by Tyler L Borko, Alanna Ritchie, Sean Selva, Stefan Sillau, Eric Engebretson, Andrew B Wolf, Enrique Alvarez, Robert Gross, Eric Gutierrez, John Corboy, Diego Silva, Ryan Kammeyer, Amanda L Piquet, Timothy L Vollmer, Salim Chahin, Gregory Owens, Jeffrey L Bennett and Kavita V Nair in Multiple Sclerosis Journal – Experimental, Translational and Clinical

Acknowledgements

This publication is dedicated to Brooke Valdez whose tireless dedication makes this work possible.

Footnotes

Data availability: Anonymized data published within this article can be made available by request for any qualified investigator.

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding: The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Bristol Myers Squibb Foundation.

Supplemental material: Supplemental material for this article is available online.

ORCID iDs: Tyler L. Borko https://orcid.org/0000-0002-6679-1867

Eric Engebretson https://orcid.org/0009-0006-8875-5521

Andrew B. Wolf https://orcid.org/0000-0002-1406-4107

Eric Gutierrez https://orcid.org/0000-0002-9722-1699

John Corboy https://orcid.org/0000-0003-4169-1461

Ryan Kammeyer https://orcid.org/0000-0001-8657-3106

Contributor Information

Tyler L Borko, Department of Neurology, University of Colorado Anschutz School of Medicine, Aurora, CO, USA; Midwestern University Arizona College of Osteopathic Medicine, Glendale, AZ, USA.

Alanna Ritchie, Department of Neurology, University of Colorado Anschutz School of Medicine, Aurora, CO, USA.

Sean Selva, Department of Neurology, University of Colorado Anschutz School of Medicine, Aurora, CO, USA.

Stefan Sillau, Department of Neurology, University of Colorado Anschutz School of Medicine, Aurora, CO, USA.

Eric Engebretson, Department of Neurology, University of Colorado Anschutz School of Medicine, Aurora, CO, USA.

Andrew B Wolf, Department of Neurology, University of Colorado Anschutz School of Medicine, Aurora, CO, USA.

Enrique Alvarez, Department of Neurology, University of Colorado Anschutz School of Medicine, Aurora, CO, USA.

Robert Gross, Department of Neurology, University of Colorado Anschutz School of Medicine, Aurora, CO, USA.

Eric Gutierrez, Department of Neurology, University of Colorado Anschutz School of Medicine, Aurora, CO, USA.

John Corboy, Department of Neurology, University of Colorado Anschutz School of Medicine, Aurora, CO, USA.

Diego Silva, Bristol Myers Squibb, New York, NY, USA.

Ryan Kammeyer, Department of Neurology, University of Colorado Anschutz School of Medicine, Aurora, CO, USA.

Amanda L Piquet, Department of Neurology, University of Colorado Anschutz School of Medicine, Aurora, CO, USA.

Timothy L Vollmer, Department of Neurology, University of Colorado Anschutz School of Medicine, Aurora, CO, USA.

Salim Chahin, Department of Neurology, Washington University in St. Louis, St. Louis, MO, USA.

Gregory Owens, Department of Neurology, University of Colorado Anschutz School of Medicine, Aurora, CO, USA.

Jeffrey L Bennett, Department of Neurology, University of Colorado Anschutz School of Medicine, Aurora, CO, USA; Department of Ophthalmology, Programs in Neuroscience and Immunology, University of Colorado Anschutz School of Medicine, Aurora, CO, USA.

Kavita V Nair, Department of Neurology, University of Colorado Anschutz School of Medicine, Aurora, CO, USA; Skaggs School of Pharmacy and Pharmaceutical Sciences, University of Colorado Anschutz School of Medicine, Aurora, CO, USA.

References

Associated Data

Supplementary Materials

sj-docx-1-mso-10.1177_20552173261466440 - Supplemental material for Establishing demographic-based normative values for neurofilament light chain and glial fibrillary acidic protein in serum and plasma samples

Supplemental material, sj-docx-1-mso-10.1177_20552173261466440 for Establishing demographic-based normative values for neurofilament light chain and glial fibrillary acidic protein in serum and plasma samples by Tyler L Borko, Alanna Ritchie, Sean Selva, Stefan Sillau, Eric Engebretson, Andrew B Wolf, Enrique Alvarez, Robert Gross, Eric Gutierrez, John Corboy, Diego Silva, Ryan Kammeyer, Amanda L Piquet, Timothy L Vollmer, Salim Chahin, Gregory Owens, Jeffrey L Bennett and Kavita V Nair in Multiple Sclerosis Journal – Experimental, Translational and Clinical

sj-docx-2-mso-10.1177_20552173261466440 - Supplemental material for Establishing demographic-based normative values for neurofilament light chain and glial fibrillary acidic protein in serum and plasma samples

Supplemental material, sj-docx-2-mso-10.1177_20552173261466440 for Establishing demographic-based normative values for neurofilament light chain and glial fibrillary acidic protein in serum and plasma samples by Tyler L Borko, Alanna Ritchie, Sean Selva, Stefan Sillau, Eric Engebretson, Andrew B Wolf, Enrique Alvarez, Robert Gross, Eric Gutierrez, John Corboy, Diego Silva, Ryan Kammeyer, Amanda L Piquet, Timothy L Vollmer, Salim Chahin, Gregory Owens, Jeffrey L Bennett and Kavita V Nair in Multiple Sclerosis Journal – Experimental, Translational and Clinical