Blood-Based Mechanistic Biomarkers Improve Prognostication of Functional Outcome After Mild Traumatic Brain Injury

Performance of currently available prognostic models for predicting incomplete functional recovery following mild traumatic brain injury (mTBI) is only modest. Blood-based biomarkers may improve model performance. In this study we aimed to assess the incremental discriminative value of promising biomarkers reflecting activation of relevant pathophysiological processes (mechanistic markers), interleukin (IL) -6, -8, and -10 (inflammation), free thiols (FTs, oxidative stress), and tryptophan (Trp; kynurenine pathway [KP]), to prognostic models containing pertinent clinical predictors. Established mTBI biomarkers, glial fibrillary acidic protein, ubiquitin carboxy-terminal hydrolase L1 (UCH-L1), and neurofilament light were also included to compare prognostic performance with the less well characterized mechanistic markers. In a prospective longitudinal cohort study (Acute Injury Markers for mild traumatic brain injury), 257 patients with mTBI were recruited. Blood was sampled within 24 h of injury at the emergency department (ED) and again 4-6 weeks later. All participants underwent acute head CT scans. Two models were used as reference models: (1) ED model, including only predictors obtained at the ED, and (2) ED+ model, including additional predictors obtained through symptom questionnaires 2 weeks after injury. The primary outcome was the Glasgow Outcome Scale Extended (GOS-E) at 6 months. The incremental prognostic value of biomarkers, relative to the ED or ED+ model, was assessed via optimism-corrected differences (Δ) in the area under the receiver operating curve (AUC) and Nagelkerke's R2. Outcome data was available for 208 patients, with 49% having incomplete recovery (GOS-E < 8). Plasma IL-6 was the only individual biomarker to improve ED model performance (ΔAUC 0.039 (95% confidence interval, 0.017, 0.057), and ΔR2 4.3% (1.2%, 6.7%), albeit modestly. UCH-L1 was the best-performing established marker (ΔAUC 0.005 [-0.008, 0.027] and ΔR2 0.1% [-3.8%, 2.1%]). Combining IL-6 with plasma Trp resulted in the largest ED model improvement (ΔAUC 0.041 [0.021, 0.065], ΔR2 4.2% [0.0%, 8.1%]). While individual biomarkers provided negligible improvement to the ED+ model, the combination of IL-6, Trp, and plasma FTs resulted in improved performance of the ED+ model (ΔAUC 0.034 [0.012, 0.055] and ΔR2 5.4% [-8.2%, 12.4%]). Blood-based mechanistic markers marginally improve performance of prognostic models for incomplete functional recovery following mTBI. The findings of this study highlight the need for further assessment of mechanistic markers, and especially IL-6, in the context of mTBI prognosis.