Two Serum Biomarkers Identify Sustained injury in mild TBI cohorts and American Football Players.Timothy Van Meter1, Nazanin Mirshahi1, Hayley Falk2, Vani Rao3, Haris Sair4, W. Frank Peacock5, Ramon Diaz-Arrastia6, Ronel Veksler8, Frederick Korley2,7, and Alon Friedman8,9

.1Program for Neurological Diseases, ImmunArray, Inc., Richmond, VA. 2Departments of Emergency Medicine, 3Psychiatry, and 4Radiology, Johns Hopkins University School of Medicine, Baltimore, MD. 5Department of Emergency Medicine, Baylor College of Medicine, Houston, TX. 6Center for Neuroscience and Regenerative Medicine, USUHS, Rockville, MD. 7Department of Emergency Medicine, University of Michigan, Ann Arbor, MI. 8Departments of Physiology and Cell Biology; Brain and Cognitive Sciences, Ben-Gurion University of the Negev, Beer-Sheva,

Israel. 9 Departments of Medical Neuroscience and Pediatrics, Dalhousie University, Nova Scotia, Canada.

References Cited1. Korley FK. Emergency Department Evaluation of Traumatic Brain Injury in the United States, 2009-2010. J. Head Trauma Rehabil. (Sept 2015, Epub ahead of print) 2. Yang J, Korley FK, et al.. Serum neurogranin measurement as abiomarker of acute traumatic brain injury. Clinical Biochemistry. 48(13-14):843-8, 2015.

TBI: The Continuum of Injury and Recovery

First 30 Days (approx.)Neurological dysfunction occurs with initial insult and as the

injured brain rebuilds itself.

Beyond 30 Days (approx.)Neurological dysfunction may re-emerge with the potential

induction of autoantibodies against circulating signaling molecules and brain structures.

Figure 2. Conceptual View of the TBI Continuum of Care. Following brain injury, short term damage may be observed in the first few days following initialinjury. Patient samples were analyzed by ELISA for independent detection of 5 biomarkers, of the more than 20 that are being developed by ImmunArray, Inc.

Figure 1. Trial Design and Time of Blood Sampling from the TBI Cohort Used in the Study.

Serum Biomarker Trial Design: HeadSMART Mild TBI Cohort

0 4 24 72 168 1 month 3 month 6 month

HeadSMART

Time point: (hour/month)

6 M o n t h S t u d y T i m e C o u r s e

Time 0 Day 7 Day 30 Year 30

Acute Event

Long-term Damage/ImpactShort-term Damage

Day 1 Circuit dismantling and acute cell injury and death

Analytes/Autoantigens Autoantibodies

> Day 30 Auto-antibody attachmentDay 7-30 Circuit reorganization

Introduction• Head injury brings nearly 5 million patients into emergency departments per year in the US1. The majority of patients receiving a CT (90%) have a negative CT result.

Structural MRI scanning reveals structural abnormalities in up to one third of those patients, and advanced neuroimaging methods, such as diffusion tensor imaging (DTI),detects abnormalities in an even larger fraction. There is a great need to identify TBI in patients using objective laboratory tests, as these patients are at risk for persistentpost-concussive symptoms that may affect their overall quality of life. These patients could be eligible for clinical trials with novel therapies. Blood based diagnostic testsmeasuring changes in physiological levels of circulating biomarkers may aid in identifying patients at risk for long-term symptoms of TBI, and allow stratification of patientsfor more effective treatment planning. In this study, we contrast mild traumatic brain injury patients with football players sustaining repetitive head injuries.

• Several protein biomarkers discovered in serum and CSF have been detected in TBI, including Brain Derived Neurotrophic Factor (BDNF), Neurogranin (NRGN) and GlialFibrillary Acidic Protein (GFAP)2. However, none of the antigens tested to date have been useful as single biomarkers to help confirm a diagnosis of brain injury. The currentstudy evaluated 5 brain-specific serum protein biomarkers, tested individually and in combination, to diagnose brain injury.

Results• Wilcoxon Rank Sum tests were used to compare median biomarker levels in healthy controls, athletes, and mTBI patients (Figure 3). Significant differences in

biomarker levels for SNCB were found in acute (p< 0.0001) and 1 month time points (p=0.0004) in mild TBI patients; whereas NRGN was significant at the 1 monthblood draw (p= < 0.0001), but not during acute testing (p=0.3024). In contrast, athletes sustaining repetitive injury showed significantly elevated NRGN (p<0.0001),and no difference in SNCB (p=0.601). All median comparisons were performed in RStudio version 0.99.896.

• Results from ROC analysis indicate area under the curve (AUC) values for individual biomarkers were below clinically relevant levels for sensitivity and specificity(i.e., where both <0.9) in distinguishing healthy individuals from mTBI and athletes. Combining NRGN and SNCB biomarkers in a panel gave AUC values greater than0.95 (AUC= 0.969; sensitivity=0.997; specificity =0.922).

• Significant improvements in the combined sensitivity and specificity for the classification of athletes as brain-injured are observed with three- and four-analytepanels (see Future Directions). These results were obtained using logistic regression analysis (model building with 30 bootstrap replicates, in RStudio, “pROC”package, Table 1, below).

Discussion• An objective blood-based biomarker test for diagnosis requires a combination of both high sensitivity and specificity for the successful identification of brain-injured patients

who are at risk following suspected brain injury or concussion. None of the individual 5 markers that we tested was sufficiently sensitive and specific to accurately diagnosebrain injury patients. Three- and four-marker panels were able to classify subjects with repetitive injury with high sensitivity and specificity (> 0.93).

• Athletes sustaining repetitive injury during the playing season had elevated biomarkers that are also elevated in mild TBI patients, particularly at 1 month after injury (NRGN).This may indicate the utility of chronic injury markers, such as Neurogranin, may be more appropriate for use in athletes with long term repetitive injuries from concussion orsub-concussive injury. These results should be investigated in larger athletic cohorts, and correlated with additional neuroimaging findings such as DTI-MRI.

• Further validation of these biomarker assay panels is warranted, in order to evaluate whether testing can provide benefit to clinical decision making and patient stratification.

Methods Samples: Biomarker assays were performed on three separate cohorts. Football Players: Athletes playing American football at Ben Gurion University were enrolled in the

study and blood samples were taken on and off season. All players were aged 18-39 years (median 26.9 years). MRI studies including blood brain barrierpermeability (BBBP) and diffusion weighted tensor imaging (DTI) were conducted, as well as collection of clinical information and administration of the NFLAssessment questionnaire. Mild Traumatic Brain Injury: Mild traumatic brain injury (mTBI) patients were enrolled in the HeadSMART trial at Johns HopkinsUniversity (2 sites). The HeadSMART trial is an IRB-approved prospective clinical trial that was initiated in 2014 and is still accruing. For the current biomarkerstudies, 254 mTBI patients were enrolled, all patients met the ACEP guidelines for eligibility for head CT due to head injury. Clinical data, outcome measures (GOSE,PCS), neuropsychiatric testing and neuroimaging findings were collected on all patients. Median baseline blood draw was taken 4.2 hours from injury.Subsequently, patients were evaluated at 7 additional time points out to 6 months post injury. Healthy Controls. Healthy controls were enrolled using IRB-approved screening questionnaires as siblings and family members of ED patients. Major diseases and health concerns were excluded. Male age matched controlsfrom this cohort (Median age 28, range 18-39; n=52) were used to compare with athletes from the football cohort.

Biomarker Assays: Serum biomarker concentrations for Brain-Derived Neurotrophic Factor (BDNF), Glial Fibrillary Acidic Protein (GFAP), Neurogranin (NRGN), Neuron Specific Enolase

(NSE), and Beta-Synuclein (SNCB) were assessed in duplicate tests using high sensitivity sandwich ELISA tests across replicate assays. Detection technologies wereeither Meso Scale Discovery (MSD) electro-chemiluminescence or peroxidase-mediated colorimetric detection with 3,3’,5,5’-tetramethylbenzidine (TMB). Modelbuilding was performed using logistic regression with 30 bootstrap replicates and the models were tested on the out-of-bag sample sets. Model performance wasestimated in RStudio version 0.99.896 using the “pROC” package.

Future Directions: Additional Biomarkers and Panels Enhance Sensitivity and Specificity.

Figure 4. Comparison of Biomarker Levels in Football Athletes and Age-Matched Healthy Controls. BDNF, GFAP, and NSE were tested in immunoassays.

Table 1: Performance of Biomarker Panels in Athletes

Biomarkers AUC Sensitivity Specificity# of HC

# of Athletes

NRGN, SNCB 0.969 0.997 0.922 52 23

NRGN, GFAP 0.956 0.989 0.898 41 25

NRGN, BDNF 0.967 0.978 0.913 52 19

NRGN, SNCB, NSE 0.957 0.990 0.915 52 17

NRGN, SNCB, GFAP 0.953 0.942 0.911 41 23

NRGN, SNCB, BDNF 0.975 0.990 0.928 52 17

NRGN, SNCB, GFAP, BDNF 0.969 0.980 0.937 41 17

Figure 3. Comparison of biomarker levels in mTBI, athletes, and healthy controls. Box plots demonstrate data distributions for SNCB (top row) and NRGN(bottom row) in healthy controls (n=200) versus mild TBI patients at acute evaluation (A,D; n=239 SNCB; n=227 NRGN; n=200 HC for both), and at 1 month post-injury (B, E; for SNCB, n= 59 mTBI; for NRGN n=51; n=200 controls), and in age-matched male healthy controls (n=52) versus football athletes (n= 25, C,F) Resultswere plotted in Rstudio using “ggplot2” package.

Biomarker Panel performance. • Testing of three additional biomarkers, BDNF, GFAP, and NSE, in age-matched

controls and football players showed significant median differences in NSE(p<0.0001), but not GFAP (p=0.053) or BDNF (p=0.059); Wilcoxon Rank Sum test.

• ROC analyses performed with two- , three-, and four-analyte biomarker panels wasused to optimize the balance of sensitivity and specificity in detecting injury inathletes during the playing season.

• Addition of additional markers to NRGN and SNCB enhances sensitivity andspecificity by more broadly covering the spectrum of injury in athletes sufferingrepetitive injury.

A B C

D E F