Evaluation of Blood Glial Fibrillary Acidic Protein as a Potential Marker in Huntington's Disease

Abstract

Objective: Huntington's disease (HD) is an autosomal dominant neurodegenerative disorder. Neurofilament light protein (NfL) is correlated with clinical severity of HD but relative data are the lack in the Chinese population. Reactive astrocytes are related to HD pathology, which predicts their potential to be a biomarker in HD progression. Our aim was to discuss the role of blood glial fibrillary acidic protein (GFAP) to evaluate clinical severity in patients with HD. Methods: Fifty-seven HD mutation carriers (15 premanifest HD, preHD, and 42 manifest HD) and 26 healthy controls were recruited. Demographic data and clinical severity assessed with the internationally Unified Huntington's Disease Rating Scale (UHDRS) were retrospectively analyzed. Plasma NfL and GFAP were quantified with an ultra-sensitive single-molecule (Simoa, Norcross, GA, USA) technology. We explored their consistency and their correlation with clinical severity. Results: Compared with healthy controls, plasma NfL (p < 0.0001) and GFAP (p < 0.001) were increased in Chinese HD mutation carriers, and they were linearly correlated with each other (r = 0.612, p < 0.001). They were also significantly correlated with disease burden, Total Motor Score (TMS) and Total Functional Capacity (TFC). The scores of Stroop word reading, symbol digit modalities tests, and short version of the Problem Behaviors Assessments (PBAs) for HD were correlated with plasma NfL but not GFAP. Compared with healthy controls, plasma NfL has been increased since stage 1 but plasma GFAP began to increase statistically in stage 2. Conclusions: Plasma GFAP was correlated with plasma NfL, disease burden, TMS, and TFC in HD mutation carriers. Plasma GFAP may have potential to be a sensitive biomarker for evaluating HD progression.

Keywords: Huntington's disease; biomarker; clinical severity; glial fibrillary acidic protein; neurofilament light protein.

Figure 1

Comparison and correlation of plasma NfL and GFAP. (A) Correlation of plasma NfL and GFAP in HD mutation carriers (n = 57). (B) Concentration of plasma NfL across disease stages (controls: n = 26; preHD: n = 15; stage 1: n = 20; stage 2: n = 11; stage 3: n = 10). (C) Concentration of plasma GFAP across disease stages (controls: n = 26; preHD: n = 15; stage 1: n = 20; stage 2: n = 11 stage 3: n = 10). (D) ROC curves for discrimination between controls (n = 26) and HD mutation carriers (n = 57) [95% confidence intervals (CIs) for AUCs: plasma NfL, 0.810–0.958, p < 0.0001; plasma GFAP, 0.657–0.868, p < 0.0001]. (E) ROC curves for discrimination between controls (n = 26) and HD mutation carriers (n = 57) with combination of plasma NfL and GFAP (95% CIs for AUCs: 0.814–0.960, p < 0.0001). (F) ROC curves for discrimination between preHD (n = 15) and manifest HD participants (n = 42) (95% CIs for AUCs: plasma NfL, 0.929–1.000, p < 0.0001; plasma GFAP, 0.496–0.822, p = 0.070). (G) ROC curves for discrimination between preHD (n = 15) and manifest HD participants (n = 42) with combination of plasma NfL and GFAP (95% CIs for AUCs: 0.934–1.000, p < 0.0001). Significance level was defined as p < 0.05. ^*p < 0.05, ^**p < 0.01, and ^****p < 0.0001. NfL, neurofilament light protein; GFAP, glial fibrillary acidic protein; HD, Huntington's disease; ROC, receiver operating characteristic.

Figure 2

Correlation of plasma NfL and clinical measures. (A) The concentration of plasma NfL in healthy controls (n = 26) and HD mutation carriers (n = 57). (B) Plasma NfL was significantly correlated with disease burden in HD mutation carriers. (C) Plasma NfL was significantly correlated with Total Motor Score in HD mutation carriers. (D) Plasma NfL was significantly correlated with Total Functional Capacity in HD mutation carriers. (E) Plasma NfL was significantly correlated with the scores of symbol digit modalities and Stroop word reading tests in HD mutation carriers. (F) Plasma NfL was significantly correlated with the score of the short version of the Problem Behaviors Assessment for HD. Scatter plots show unadjusted values. R and p values are age-adjusted, generated from Pearson's partial correlations including age as a covariate. The significance level was defined as p < 0.05. NfL, neurofilament light protein.

Figure 3

Correlation of plasma GFAP and clinical measures. (A) The concentration of plasma GFAP in healthy controls (n = 26) and HD mutation carriers (n = 57). (B) Plasma GFAP was significantly correlated with disease burden in HD mutation carriers. (C) Plasma GFAP was significantly correlated with Total Motor Score in HD mutation carriers. (D) Plasma GFAP was significantly correlated with Total Functional Capacity in HD mutation carriers. (E) Plasma GFAP had no correlation with the scores of symbol digit modalities and Stroop word reading tests in HD mutation carriers. (F) Plasma GFAP had no correlation with the score of the short version of the Problem Behaviors Assessment for HD. Scatter plots show unadjusted values. R and p values are age-adjusted, generated from Pearson's partial correlations including age as a covariate. The significance level was defined as p < 0.05. GFAP, glial fibrillary acidic protein.