A-094 Performance Evaluation of GFAP and UCH-L1 Biomarkers for Traumatic Brain Injury in the Alinity i TBI Test (in development)

Abstract Background Traumatic Brain Injury (TBI) has been recognized as an important cause of death and disability and is a growing public health problem. An estimated 69 million people globally experience a TBI annually. Blood-based biomarkers such as glial fibrillary acidic protein (GFAP) and ubiquitin C-terminal hydrolase L1 (UCH-L1) have shown utility to predict acute traumatic intracranial injury on head CT scan after TBI. Methods The Alinity i TBI test is a panel of in vitro diagnostic chemiluminescent microparticle immunoassays for the measurement of GFAP and UCH-L1 in plasma and serum. Performance characteristics such as detection limits, imprecision, linearity, measuring interval, expected values, and interferences were established following CLSI guidance. A feasibility cohort of 354 subjects was used to establish cutoffs which assist in determining the need for a CT (computed tomography) scan of the head. A prospective study was conducted to assess the clinical performance of the TBI test. Fresh plasma specimens from 97 mild TBI subjects presenting to Level I trauma centers were collected within 12 h of injury from subjects ≥18 years of age who presented with suspected TBI and had a head CT scan performed. The TBI interpretation is considered positive if either GFAP or UCH-L1 is greater than or equal to the cutoffs. The TBI interpretation is considered negative if both GFAP and UCH-L1 are less than the cutoffs. Results The analytical measuring interval (AMI) extends from the limit of quantitation (LoQ) to the upper LoQ and is determined by the range that demonstrates acceptable performance for linearity, imprecision, and bias. The AMI is 6.1 to 42 000.0 pg/mL for GFAP and 26.3 to 25 000.0 pg/mL for UCH-L1. Within-laboratory imprecision (20 day) ranged from 2.8% to 5.3% CV for GFAP and 2.1% to 5.6% CV for UCH-L1. Cutoffs for GFAP and UCH-L1 of 35.0 pg/mL and 400.0 pg/mL, respectively, were selected using both 10-fold cross validation and bootstrapping methods and applying a selection criterion of negative predictive value (NPV) (prevalence adjusted) ≥99% and Sensitivity ≥96%. Of the 97 mild TBI subjects evaluated, 14 had positive head CT scan results. All 14 of the subjects with positive head CT results had a positive TBI interpretation (Sensitivity 100.0%; 95% CI:78.5%,100.0%). There were 23 subjects with a negative TBI interpretation, all 23 had a negative head CT scan result. The NPV of the test was 100% (95% CI:85.7%,100.0%). Conclusion The Al