Biomarkers of Neurobiologic Recovery in Adults With Sport-Related Concussion

Abstract

Importance: Sport-related concussion (SRC), a form of mild traumatic brain injury, is a prevalent occurrence in collision sports. There are no well-established approaches for tracking neurobiologic recovery after SRC.

Objective: To examine the levels of serum glial fibrillary acidic protein (GFAP) and neurofilament light (NfL) in Australian football athletes who experience SRC.

Design, setting, and participants: A cohort study recruiting from April 10, 2021, to September 17, 2022, was conducted through the Victorian Amateur Football Association, Melbourne, Australia. Participants included adult Australian football players with or without SRC. Data analysis was performed from May 26, 2023, to March 27, 2024.

Exposure: Sport-related concussion, defined as at least 1 observable sign and/or 2 or more symptoms.

Main outcomes and measures: Primary outcomes were serum GFAP and NfL levels at 24 hours, and 1, 2, 4, 6, 8, 12, and 26 weeks. Secondary outcomes were symptoms, cognitive performance, and return to training times.

Results: Eighty-one individuals with SRC (median age, 22.8 [IQR, 21.3-26.0] years; 89% male) and 56 control individuals (median age, 24.6 [IQR, 22.4-27.3] years; 96% male) completed a total of 945 of 1057 eligible testing sessions. Compared with control participants, those with SRC exhibited higher GFAP levels at 24 hours (mean difference [MD] in natural log, pg/mL, 0.66 [95% CI, 0.50-0.82]) and 4 weeks (MD, 0.17 [95% CI, 0.02-0.32]), and NfL from 1 to 12 weeks (1-week MD, 0.31 [95% CI, 0.12-0.51]; 2-week MD, 0.38 [95% CI, 0.19-0.58]; 4-week MD, 0.31 [95% CI, 0.12-0.51]; 6-week MD, 0.27 [95% CI, 0.07-0.47]; 8-week MD, 0.36 [95% CI, 0.15-0.56]; and 12-week MD, 0.25 [95% CI, 0.04-0.46]). Growth mixture modeling identified 2 GFAP subgroups: extreme prolonged (16%) and moderate transient (84%). For NfL, 3 subgroups were identified: extreme prolonged (7%), moderate prolonged (15%), and minimal or no change (78%). Individuals with SRC who reported loss of consciousness (LOC) (33% of SRC cases) had higher GFAP at 24 hours (MD, 1.01 [95% CI, 0.77-1.24]), 1 week (MD, 0.27 [95% CI, 0.06-0.49]), 2 weeks (MD, 0.21 [95% CI, 0.004-0.42]) and 4 weeks (MD, 0.34 [95% CI, 0.13-0.55]), and higher NfL from 1 week to 12 weeks (1-week MD, 0.73 [95% CI, 0.42-1.03]; 2-week MD, 0.91 [95% CI, 0.61-1.21]; 4-week MD, 0.90 [95% CI, 0.59-1.20]; 6-week MD, 0.81 [95% CI, 0.50-1.13]; 8-week MD, 0.73 [95% CI, 0.42-1.04]; and 12-week MD, 0.54 [95% CI, 0.22-0.85]) compared with SRC participants without LOC. Return to training times were longer in the GFAP extreme compared with moderate subgroup (incident rate ratio [IRR], 1.99 [95% CI, 1.69-2.34]; NfL extreme (IRR, 3.24 [95% CI, 2.63-3.97]) and moderate (IRR, 1.43 [95% CI, 1.18-1.72]) subgroups compared with the minimal subgroup, and for individuals with LOC compared with those without LOC (IRR, 1.65 [95% CI, 1.41-1.93]).

Conclusions and relevance: In this cohort study, a subset of SRC cases, particularly those with LOC, showed heightened and prolonged increases in GFAP and NfL levels, that persisted for at least 4 weeks. These findings suggest that serial biomarker measurement could identify such cases, guiding return to play decisions based on neurobiologic recovery. While further investigation is warranted, the association between prolonged biomarker elevations and LOC may support the use of more conservative return to play timelines for athletes with this clinical feature.

Figure 1.. The Profiles of Serum Glial Fibrillary Acidic Protein (GFAP) and Neurofilament Light (NfL) Levels After Sport-Related Concussion

At a group level, serum GFAP levels were significantly increased in the sport-related concussion (SRC) participants vs control participants at 24 hours and 4 weeks (A). Serum NfL levels were increased at 1 week, 2 weeks, 4 weeks, 6 weeks, 8 weeks, and 12 weeks (B). While the control group had a relatively consistent profile of GFAP over time (C), the SRC group showed heterogeneity in magnitude of these changes, with some participants appearing to have a secondary increase at approximately 4 weeks (D). Similarly for serum NfL, a comparatively consistent profile was observed in control participants (E), whereas in the SRC participants, there was clear heterogeneity in the extent and timing of peak NfL levels (F). Box plots show the minimum, lower quartile, median, upper quartile, and maximum value. ^aP < .001. ^bP < .05. ^cP < .01.

Figure 2.. Glial Fibrillary Acidic Protein (GFAP) and Neurofilament Light (NfL) Trajectory Subgroups: Biomarker Profiles, Symptoms and Return to Training Time After Sport-Related Concussion (SRC)

Symbols and error bars in A and B represent the median and IQR. Data for C, D, E, and F are presented using box plots showing the minimum, lower quartile, median, upper quartile, and maximum value. ^aP < .001 for the extreme GFAP subgroup vs control participants. ^bP < .001 for the moderate GFAP subgroup vs control participants. ^cP < .05 for the extreme GFAP subgroup vs control participants. ^dP < .01 for the extreme GFAP subgroup vs control participants. ^eP < .05 for the moderate NfL subgroup vs control participants. ^fP < .05 for the extreme NfL subgroup vs control participants. ^gP < .05 for the extreme GFAP subgroup vs moderate GFAP subgroup. ^hP < .05 for the moderate GFAP subgroup vs control participants. ⁱP < .05 for the minimal or no change in NfL subgroup vs control participants. ^jP < .05 for the moderate NfL subgroup vs the minimal or no change in NfL subgroup. ^kP < .001.

Figure 3.. Biomarker and Clinical Recovery in Sport-Related Concussion (SRC) Participants With and Without Loss of Consciousness (LOC)

Serum glial fibrillary acidic protein (GFAP) levels were higher at 24 hours, 1 week, 2 weeks, and 4 weeks in SRC participants with LOC compared with those without (A). Similarly for serum neurofilament light (NfL), levels were higher in participants with LOC at 1 week, 2 weeks, 4 weeks, 6 weeks, 8 weeks, and 12 weeks compared with participants without LOC (B). Symptom evaluation with the Rivermead Post Concussion Questionnaire (RPQ) revealed a greater severity of symptoms in SRC participants with vs without LOC at 24 hours, but no other time points (C). SRC participants with LOC reported a greater time to return to training than participants without LOC (D). With return to training affected by end of season, COVID-19 lockdowns, or failure to return to training, 15 cases were not included. Box plots show the minimum, lower quartile, median, upper quartile, and maximum value. ^aP < .001. ^bP < .05. ^cP < .01.