Inflammatory blood-based biomarkers to aid in the assessment and prognostication of traumatic brain injury: a TRACK-TBI study

Abstract Background Inflammatory proteins detectable in blood reflect pathoanatomic injury patterns after traumatic brain injury (TBI). Identifying biomarkers of secondary neurologic and systemic injury may improve detection of patients at risk for clinical decline and chronic disability. This study examined the utility of acute and subacute inflammatory biomarkers to differentiate TBI diagnosis and severity, and predict 6-month outcomes. Methods The Transforming Research and Clinical Knowledge in Traumatic Brain Injury (TRACK-TBI) Study prospectively enrolled TBI patients presenting to 18 United States trauma centers with head computed tomography (CT) on day 1 (D1; < 24-h post-injury). Data were extracted from TRACK-TBI subjects with D1 and 2-week (W2) plasma samples and 6-month functional outcomes, yielding 369 TBI subjects, 100 orthopedic trauma controls (OC), and 69 healthy controls. Twenty-seven inflammatory biomarkers were analyzed (MesoScale Diagnostics). TBI severity was defined using Glasgow Coma Scale (GCS; 3–12/13–15) and radiographic intracranial injury (CT-positive/CT-negative). Multivariable logistic regressions examined biomarkers as predictors of 6-month unfavorable outcomes (Glasgow Outcome Scale-Extended = 1–4 (death/severe-disability)), and adjusted for clinico-demographic factors and multiple comparisons. Adjusted odds ratios (AOR 95% confidence interval) per log 2 -unit increase in biomarker level were reported. Results Ten biomarkers (c-reactive protein (CRP), serum amyloid A (SAA), interleukin (IL)-1ꞵ, IL-2, IL-4, IL-6, IL-10, IL-15, IL-17A, tumor necrosis factor (TNF)-α) differed significantly between GCS 3–12 vs. 13–15 TBI, CT-positive vs. CT-negative TBI, and GCS 3–12 TBI vs. OC, at both D1 and W2 ( p < 0.001). IL-6, CRP, and SAA showed good discrimination of clinical TBI severity (D1/W2 area under-the-curve (AUC): 0.87/0.87, 0.82/0.88, 0.80/0.85, respectively), and moderate-to-good discrimination of radiographic TBI severity (D1/W2 AUC: 0.81/0.82, 0.78/0.83, 0.77/0.79, respectively). Five W2 biomarkers emerged as multivariable predictors of 6-month unfavorable outcomes (IL-15: AOR = 2.26 1.14–4.49; SAA: AOR = 1.91 1.37–2.67; IL-6: AOR = 1.80 1.25–2.61; IL-17A: AOR = 1.72 1.24–2.39; CRP: AOR = 1.40 1.06–1.85). Conclusions Ten circulating inflammatory proteins were associated with TBI diagnosis and severity at D1 and W2. Of these, five biomarkers expressed subacute (W2) levels predictive of 6-month death/severe-disability, underscoring their potential for validation as a novel biomarker class and integration into TBI prognostic models. Distillation of pro- and anti-inflammatory biomarker cascades in TBI could facilitate precision medicine approaches for risk stratification and therapeutic modulation.