Overall Summary of Evidence

Three studies were conducted in support of the FDA licensure request.1-4  The first study used frozen plasma samples obtained from the study conducted in support of the Banyan Biomarkers TBI Assay.2,3  This study (n=1901) demonstrated comparable performance to the Banyan Biomarkers assay with a sensitivity of 95.8%, negative predictive value of 99.3%, and specificity of 40.4% [95%CI 38.2%, 42.7%]. There were n=5 false negative test results; none of these were abnormalities requiring surgical intervention.2

The second study used a small sample of fresh plasma samples (n=88) obtained from a small subset of patients a study of TBI at Level 1 Trauma Centers who had head CT performed.3 In this sample the sensitivity and negative predictive value were 100%, but the specificity was lower - 23.7%. The reason for the lower specificity in this population is not clear but may have been due to the small sample size or due to differences in the study population. For example, only 6% of the population in the first study had findings on head CT as compared to this study where 33% had findings on head CT. The positive predictive values for the two studies were comparable after adjustment for the prevalence of positive head CT.

The third study used venous whole blood specimens collected within 24 hours of head injury. The study (n=970) demonstrated comparable performance to the earlier plasma studies, with a sensitivity of 96.5%, negative predictive value of 96.5% (adjusted NPV at 6% prevalence of CT+ was 99.4%), and specificity was 40.3%. There were n=10 false negative results and none of these required surgical intervention.

While the specificity of the assay is low to moderate for brain injury and hemorrhage visible on CT, additional research suggests that many individuals with elevated Glial Fibrillary Acidic Protein using a prototype of the assay have evidence of brain injury on MRI.5  It is important to note the assay is not FDA approved for this purpose.

Summary of Study #1 

Stored frozen plasma samples were obtained from Bazarian, et al study used for approval of Banyan Biomarkers TBI assay, an early version of this assay.1,2

  • Inclusion/Exclusion Criteria: Adults ≥18 years, presenting to Emergency Department with non-penetrating TBI and GCS 13-15, where the referring provider felt a head CT was indicated, with blood test within 24 hours of injury at 22 sites in the U.S. and Europe
  • The study population (n=1901) had a median age of 49.0 years and ranged from 18 to 98 years. Over half (56.6%) were male and 70.6% were white and 26.2% were African American.
  • Nearly all had a GCS of 15 (94.1%) and around half had Loss of Consciousness (42.2%), Alteration of Consciousness (56.3%), or visible trauma above the clavicle (63.3%). Around one-third reported post-traumatic amnesia (33.0%).
  • The median time between injury and blood draw was 3.2 hours with a range of 0.3 to 11.9 hours.
Table 1. Clinical Performance 

There were five individuals with false negative results (i.e. not elevated biomarker result and findings on head CT). None of these individuals required surgical intervention. Findings included a small sub-arachnoid hemorrhage, a small subdural hemorrhage and a venous angioma thought to be a congenital anomaly.  See Figure 1 for the 5 head CT scans from false negative subjects.

Figure 1. Head CT Scans from false negative subjects                    

Summary of Study #2

Fresh plasma samples were obtained from four clinical sites of the Transforming Research and Clinical Knowledge in

Traumatic Brain Injury (TRACK-TBI) study in the United States.3

  • Inclusion/Exclusion Criteria: Adults ≥18 years, presenting to Emergency Department with GCS 13-15, who had a head CT performed at 4 sites in the U.S.
  • The study population (n=88) had a mean age of 42.5 years and ranged from 18 to 85 years. Nearly three quarters (71.2%) were male.
  • Most had a GCS of 15 (81.8%) and over half had loss of consciousness (68.2%), presence of confusion (67.0%), or post-traumatic amnesia (68.2%).
  • The median time between injury and blood draw was 4.3 hours with a range of 2.0 to 11.8 hours.
Table 2. Clinical Performance - Supplemental Fresh Specimen Study 

Summary of Study #3 

A prospective, multi-center, observational study was conducted to evaluate the clinical performance of the i-STAT TBI cartridge in classifying intended use population subjects with suspected mild TBI for the likely absence of acute intracranial lesions visualized by a head CT scan. Venous whole blood specimens were collected in K2EDTA within 24 hours of the head injury.4

  • Inclusion/Exclusion Criteria: Adults ≥18 years, presenting to Emergency Department with non-penetrating TBI and GCS 13-15, who had a head CT ordered as part of their standard of care, with blood test within 24 hours of injury at 20 Transforming Research and Clinical Knowledge in Traumatic Brain Injury (TRACK-TBI)
  •  The study population (n=970) had a median age of 46.0 years and ranged from 18 to 97 years. Over half (64.0%) were male and 68.6% were white and 17.7% were African American.
  • A majority had a GCS of 15 (78.6%) and about two thirds had Loss of Consciousness (69.6%), Alteration of Consciousness (72.1%), visible trauma above the clavicle (65.6%) and post-traumatic amnesia (62.4%).
  • The median time between injury and blood draw was 8.1 hours with a range of 1.5 to 24 hours.
Table 3. Clinical Performance 

Summary of GFAP/MRI Study 

Yue JK, Yuh EL, Korley FK, et al. Association between plasma GFAP concentrations and MRI abnormalities in patients with CT-negative traumatic brain injury in the TRACK-TBI cohort: a prospective multicentre study. Lancet Neurol. 2019;18(10): 953-961. https://doi.org/10.1016/s1474-4422(19)30282-0

The study enrolled adults with GCS 13-15 between 2014-2018 at 18 participating Level 1 U.S. trauma centers who presented within 24 hours of injury and had a head CT as well as an MRI within 7-18 days post injury.5

  • Of 1375 individuals enrolled in TRACK-TBI study, 794 had negative head CT and a GCS of 13-15 of whom 450 had an MRI scan performed within 7-18 days. Of these, 27% had positive MRI.
  • Individuals with positive head CT had the highest GFAP levels (Median 786.0 pg/mL); followed by those with negative head CT and positive MRI (Median 414.4 pg/mL); followed by those with negative head CT and negative MRI (Median 74.0 pg/mL). Of note, healthy (Median 8.0 pg/mL) and orthopedic trauma (Median 13.1 pg/mL) controls had significantly lower GFAP levels (Table 4/Figure 2).
  • From a pathophysiologic perspective, elevated GFAP levels were noted in traumatic axonal injury, diffuse axonal injury, extra-axial and mixed lesions (See Figure 2).
  • The sensitivity and specificity of GFAP levels for positive MRI is listed in Table 5. Of note, GFAP levels greater than 282.70 pg/mL has a specificity of 80.3% for positive MRI.
Table 4. WB GFAP concentrations by imaging modality and findings. 
Figure 2. GFAP concentration by MRI pathology 
Table 5. Cutoff concentrations of WB GFAP to predict MRI-positive vsersus MRI-negative findings in patients with negative CT

REFERENCES 

  1. Bazarian JJ, Biberthaler P, Welch RD, et al. Serum GFAP and UCH-L1 for prediction of absence of intracranial injuries on head CT (ALERT-TBI): a multicentre observational study. Lancet Neurol. 2018;17(9): 782-789. https://doi.org/10.1016/s1474-4422(18)30231-x.
  2. Bazarian JJ, Welch RD, Caudle K, et al. Accuracy of a rapid glial fibrillary acidic protein/ubiquitin carboxyl-terminal hydrolase L1 test for the prediction of intracranial injuries on head computed tomography after mild traumatic brain injury. Acad Emerg Med. 2021 Nov;28(11):1308-1317. doi: 10.1111/acem.14366. Epub 2021 Sep 7.
  3. Abbott Point of Care Inc., 2021, 510(k) Summary: i-STAT TBI Plasma Cartridge with the i-STAT Alinity System (K201778; Approved 8JAN2021); 2021. https://www.accessdata.fda.gov/cdrh_docs/pdf20/K201778.pdf . Accessed 16 February 2021.
  4. Abbott Point of Care, Inc., 2024, 510(k) Summary: i-STAT TBI cartridge with the i-STAT Alinity System (K234143; Approved 27MAR2024). https://www.accessdata.fda.gov/cdrh_docs/reviews/K234143.pdf . Accessed 19 Aug 2024.
  5. Yue JK, Yuh EL, Korley FK, et al. Association between plasma GFAP concentrations and MRI abnormalities in patients with CT-negative traumatic brain injury in the TRACK-TBI cohort: a prospective multicentre study. Lancet Neurol. 2019;18(10): 953-961. https://doi.org/10.1016/s1474-4422(19)30282-0 .