Glial Fibrillary Acidic Protein as a Plasma Biomarker of Brain Injury in Children with Sickle Cell Disease.

In children with sickle cell disease (SCD), silent cerebral infarct (SCI) is an independent risk factor for lower IQ, poorer school performance, and overt stroke. MRI is the only method to identify patients with SCI. A blood biomarker of SCI in patients with sickle cell disease (SCD) would fill a clinical void because blood is easy to obtain and measure, a biomarker may determine risk of or progression of neurologic injury to overt stroke, and a biomarker could benchmark current and new therapies for SCI and other subclinical forms of neurologic injury in SCD. We hypothesized that an unbiased proteomics screen of plasma from sickle cell patients would yield biomarkers of brain injury.

A cross-sectional sample of children 5-14 years old with sickle cell disease (HbSS and HbSB⁰) who were screened for the Silent Infarct Transfusion (SIT) Trial (ClinicalTrials.gov NCT00072761) were studied (n=258) along with 60 age-matched healthy controls and 28 adults with overt brain injury. For biomarker discovery, plasma underwent depletion of hemoglobin with nickel- nitrilotriacetic acid and immunoaffinity depletion of the 12 most abundant plasma proteins. After fractionation by reverse phase liquid chromatography over a C18 column and acetonitrile gradient and trypsin digestion, peptide spectra from each fraction were obtained by LC/MS/MS (LTQ-Orbitrap) and were searched using X!Tandem and human IPI database version 3.5. Post search analysis was performed using Proteomics Analysis Software System (Integrated Analysis Inc., Bethesda, MD). An electrochemiluminescent immunoassay was developed for one of the candidate proteins identified, glial fibrillary acidic protein (GFAP), for protein validation (MesoScale Discovery).

GFAP, a brain-specific intermediate filament known to be a marker of acute stroke and head trauma in adults, was identified in discovery sample plasma. Four percent of sickle cell subjects in steady-state had plasma GFAP concentrations similar to stroke and brain surgery controls (>0.45 ng/mL). Among sickle cell subjects 5-14 years old screened for the SIT Trial, 9.3% had GFAP concentrations above the 95^th percentile of age-matched controls (95^th percentile cutoff: 0.227 ng/mL; p=0.08). Sickle cell subjects with silent cerebral infarct (SCI) had more elevations above the normal 95^th percentile (10/69; 14.5%) than those without SCI (9/131; 6.8%), although this difference was not statistically significant (p=0.08). The sensitivity of GFAP as a marker of brain injury was demonstrated in a case of an 11 year-old child with HbSS and acute stroke in which plasma GFAP was 1.52 ng/mL (6 times the normal control 95^th percentile) before the stroke was clinically evident and peaked at 2.83 ng/mL.

We discovered GFAP as a circulating brain protein in plasma of children with SCD. Plasma GFAP is a marker of acute stroke in sickle cell disease but discriminates only minimally between SCI and non-SCI status in the SIT Trial using baseline cross-sectional samples. Four percent of children with sickle disease 5-14 years old have plasma GFAP levels similar to controls with severe brain injury without clinical evidence of overt stroke.

Elevations in circulating brain proteins such as GFAP show promise as indicators of subacute brain injury in children with SCI, but will require longitudinal studies of plasma GFAP in children with SCI to clarify the utility of GFAP as a plasma biomarker of SCI and a predictor of neurologic risk.

No relevant conflicts of interest to declare.