Zhu et al. Translational Neurodegeneration (2021) 10:50 https://doi.org/10.1186/s40035-021-00275-w RESEARCH Open Access Plasma glial fibrillary acidic protein and neurofilament light chain for the diagnostic and prognostic evaluation of frontotemporal dementia Nuole Zhu1,2,3, Miguel Santos‑Santos1,2, Ignacio Illán‑Gala1,2, Victor Montal1, Teresa Estellés1,2, Isabel Barroeta1,2, Miren Altuna1,2, Javier Arranz1,2, Laia Muñoz2, Olivia Belbin1,2, Isabel Sala1,2, Maria Belén Sánchez‑Saudinós1,2, Andrea Subirana1,2, Laura Videla1,2,4, Jordi Pegueroles1, Rafael Blesa1, Jordi Clarimón1,2, Maria Carmona‑Iragui1,2,4, Juan Fortea1,2,4, Alberto Lleó1,2,3* and Daniel Alcolea1,2,3* Abstract Background: Astrocytes play an essential role in neuroinflammation and are involved in the pathogenesis of neuro‑ denegerative diseases. Studies of glial fibrillary acidic protein (GFAP), an astrocytic damage marker, may help advance our understanding of different neurodegenerative diseases. In this study, we investigated the diagnostic performance of plasma GFAP (pGFAP), plasma neurofilament light chain (pNfL) and their combination for frontotemporal dementia (FTD) and Alzheimer’s disease (AD) and their clinical utility in predicting disease progression. Methods: pGFAP and pNfL concentrations were measured in 72 FTD, 56 AD and 83 cognitively normal (CN) par‑ ticipants using the Single Molecule Array technology. Of the 211 participants, 199 underwent cerebrospinal (CSF) analysis and 122 had magnetic resonance imaging. We compared cross-sectional biomarker levels between groups, studied their diagnostic performance and assessed correlation between CSF biomarkers, cognitive performance and cortical thickness. The prognostic performance was investigated, analyzing cognitive decline through group com‑ parisons by tertile. Results: Unlike pNfL, which was increased similarly in both clinical groups, pGFAP was increased in FTD but lower than in AD (all P < 0.01). Combination of both plasma markers improved the diagnostic performance to discriminate FTD from AD (area under the curve [AUC]: combination 0.78; pGFAP 0.7; pNfL 0.61, all P < 0.05). In FTD, pGFAP corre‑ lated with cognition, CSF and plasma NfL, and cortical thickness (all P < 0.05). The higher tertile of pGFAP was associ‑ ated with greater change in MMSE score and poor cognitive outcome during follow-up both in FTD (1.40 points annually, hazard ratio [HR] 3.82, P < 0.005) and in AD (1.20 points annually, HR 2.26, P < 0.005). Conclusions: pGFAP and pNfL levels differ in FTD and AD, and their combination is useful for distinguishing between the two diseases. pGFAP could also be used to track disease severity and predict greater cognitive decline during follow-up in patients with FTD. *Correspondence: alleo@santpau.cat; dalcolea@santpau.cat 1 Sant Pau Memory Unit, Department of Neurology, Institut d’Investigacions Biomèdiques Sant Pau – Hospital de Sant Pau, Universitat Autònoma de Barcelona, 08041 Barcelona, Spain Full list of author information is available at the end of the article © The Author(s) 2021. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​ mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Zhu et al. Translational Neurodegeneration (2021) 10:50 Page 2 of 12 Keywords: Glial fibrillary acidic protein, Neurofilament, Frontotemporal dementia, Plasma biomarkers Introduction secondary aim was to study the correlation of pGFAP Frontotemporal dementia (FTD) is a progressive neuro- with cognition, other biomarkers, and structural meas- degenerative condition characterized by clinical, genetic ures in neuroimaging. and neuropathologic heterogeneity. The clinical mani- festation of FTD may overlap with psychiatric or other Methods neurodegenerative disorders, such as Alzheimer’s disease Study participants and classification (AD). Diagnosis is thus a clinical challenge. We collected clinical and biomarker information of 211 In the past two decades, many efforts have been made participants with available plasma samples from the Sant to find imaging or fluid biomarkers for FTD [1–6]. In Pau Initiative on Neurodegeneration (SPIN) cohort, a spite of the notable advances, we still do not have patho- multimodal biomarker platform for the study of neurode- physiological markers of FTD to be used in clinical generative diseases [28]. The participants were classified practice. Recently, some studies have shown that the con- into the following clinical groups according to the inter- centrations of neurofilament light chain (NfL), a marker nationally accepted diagnostic criteria: 72 patients with of axonal damage, are increased in the cerebrospinal fluid probable FTD-related clinical syndromes: 33 patients (CSF) and blood of FTD patients [6]. Moreover, their lev- with behavioral variant of frontotemporal dementia els can reflect disease severity and predict clinical pro- (bvFTD) [29], 7 semantic variant of primary progressive gression [1–6]. However, NfL is not specific for FTD and aphasia (svPPA) [30], 14 nonfluent variant of primary can be increased in other neurodegenerative diseases, progressive aphasia (nfvPPA) [30] and 18 progressive such as AD and Lewy body dementia [6]. There is a con- supranuclear palsy-corticobasal syndrome spectrum siderable overlap of NfL level between different condi- (PSP-CBD) [31, 32]. CSF AD biomarkers and/or amyloid tions [1–6]. PET were taken into account during the diagnostic classi- Neuroinflammation and neurodegeneration are fication of FTD patients, and those with positive AD bio- highly interrelated processes in neurological diseases markers were excluded from this group. We also included [7]. Increased levels of astrogliosis marker glial fibril- 56 AD patients with evidence of AD pathophysiology lary acidic protein (GFAP) in CSF and plasma, have [33] either through CSF biomarkers (n = 51) or amyloid- been described in different neurodegenerative diseases PET (n = 5), and 83 CN that had normal CSF values of [8–19, 20, 23–27]. In AD, recent studies showed that the Aβ42/Aβ40 and pTau181, and neuropsychological evalua- plasma GFAP (pGFAP) levels are associated with amy- tion within normal range [28]. loid pathology [15–18]. Furthermore, some studies have Plasma availability was the prerequisite to include the suggested that GFAP could be a marker of disease sever- participants in this study. CSF samples were available ity [8–22] and be a prognostic marker for progression to from 199 participants (94%). A subset of 122 participants AD dementia in cognitively normal older adults [15–22]. (58%) underwent 3 Tesla structural brain MRI. APOE Likewise, other studies have reported increased CSF genotype was available in 203 participants (96%), and 202 GFAP and pGFAP concentrations in FTD, even though participants (96%) were longitudinally followed up and the results are not consistent across studies [10, 12, 13, underwent a comprehensive evaluation. All participants 23–27]. Although previous studies did not detect changes had a Mini-Mental State Examination (MMSE) score of blood GFAP level in FTD [10, 12, 13], some studies and a Global Deterioration Scale of Reisberg score at the have observed elevated pGFAP in all different FTD sub- time of diagnosis, and repeated measures of MMSE were groups [24,25, 26]. Higher pGFAP concentrations have obtained during the follow-up. In a subset of 45 FTD been associated with greater functional impairment and patients, Frontotemporal Dementia Rating Scale (FTD- disease severity in two studies [13, 25], but the role of FRS) was available at the time of diagnosis. pGFAP in predicting disease progression of FTD and its potential use for diagnosis, either alone or in combina- Blood and CSF sample analysis tion with other biomarkers, remain unclear. Blood samples were collected in 10-ml EDTA tubes and In this study, our primary research aim was to deter- immediately transferred to our laboratory where they mine the potential use of pGFAP, pNfL or their combi- were centrifuged and aliquoted within 2 h after collec- nation to distinguish patients with FTD from AD and tion. CSF samples were collected on the same day of from cognitively normal participants (CN), and to study blood extraction and processed in polypropylene tubes their association with disease progression in FTD. The following international recommendations [33]. All Zhu et al. Translational Neurodegeneration (2021) 10:50 Page 3 of 12 samples were processed and aliquoted within the first Statistical significance for all tests was set at 5% two hours after lumbar puncture. The plasma and CSF (P = 0.05) and corrected for multiple comparisons. All aliquots were stored at − 80 °C until analysis. pGFAP statistical analyses were performed using packages and pNfL concentrations were measured using the SR-X “psych” (v. 2.0.8), “ggplot2” (v. 3.3.3), “pROC” (v. 1.16.2), single molecule array. CSF AD core biomarkers (Aβ42, “lmerTest” (v. 3.1-3), “nlme” (v. 3.1-148), “multcomp” (v. Aβ40, tTau and pTau181) were measured in the fully- 1.4-13), “survival” (v. 3.2-11) and “survminer” (v. 0.4.9) automated platform Lumipulse (Fujirebio-Europe), and as implemented in R statistical software version R 4.0.2 (MicroVue™, Quidel) were measured by ELISA according levels of CSF NfL (Uman Diagnostics) and CSF YKL-40 (http://​www.R-​proje​ct.​org). to previously reported methods [7, 33]. All samples and calibration curves were measured in duplicate. Intra- and Results inter-assay coefficients of variations were 5.3% and 9.6%, Demographics and clinical data respectively. The pre-analytical processing protocol for Table 1 shows the demographics and clinical data, CSF blood and CSF in the SPIN cohort has been described in and plasma biomarker levels in the FTD, AD and CN detail [28, 29]. groups. Controls were significantly younger than symp- tomatic patients, but there was no difference in age between the symptomatic groups. Controls had more Statistical analysis years of education than the symptomatic groups, and Chi-square test was used to compare sex and APOE ε4 there were no differences in sex between the groups. genotype frequency between the groups. Continuous APOEε4 genotype was more frequent in the AD group. variables are expressed as means and standard devia- Both baseline and last MMSE scores were lower in the tion. Distributions of demographic and biomarker data disease groups compared with the control group. Within were assessed using Shapiro–Wilk’s test and homogene- the FTD subgroups, the baseline and last MMSE scores ity of variances was checked by Levene’s test. Biomarker were lower in svPPA compared to other subgroups. The raw values not following a normal distribution were log- mean follow-up period was 3.4 (± 2.3) years in FTD, 3.9 transformed to achieve a normal distribution. Age- and (± 2) in AD and 4 (± 1.7) in CN. The follow-up period did sex-adjusted analysis of covariance (ANCOVA) followed not differ between the groups. by post-hoc Tukey’s test was used to compare the pGFAP Age correlated with the pGFAP level (Rho = 0.35, and pNfL levels between groups. The potential use of P < 0.001) in the control group, but not with the pNfL pGFAP, pNfL or their combination to distinguish patients level. No difference in pGFAP level was found between with FTD from AD and CN was assessed through areas male and female participants. pNfL levels were higher in under the curve (AUC) by receiver operating character- males in the whole sample (P = 0.05) and in the AD group istic (ROC) analyses. ROC curves were compared using (P = 0.04), but not in the CN or the FTD group. DeLong’s test. To study the association of pGFAP with disease progression, we divided participants in 3 tertile groups according to their pGFAP levels. A linear-mixed Plasma GFAP and NfL concentrations differ between FTD model was used to assess the association of pGFAP with and AD cognitive decline during the follow-up. We included age, After adjusting for age and sex, pGFAP and pNfL lev- sex, MMSE score at the time of diagnosis, pGFAP tertile els differed between FTD and AD patients (Fig. 1). and its interaction with time as fixed factors, and mod- The level of pGFAP was higher in patients with FTD eled random intercepts and slopes at the participant (234.9 ± 141.9 pg/ml) than in CN (134.3 ± 45.4 pg/ml, level to account for repeated measures. Multivariate Cox P = 0.0008) but lower than in AD (319.8 ± 135.1 pg/ml, regression analysis adjusting for sex and baseline age and P < 0.001). Within the FTD subgroups, the pGFAP level Kaplan Meier curves were performed to analyze the pre- was significantly higher in svPPA and PSP-CBD com- dictive value of pGFAP for significant cognitive impair- pared with controls (P = 0.03), and significantly lower in ment (MMSE score < 20). To investigate the relationship bvFTD than in AD (P < 0.001) (Fig. 1a, b). of pGFAP level with demographics, cognitive scores, CSF As expected, the pNfL levels were higher in both FTD biomarkers and pNfL level, Spearman’s correlations were (37.6 ± 42.3 pg/ml, P < 0.001) and AD (26.5 ± 13.3 pg/ml, assessed. Cortical thickness was computed with Free- P = 0.004) groups compared to CN (18 ± 20.2 pg/ml). surfer software (version 5.1) [35] (https://​surfer.​nmr.​mgh.​ The pNfL levels were increased in all FTD subgroups harva​rd.​edu/​fswiki), and its correlation with pGFAP was (P < 0.05) compared to CN, except in svPPA, which assessed using the general linear model, including age showed a similar trend as other FTD subgroups but did and sex as fixed factors. not reach statistical significance (Fig. 1c, d). Zhu et al. Translational Neurodegeneration (2021) 10:50 Page 4 of 12 Table 1 Demographics and clinical characteristics, CSF and plasma biomarker concentrations in CN, FTD and AD groups CN FTD AD bvFTD nfvPPA svPPA PSP-CBD All FTD n 83 33 14 7 18 72 56 Age, years 58 (8.5)b,c 68.3 (10.6)a 72.8 (5.6)a 73.1 (11.1)a 72.9 (5.4)a 70.8 (8.9)a 70.8 (7.2)a Sex Female 49 10 9 4 11 34 31 Male 34 23 5 3 7 38 25 APOE ε4 status ε4+ 23c 3c 3c 2c 2c 10c 27a,b,d−g c c c c c c ε4− 57 29 11 5 14 59 27a,b,d−g Follow up time, years 3.9 (1.7) 4.1 (2.5) 3.1 (2.3) 4.8 (1.7) 1.9 (1) 3.4 (2.3) 4 (2) Education, years 15.4 (4.1)b−g 12.9 (5.6)a 13.3 (4.9)a 14.8 (5.2)a 11.3 (3.7)a 12.7 (5)a 10.4 (4.7)a Baseline MMSE score 29.2 (1)b−g 25.1 (4.2)a,f 24.1 (6.8)a,f 16.5 (12.2)a,c−e,g 22.6 (6.3)a,f 23.5 (6.5)a 23.7 (3.8)a b−g a,f a,f a,c−e,g a a Last MMSE score 29.3 (1) 16.6 (10) 14 (12.2) 0.4 (0.9) 11.6 (12.4) 13.3 (11.4) 15.5 (9.5)a b−d,f−g a a a a,b pGFAP, pg/ml 134.3 (45.4) 193.1 (100.1) 224.6 (77.2) 159.3 (279.1) 271.2 (148.2) 234.9 (141.9) 319.8 (135.1)a,c b−e,g a a a a pNfL, pg/ml 18 (20.2) 41.9 (60.1) 34.9 (17.2) 31.3 (14.5) 34.3 (18.1) 37.6 (42.3) 26.5 (13.3)a b,c,f,g a a a a,c CSF NfL, pg/ml 494.8 (274.3) 1436.3 (930.6) 1944.5 (1146.7) 2394.6 (637.8) 1739.3 (1607.3) 1654.1 (1187.6) 1286.4 (1136.9)a,b,f b,c a a a a CSF YKL-40, ng/ml 190.3 (49.4) 265.8 (64.7) 274.84 (61) 311.8 (37.1) 262.3 (79.3) 268.2 (63.7) 287.3 (68.1)a c c c c c c CSF Aβ42/Aβ40 0.099 (0.016) 0.098 (0.015) 0.092 (0.012) 0.103 (0.0002) 0.095 (0.012) 0.097 (0.013) 0.046 (0.0099)a,b,d−g c c c c c CSF tTau, pg/ml 264 (103.9) 388.9 (182.3) 275.4 (144.5) 367.5 (0.7) 276.1 (122.7) 332.7 (162) 774.2 (390.8)a,b,d,f,g CSF pTau, pg/ml 39.8 (17.8)c 45.7 (16.1)c 42.8 (21.9)c 47.2 (10) 33.9 (15)c 42.1 (16.4)c 121.2 (71)a,b,d,f,g MMSE, Mini-Mental State Examination; pGFAP, plasma glial fibrillary acidic protein; pNfL, plasma neurofilament light chain; CN, cognitivelly normal participants; FTD, Frontotemporal dementia; AD, Alzheimer disease; bvFTD, behavioral variant of frontotemporal dementia; nfvPPA, nonfluent variant of primary progressive aphasia; svPPA, semantic variante of primary progressive aphasia; PSP-CBD, progressive supranuclear palsy-corticobasal syndrome spectrum a Different from Control (P < 0.05) b Different from FTD (P < 0.05) c Different from AD (P < 0.05) d Different from bvFTD (P < 0.05) e Different from fnvPPA (P < 0.05) f Different from svPPA (P < 0.05) g Different from PSP-CBD (P < 0.05). Data are shown as mean (standard deviation) Diagnostic accuracy of plasma biomarkers and their of the two plasma markers, individually or in combi- combination to differentiate between FTD, AD and CN nation, showed higher accuracy than the basic model. We next explored the diagnostic performance of However, the addition of pGFAP and pNfL to the basic pGFAP, pNfL and their combination using logistic model (panel model) led to better diagnostic perfor- regression to discriminate across the different groups. mance than the basic model both in differentiating FTD The basic model included age, sex and APOEε4 allele from AD (AUC 0.83 [95%CI 0.75–0.9] vs AUC 0.69 status. The diagnostic performance of this basic model [95%CI 0.6–0.79], P = 0.003) and AD from CN (AUC was then compared with a panel including the variables 0.96 [95%CI 0.93–1] vs AUC 0.89 [95%CI 0.83–0.95], in the basic model together with pGFAP and pNfL. P = 0.0007). For differentiation of FTD from CN, the ROC curve analysis (Fig. 2) showed that the combina- panel showed highest AUC compared to other meth- tion of pGFAP and pNfL had higher accuracy than the ods (all P < 0.05), but not compared to the basic model two plasma markers separately in differentiating FTD (AUC 0.88 [95%CI 0.83–0.94] vs AUC 0.87 [95%CI from AD (combination: AUC 0.78 [95%CI 0.70–0.86]; 0.81–0.93], P = 0.08). pGFAP: AUC 0.7 [95%CI 0.61–0.71]; pNfL: AUC 0.61 Taking into account that pGFAP increases with age, we [95%CI 0.51–0.71], P = 0.04 and P = 0.002, respec- repeated the same analyses in an age-matched subgroup. tively) and for the discrimination between FTD and CN In this new comparison, similar results were observed (combination: AUC 0.82 [95%CI 0.75–0.89]; pGFAP: except in the comparison between FTD and CN, in AUC 0.76 [95%CI 0.68–0.84]; pNfL: AUC 0.81 [95%CI which both pNfL and the combination showed higher 0.74–0.88], P = 0.02 and P = 0.51, respectively). None diagnostic performance compared to this basic model Zhu et al. Translational Neurodegeneration (2021) 10:50 Page 5 of 12 Fig. 1 Comparison of pGFAP and pNfL in different groups. a pGFAP in the FTD, AD and cognitively normal groups; b pGFAP concentration in all FTD subgroups, AD and cognitively normal groups; c pNfL concentration in the FTD, AD and cognitively normal groups; d pNfL concentration in all FTD subgroups, AD and cognitively normal groups. *P < 0.05, **P < 0.01, ***P < 0.001. pGFAP, plasma glial fibrillary acidic protein; pNfL, plasma neurofilament light chain; CN, cognitively normal participants; FTD, frontotemporal dementia; AD, Alzheimer’s disease (Combination: AUC 0.8 [95%CI 0.7–0.9]; pNfL: AUC 0.8 The pGFAP level was significantly correlated with AD- [95%CI 0.7–0.9]; Basic model: AUC 0.59 [95%CI 0.47– specific CSF biomarkers, such as the Aβ42/Aβ40 ratio, 0.72], P = 0.05 and P = 0.005) (Additional file 1). amyloid-PET centiloid, total tau and phosphorylated tau in the whole sample, but these correlations were not sig- nificant within diagnostic subgroups (Additional file 2). pGFAP correlated with other fluid biomarkers, cognitive After adjusting for age and sex, pGFAP was also signifi- and functional scores and neuroimaging cantly associated with the baseline MMSE scores in the In the whole sample, pGFAP level correlated with pNfL whole sample (Rho =  − 0.58, P < 0.001) and in the FTD (Rho = 0.53, P < 0.001), CSF NfL (Rho = 0.52, P < 0.001) group (Rho =  − 0.33, P = 0.005), but not in the AD or the and CSF YKL-40 levels (Rho = 0.49, P < 0.001) after CN group (Fig. 3d). In the FTD group, after adjusting for adjusting for age and sex. To avoid group effects on the age, higher pGFAP concentrations were associated with correlation assessment, we studied correlations of these lower FTD-FRS scores (r =  − 0.28, P = 0.046, Fig. 3e). biomarkers within groups. The pGFAP level correlated We assessed the correlation between pGFAP and corti- with plasma and CSF NfL in FTD (Rho = 0.49, P < 0.001; cal thickness in a subset of 122 participants (29 FTD, 25 Rho = 0.32, P < 0.02) and in CN (Rho = 0.4, P < 0.001; AD and 68 CN) with structural MRI suitable for quanti- Rho = 0.3, P = 0.016), but not with CSF YKL-40. By con- tative analyses. After adjusting for age, sex and neuroim- trast, in the AD group, pGFAP level correlated with pNfL aging acquisition center, vertex-wise regression analysis level only (Rho = 0.35, P = 0.007) (Fig. 3a–c). showed a significant association between pGFAP and Zhu et al. Translational Neurodegeneration (2021) 10:50 Page 6 of 12 Fig. 2 Receiver operating characteristic curves (ROC) of pGFAP, plasma neurofilament light chain, their combination, a basic model with risk factors (age, sex and APOE), and a combination of plasma biomarkers with additional risk factors to discriminate FTD from AD (a), FTD from normal cognitively normal (b), and AD from cognitively normal (c). AUC, area under the curve; pGFAP, plasma glial fibrillary acidic protein; pNfL, plasma neurofilament light chain; CN, cognitively normal participants; FTD, frontotemporal dementia; AD, Alzheimer’s disease cortical thickness in the orbitofrontal and occipital pole (1.40 points annually, HR = 3.82 in FTD; 1.20 points regions (Fig. 3f, P < 0.05) in the FTD group. In the AD and annually, HR = 2.26 in AD, both P < 0.001). the CN groups, no correlation between pGFAP concen- tration and cortical thickness was found (not shown). Discussion In this study, we found that pGFAP was increased in Baseline pGFAP level predicts cognitive decline FTD, and that its level differs from that in AD patients. To investigate the prognostic performance of pGFAP pGFAP also correlated with neuronal damage biomark- levels, we divided pGFAP levels in tertiles within ers, cognitive and functional scores and structural imag- each symptomatic group: low (< 154 pg/ml in FTD ing measures in FTD. Importantly, a higher pGFAP level and < 228 pg/ml in AD), medium (154–240 pg/ml in FTD was associated with a greater change in MMSE score and 228–366 pg/ml in AD) and high (> 240 pg/ml in FTD during follow-up and predicted progression to moderate and > 366 pg/ml in AD). cognitive impairment in FTD. Our findings suggest that Linear-mixed model analysis was used to assess the pGFAP could be a useful marker of disease severity and relationship between the baseline pGFAP level and cog- for prognostic assessment in FTD, in addition to its role nitive decline measured by changes in MMSE score in AD. during follow-up. After adjusting for age, sex and base- Emerging evidence suggests that astrocytes play an line MMSE score, patients in the highest pGFAP tertile essential role in neuroinflammation and are involved in showed a greater change in MMSE score compared to the pathophysiology of several neurological diseases [34, those in the lowest tertile in both FTD (mean loss of 1.40 36, 38]. Biomarkers that track astrocyte biology, such as points annually, P = 0.003) and AD (mean loss of 1.20 GFAP, have been investigated over the past few years in points annually, P < 0.001) (Fig. 4a, b). neurodegenerative diseases. In AD, previous studies have Kaplan Meier curves (Fig. 5) and Cox regression analy- shown that reactive astrocytes are closely associated ses including baseline age and sex as covariates were used with senile plaques and neurofibrillary tangles [37–39]. to assess progression to moderate cognitive impairment Elevated concentrations of GFAP in CSF and blood have (MMSE score < 20). Compared to the lowest tertile, the been observed in AD, and it has been reported that its highest tertile of pGFAP was associated with increased levels correlate with disease severity [8–16, 19]. Recent risk of poor cognitive outcome both in FTD and in AD studies have shown that the elevated CSF and plasma Zhu et al. Translational Neurodegeneration (2021) 10:50 Page 7 of 12 Fig. 3 Correlation of pGFAP with plasma and CSF neurofilament light chain (a, b), CSF YLK-40 (c), and baseline MMSE score (d) in FTD, AD and cognitively normal participants. Correlation of pGFAP with baseline FTD-FRS score (e) and cortical thickness in FTD (f, n = 29). Blue regions in f represent a direct correlation. For illustrative purposes, a scatterplot shows the individual log(pGFAP) and the value of cortical thickness in the corresponding cortical region (marked with an asterisk) (f). P-value ­10−1.3 = 0.05. All correlation analyses were adjusted by age and sex. pGFAP, plasma glial fibrillary acidic protein; pNfL, plasma neurofilament light chain; CN, cognitively normal participants; FTD, frontotemporal dementia; AD, Alzheimer’s disease GFAP are associated with amyloid pathology, also in cog- FTD diagnosis. According to this hypothesis, various nitively unimpaired subjects [15–20]. In addition, astro- studies have observed elevated pGFAP concentration in gliosis is also recognized in FTD. Astrogliosis is mainly different FTD subgroups and both in presenile and late- confined to concrete regions such as the frontal cortex onset bvFTD cases [24, 25, 27]. In addition, one study and the hippocampus [40, 41]. Interestingly, increased in genetic FTDs found increased pGFAP concentration CSF and plasma GFAP levels have been reported in both in progranulin-associated symptomatic FTD patients sporadic and genetic FTD cohorts [10–12, 14, 23]. These [23]. Here we report an increased pGFAP level in FTD findings suggest that pGFAP may be a valuable tool in patients, which was, however, lower than that of AD. Our Zhu et al. Translational Neurodegeneration (2021) 10:50 Page 8 of 12 Fig. 4 Relationship between the baseline tertile of pGFAP and longitudinal changes of cognitive score (MMSE score) in FTD (a) and AD (b). Cognitive changes were estimated through linear mixed effects regression models adjusted for age and baseline MMSE score. Red lines show the lowest tertile (below 154 pg/ml in FTD and below 228 pg/ml in AD), green lines show the medium tertile (154–240 pg/ml in FTD and 226–366 pg/ ml in AD), and blue lines show the highest tertile of pGFAP level (above 240 pg/ml in FTD and above 366 pg/ml in AD). Shaded areas indicate the 95% confidence interval for predicted cognitive scores. pGFAP, plasma glial fibrillary acidic protein; pNfL, plasma neurofilament light; CN, cognitivelly normal participants; FTD, frontotemporal dementia; AD, Alzheimer’s disease Fig. 5 Kaplan Meier curves of clinical progression to significant cognitive impairment (MMSE score < 20) in FTD (a) and AD (b) individuals with low, medium or high baseline pGFAP level. Red lines show the lowest tertile (below 154 pg/ml in FTD and below 228 pg/ml in AD), green lines show medium tertile (154–240 pg/ml in FTD and 226–366 pg/ml in AD), and blue lines show the highest tertile of biomarker concentrations (above 240 pg/ml in FTD and above 366 pg/ml in AD). In FTD, the median time of progression to significant cognitive impairment was 3.4 (0.8–7.0) years in the medium-tertile pGFAP group and 1.1 (0–2.3) years in the high-tertile pGFAP group (Log-rank test, P < 0.0001). pGFAP, plasma glial fibrillary acidic protein; FTD, frontotemporal dementia; AD, Alzheimer’s disease results are in line with previous reports of higher concen- between groups [20, 24]. These discrepancies could be trations of pGFAP in AD compared to non-AD demen- explained by the heterogeneity of the patient composi- tias [13, 18]. However, other studies found no differences tion: some studies included different types of dementias Zhu et al. Translational Neurodegeneration (2021) 10:50 Page 9 of 12 in the non-AD dementia group and little was known Previous studies have reported a weak correlation about the function of pGFAP in such contexts, such as between CSF GFAP and YKL-40 levels in neurodegen- vascular dementia or dementia with Lewy bodies. Our erative diseases [11, 12]. Similarly, we did not find signifi- results support previous clinical-pathological findings cant correlation between pGFAP and CSF YKL-40 levels. that in FTD, astrocytic activation is at a lower extent This finding supports the hypothesis that different astro- and is more confined to specific brain regions than that cyte subpopulations or different spatial distributions are in AD [40, 41]. On the other hand, the increase in pNfL, involved in the pathophysiology of FTD and AD [12, 13, a marker of axonal damage, suggests a similar degree of 37, 38–41]. neurodegeneration in FTD compared to AD. One pos- The association between elevated pGFAP level, disease sible explanation for the differences in pGFAP levels but severity and prognostic markers in AD has been reported not in pNfL between FTD and AD is that FTD and AD before [15, 17–22]. Previous studies reported that the are characterized by different neuroinflammatory pat- serum GFAP has potentials to predict future conversion terns, with different degrees of astroglial activation in dif- to dementia in cognitively normal individuals, not only to ferent subpopulations but leading to a similar degree of AD, but also to other dementias including FTD [16, 17, neuronal degeneration. Similar to a previous study [24], 22]. However, pGFAP as a marker of disease severity and we found that the pGFAP level tended to be different prognosis in FTD has seldom been investigated [13, 25, across the subtypes of the FTD spectrum, being signifi- 26]. In the current study, after adjusting for age and sex, cantly higher in svFTD and PSP-CBD compared to CN. pGFAP remained significantly correlated with cognitive This finding could indicate differences in the degree of and imaging measures in FTD. Higher levels of pGFAP astrocytic degeneration in different clinical phenotypes have been shown to be associated with lower brain vol- [23]. umes in GRN and C9orf72 presymptomatic carriers [23], The current study showed that pGFAP had an accept- and their association with smaller hippocampal volume able performance to discriminate FTD from AD in svPPA [26] and temporal atrophy in FTLD has been (AUC = 0.7) and CN (AUC = 0.76). These accuracies reported [27]. To our knowledge this is the first study to were somewhat lower than the values reported in two show that higher concentrations of pGFAP are also asso- recent studies using pGFAP to identify bvFTD from AD ciated with orbitofrontal cortical thickness in the whole (AUC = 0.85) and frontotemporal lobar degeneration FTD group. In line with other reports in AD patients [16, (FTLD) from primary psychiatric disorders (AUC = 0.82) 20, 22], we observed that pGFAP predicted faster cogni- [13, 27]. These discrepancies could be attributed to dif- tive decline both in FTD and AD. This is the first report ferences in cohort characteristics and heterogeneity due of the association between pGFAP and cognitive decline to different composition of FTD subtypes. An inter- in FTD; we found a higher rate of cognitive decline esting finding was that, contrary to pNfL, which tend among those with pGFAP level > 240 pg/ml. Surpris- to be higher in FTD compared to AD, the pGFAP level ingly, the relationship between pGFAP and disease sever- was significantly lower in FTD, suggesting the potential ity in AD was not confirmed in our cohort. One of the diagnostic utility of their combination. Indeed, the com- considerations when comparing our study with the pre- bination of pGFAP with pNfL improved the diagnostic vious ones is that other studies included both Aβ+ and performance in distinguishing FTD from AD and CN. Aβ− subjects [15, 22], whereas in our study the correla- Moreover, the incorporation of age, sex and APOEε4 tion analysis was performed in disease subgroups. Previ- allele status improved the diagnostic performance to ous studies have found that pGFAP is increased in early an AUC of 0.83 in differentiating FTD from AD and an stages of AD [15–22] and does not differ between disease AUC of 0.89 in differentiating FTD from CN. In addition, stages [19, 20], which could explain the lack of associa- pGFAP was particularly promising in identifying AD, tion with imaging measures in the AD subgroup. Con- consistent with previous observations [13, 15, 17–20]. sistent with previous studies [2, 7, 17], our results show Taken together, this study suggests that the combination that the glial biomarker increases also in FTD, in addition of plasma astrocytic and neuronal markers could be rel- to AD, and its level increases later in the disease course in evant in FTD. FTD than in AD, when the disease is more advanced. We observed that the pGFAP level increased with age The strength of our study is the relatively large sam- in controls, similar to what has been observed in previ- ple size, the inclusion of several FTD subgroups, regular ous studies [13, 17, 23, 25], and we took age into account follow-up of the participants and the use of multimodal when interpreting this marker. Consistent with various approach, with inclusion of clinical measures, plasma and recent reports [15, 17, 21, 23], pGFAP significantly cor- CSF assessments, and structural imaging. Additionally, related with CSF and plasma NfL, suggesting that glial all blood-based biomarkers were collected using the same activation and neuroaxonal degeneration are correlated. standard operating procedures and measured following Zhu et al. Translational Neurodegeneration (2021) 10:50 Page 10 of 12 a harmonized protocol. This study also has some limita- TV3” foundation grants 20141210 to JF, 044412 to RB and 20142610 to AL. This work was also supported by a grant from the Fundació Bancaria La Caixa to RB tions, in that it did not include serial longitudinal meas- (DABNI project). The sponsors of the study did not take part in the design and ures of pGFAP, lacked neuropathological confirmation, conduct of the study; collection, management, analysis and interpretation of and excluded AD copathologies in FTD groups. the data; writing and review of the report; or the decision to submit the article for publication. Conclusion Availability of data and materials The datasets used and/or analyses during the current study are available from In conclusion, our study shows that the pGFAP con- the corresponding author on reasonable request. centration may be useful in FTD and could improve its diagnosis. Furthermore, our data support that pGFAP is Declarations not an exclusive marker in AD, but also plays an essential role in other amyloid-independent processes involved in Ethics approval and consent to participate All procedures in the study were approved by the Hospital de Sant Pau ethics FTD. pGFAP could be used as a marker for FTD severity committee. All participants provided written informed consent. and prognosis. Consent for publication Not applicable. Abbreviations AD: Alzheimer’s disease; AUC​: Area under the curve; bvFTD: Behavioral variant Competing interest of frontotemporal dementia; CN: Cognitively normal participants; CSF: Cer‑ Nuole Zhu is employed by Hospital de la Santa Creu i Sant Pau. Declarations ebrospinal fluid; FTD: Frontotemporal dementia; FTLD: Frontotemporal lobar of interest: none. Miguel Santos is employed by Hospital de la Santa Creu i degeneration; FTD-FRS: Frontotemporal Dementia Rating Scale; GFAP: Glial Sant Pau. He is funded by a “Juan Rodés” research grant from the Institute of fibrillary acidic protein; HR: Hazard ratio; MMSE: Mini-Mental State Examina‑ Health Carlos III. Ignasio Illán-Gala is supported by the Global Brain Health tion; NfL: Neurofilament light chain; nfvPPA: Nonfluent variant of primary Institute (Atlantic Fellow for Equity in Brain Health and pilot award for global progressive aphasia; PET: Positron emission tomography; pGFAP: Plasma Glial brain health leaders CBHI ALZ UK-21-720973) and the “Juan Rodés” grant from fibrillary acidic protein; pNfL: Plasma neurofilament-light; PSP-CBD: Progressive the Institute of Health Carlos III /JR20/00018). Victor Montal is employed by the supranuclear palsy-corticobasal syndrome spectrum; ROC: Receiver operating Biomedical Research Institute Sant Pau. Declarations of interest: none. Teresa characteristic; SPIN: Sant Pau Initiative on Neurodegeneration; svPPA: Semantic Estellés is employed by Biomedical Research Institute Sant Pau. Declarations of variant of primary progressive aphasia. interest: Dr. Estellés is funded by a “Río Hortega” research grant from the Insti‑ tute of Health Carlos III. Isabel Barroeta is employed by Hospital de la Santa Creu i Sant Pau. Declarations of interest: none. Miren Altuna is employed by Supplementary Information Biomedical Research Institute Sant Pau. Declarations of interest: Dr. Estellés is The online version contains supplementary material available at https://​doi.​ funded by a “Río Hortega” research grant from the Institute of Health Carlos III. org/​10.​1186/​s40035-​021-​00275-w. Javier Arranz is employed by Biomedical Research Institute Sant Pau. Declara‑ tions of interest: none. Laia Muñoz is employed by Biomedical Research Insti‑ tute Sant Pau. Declarations of interest: none. Olivia Belbin is employed by Bio‑ Additional file 1. Diagnostic accuracy of plasma biomarkers and their medical Research Institute Sant Pau. Dr. Belbin is funded by a “Miguel Servet” combination to differentiate between FTD, AD and CN in age-matched research grant from the Institute of Health Carlos III. Isabel Sala is employed by group Hospital de la Santa Creu i Sant Pau. Declarations of interest: none. M. Belén Sánchez-Saudinós is employed by the Biomedical Research Institute Sant Pau. Additional file 2. The correlation of pGFAP with CSF AD-specific Declarations of interest: none. Andrea Subirana is employed by the Biomedical biomarkers Research Institute Sant Pau. Declarations of interest: none. Laura Videla is employed by Fundació Catalana Síndrome de Down. Declarations of interest: Acknowledgements none. Jordi Pegueroles is employed by the Biomedical Research Institute Sant The authors would like to acknowledge the Department of Medicine at the Pau. Declarations of interest: none. Rafael Blesa is employed by Hospital de Universitat Autònoma de Barcelona. We are grateful to all participants in the la Santa Creu i Sant Pau and received research grants from Institute of Health study and their families. We also thank all the clinical team members who Carlos III, Fundació Bancària Obra Social La Caixa and Fundació La Marató were involved in the selection and assessment of participants in the SPIN de TV3. He participated in advisory boards from Lilly and Nutricia, and he cohort, and the laboratory and neuroimaging teams for sample handling, received speakers honoraria and travel funding from Novartis and Nutricia. biomarker analyses, and structural support. Jordi Clarimón is employed by Biomedical Research Institute Sant Pau and received research grants from Generalitat de Catalunya and from Institute of Authors’ contributions Health Carlos III. Maria Carmona-Iragui is employed by Hospital de la Santa NZ, DA and AL designed the study. NZ, MS, IIG, VM, TE, IB, MA, JA, LM, OB, IS, Creu i Sant Pau. Declarations of interest: none. Juan Fortea is employed MBSS, AS, LV, JP, RB, JC, MCI, JF, AL and DA acquired data relevant for the study. by Hospital de la Santa Creu i Sant Pau and received research grants from NZ, DA and VM performed statical analysis. NZ, DA and AL contributed in Institute of Health Carlos III, Fundació La Marató de TV3, and Pla Estratègic de analysis and interpretation of data. DA and AL participated in study supervi‑ Recerca i Innovació en Salut (PERIS). Alberto LLeó is employed by Hospital de sion or coordination. NZ, DA, AL drafted the first version of the manuscript. All la Santa Creu i Sant Pau and received research grants from CIBERNED, Institute authors read and approved the final manuscript. of Health Carlos III and Fundació BBVA. He participated in advisory boards from Fujirebio-Europe, Nutricia, Biogen and received speaker honoraria from Funding Lilly. Daniel Alcolea is employed by Hospital de la Santa Creu i Sant Pau and This study was supported by the Fondo de Investigaciones Sanitario (FIS), received research grants from Pla Estratègic de Recerca i Innovació en Salut Instituto de Salud Carlos III (PI14/01126, PI17/01019 and PI20/01473 to (PERIS SLT006/17/125), and from Institute of Health Carlos III (PI18/00435 and JF, PI13/01532 and PI16/01825 to RB, PI18/00335 to MCI, PI18/00435 and INT19/00016). He participated in advisory boards from Fujirebio-Europe and INT19/00016 to DA, PI17/01896 and AC19/00103to AL) and the CIBERNED pro‑ Roche Diagnostics and received speaker honoraria from Fujirebio-Europe, gram (Program 1, Alzheimer Disease to AL), jointly funded by Fondo Europeo Roche Diagnostic, Nutricia, Esteve and from Krka Farmacéutica S.L. de Desarrollo Regional, Unión Europea, “Una manera de hacer Europa”. This work was also supported by Generalitat de Catalunya (2017-SGR-547, SLT006/17/125 to DA, SLT006/17/119 to JF, SLT002/16/408 to AL) and “Marató Zhu et al. Translational Neurodegeneration (2021) 10:50 Page 11 of 12 Author details 16. Milà-Alomà M, Salvadó G, Gispert JD, Vilor-Tejedor N, Grau-Rivera O, 1 Sant Pau Memory Unit, Department of Neurology, Institut d’Investigacions Sala-Vila A, et al. Amyloid beta, tau, synaptic, neurodegeneration, and Biomèdiques Sant Pau – Hospital de Sant Pau, Universitat Autònoma de glial biomarkers in the preclinical stage of the Alzheimer’s continuum. Barcelona, 08041 Barcelona, Spain. 2 Centro de Investigación Biomédica en Alzheimers Dement. 2020;16(10):1358–71. Red en Enfermedades Neurodegenerativas (CIBERNED), 28031 Madrid, Spain. 17. Chatterjee P, Pedrini S, Stoops E, Goozee K, Villemagne VL, Asih PR, et al. 3 Autonomous University of Barcelona, 08913 Barcelona, Spain. 4 Fundación Plasma glial fibrillary acidic protein is elevated in cognitively normal older Catalana Síndrome de Down, Centre Mèdic Down, 08029 Barcelona, Spain. adults at risk of Alzheimer’s disease. Transl Psychiatry. 2021;11(1):27. 18. Benedet AL, Milà-Alomà M, Vrillon A, Ashton NJ, Pascoal TA, Lussier F, et al. Received: 7 October 2021 Accepted: 1 December 2021 Differences between plasma and cerebrospinal fluid glial fibrillary acidic protein levels across the Alzheimer disease continuum. JAMA Neurol. 2021:e213671. 19. Cicognola C, Janelidze S, Hertze J, Zetterberg H, Blennow K, Mattsson- Carlgren N, et al. Plasma glial fibrillary acidic protein detects Alzheimer References pathology and predicts future conversion to Alzheimer dementia 1. Rohrer JD, Woollacott IO, Dick KM, Brotherhood E, Gordon E, Fellows A, in patients with mild cognitive impairment. Alzheimers Res Ther. et al. Serum neurofilament light chain protein is a measure of disease 2021;13(1):68. intensity in frontotemporal dementia. Neurology. 2016;87(13):1329–36. 20. Pereira JB, Janelidze S, Smith R, Mattsson-Carlgren N, Palmqvist S, Teunis‑ 2. Alcolea D, Vilaplana E, Suárez-Calvet M, Illán-Gala I, Blesa R, Clarimón J, sen CE, et al. Plasma GFAP is an early marker of amyloid-β but not tau et al. CSF sAPPβ, YKL-40, and neurofilament light in frontotemporal lobar pathology in Alzheimer’s disease. Brain. 2021:awab223. degeneration. Neurology. 2017;89(2):178–88. 21. Verberk IM, Laarhuis MB, van den Bosch KA, Ebenau JL, van Leeuwenstijn 3. Meeter LHH, Vijverberg EG, Del Campo M, Rozemuller AJM, Donker M, Prins ND, et al. Serum markers glial fibrillary acidic protein and neuro‑ Kaat L, de Jong FJ, et al. Clinical value of neurofilament and phospho- filament light for prognosis and monitoring in cognitively normal older tau/tau ratio in the frontotemporal dementia spectrum. Neurology. people: a prospective memory clinic-based cohort study. Lancet Healthy 2018;90(14):e1231–9. Long. 2021;179:1373–84. 4. Bridel C, van Wieringen WN, Zetterberg H, Tijms BM, Teunissen CE; and 22. Rajan KB, Aggarwal NT, McAninch EA, Weuve J, Barnes LL, Wilson RS, the NFL Group, et al. Diagnostic value of cerebrospinal fluid neurofila‑ et al. Remote blood biomarkers of longitudinal cognitive outcomes in a ment light protein in neurology: a systematic review and meta-analysis. population study. Ann Neurol. 2020;88(6):1065–76. JAMA Neurol. 2019;76(9):1035–48. 23. Heller C, Foiani MS, Moore K, Convery R, Bocchetta M, Neason M, 5. van der Ende EL, Meeter LH, Poos JM, Panman JL, Jiskoot LC, Dopper EGP, et al. Plasma glial fibrillary acidic protein is raised in progranulin- et al. Genetic Frontotemporal dementia Initiative (GENFI). Serum neuro‑ associated frontotemporal dementia. J Neurol Neurosurg Psychiatry. filament light chain in genetic frontotemporal dementia: a longitudinal, 2020;91(3):263–70. multicentre cohort study. Lancet Neurol. 2019;18(12):1103–11. 24. Marelli C, Hourregue C, Gutierrez LA, Paquet C, Menjot de Champfleur N, 6. Ashton NJ, Janelidze S, Al Khleifat A, Leuzy A, van der Ende EL, Karikari De Verbizier D, et al. Cerebrospinal fluid and plasma biomarkers do not TK, et al. A multicentre validation study of the diagnostic value of plasma differ in the presenile and late-onset behavioral variants of frontotempo‑ neurofilament light. Nat Commun. 2021;12(1):3400. ral dementia. J Alzheimers Dis. 2020;74(3):903–11. 7. Alcolea D, Martínez-Lage P, Sánchez-Juan P, Olazarán J, Antúnez C, Izagirre 25. Benussi A, Ashton NJ, Karikari TK, Gazzina S, Premi E, Benussi L, et al. A, et al. Amyloid precursor protein metabolism and inflammation markers Serum glial fibrillary acidic protein (GFAP) is a marker of disease severity in in preclinical Alzheimer disease. Neurology. 2015;85(7):626–33. frontotemporal lobar degeneration. J Alzheimers Dis. 2020;77(3):1129–41. 8. Fukuyama R, Izumoto T, Fushiki S. The cerebrospinal fluid level of glial 26. Heller C, Chan E, Foiani MS, Todd E, Russell LL, Greaves CV, et al. Plasma fibrillary acidic protein is increased in cerebrospinal fluid from Alzheimer’s glial fibrillary acidic protein and neurofilament light chain are measures disease patients and correlates with severity of dementia. Eur Neurol. of disease severity in semantic variant primary progressive aphasia. J 2001;46(1):35–8. Neurol Neurosurg Psychiatry. 2020:jnnp-2020-325085. 9. Jesse S, Steinacker P, Cepek L, von Arnim CA, Tumani H, Lehnert S, et al. 27. Katisko K, Cajanus A, Huber N, Jääskeläinen O, Kokkola T, Kärkkäinen Glial fibrillary acidic protein and protein S-100B: different concentra‑ V, et al. GFAP as a biomarker in frontotemporal dementia and primary tion pattern of glial proteins in cerebrospinal fluid of patients with psychiatric disorders: diagnostic and prognostic performance. J Neurol Alzheimer’s disease and Creutzfeldt-Jakob disease. J Alzheimers Dis. Neurosurg Psychiatry. 2021;92(12):1305–12. 2009;17(3):541–51. 28. Alcolea D, Clarimón J, Carmona-Iragui M, Illán-Gala I, Morenas-Rodríguez 10. Ishiki A, Kamada M, Kawamura Y, Terao C, Shimoda F, Tomita N, et al. Glial E, Barroeta I, et al. The Sant Pau Initiative on Neurodegeneration (SPIN) fibrillar acidic protein in the cerebrospinal fluid of Alzheimer’s disease, cohort: a data set for biomarker discovery and validation in neurodegen‑ dementia with Lewy bodies, and frontotemporal lobar degeneration. J erative disorders. Alzheimers Dement (N Y). 2019;5:597–609. Neurochem. 2016;136(2):258–61. 29. Rascovsky K, Hodges JR, Knopman D, Mendez MF, Kramer JH, Neuhaus J, 11. Abu-Rumeileh S, Steinacker P, Polischi B, Mammana A, Bartoletti-Stella et al. Sensitivity of revised diagnostic criteria for the behavioural variant of A, Oeckl P, et al. CSF biomarkers of neuroinflammation in distinct forms frontotemporal dementia. Brain. 2011;134(Pt 9):2456–77. and subtypes of neurodegenerative dementia. Alzheimers Res Ther. 30. Gorno-Tempini ML, Hillis AE, Weintraub S, Kertesz A, Mendez M, Cappa SF, 2019;12(1):2. et al. Classification of primary progressive aphasia and its variants. Neurol‑ 12. Oeckl P, Weydt P, Steinacker P, Anderl-Straub S, Nordin F, Volk AE, et al. ogy. 2011;76(11):1006–14. Different neuroinflammatory profile in amyotrophic lateral sclerosis and 31. Armstrong MJ, Litvan I, Lang AE, Bak TH, Bhatia KP, Borroni B, et al. frontotemporal dementia is linked to the clinical phase. J Neurol Neuro‑ Criteria for the diagnosis of corticobasal degeneration. Neurology. surg Psychiatry. 2019;90(1):4–10. 2013;80(5):496–503. 13. Oeckl P, Halbgebauer S, Anderl-Straub S, Steinacker P, Huss AM, Neuge‑ 32. Höglinger GU, Respondek G, Stamelou M, Kurz C, Josephs KA, Lang AE, bauer H, et al. Glial fibrillary acidic protein in serum is increased in Alzhei‑ et al. Clinical diagnosis of progressive supranuclear palsy: The movement mer’s disease and correlates with cognitive impairment. J Alzheimers Dis. disorder society criteria. Mov Disord. 2017;32(6):853–64. 2019;67(2):481–8. 33. Alcolea D, Pegueroles J, Muñoz L, Camacho V, López-Mora D, Fernández- 14. Elahi FM, Casaletto KB, La Joie R, Walters SM, Harvey D, Wolf A, et al. León A, et al. Agreement of amyloid PET and CSF biomarkers for Alzhei‑ Plasma biomarkers of astrocytic and neuronal dysfunction in early- and mer’s disease on Lumipulse. Ann Clin Transl Neurol. 2019;6(9):1815–24. late-onset Alzheimer’s disease. Alzheimers Dement. 2020;16(4):681–95. 34. Kwon HS, Koh SH. Neuroinflammation in neurodegenerative disorders: 15. Verberk IMW, Thijssen E, Koelewijn J, Mauroo K, Vanbrabant J, de Wilde the roles of microglia and astrocytes. Transl Neurodegener. 2020;9(1):42. A, et al. Combination of plasma amyloid b ­ eta(1–42/1-40) and glial fibrillary 35. Fischl B, Dale AM. Measuring the thickness of the human cerebral acidic protein strongly associates with cerebral amyloid pathology. cortex from magnetic resonance images. Proc Natl Acad Sci U S A. Alzheimers Res Ther. 2020;12(1):118. 2000;97(20):11050–5. Zhu et al. Translational Neurodegeneration (2021) 10:50 Page 12 of 12 36. Liddelow SA, Barres BA. Reactive astrocytes: production, function, and therapeutic potential. Immunity. 2017;46(6):957–67. 37. Beach TG, Walker R, McGeer EG. Patterns of gliosis in Alzheimer’s disease and aging cerebrum. Glia. 1989;2(6):420–36. 38. Delacourte A. General and dramatic glial reaction in Alzheimer brains. Neurology. 1990;40(1):33–7. 39. Ross GW, O’Callaghan JP, Sharp DS, Petrovitch H, Miller DB, Abbott RD, et al. Quantification of regional glial fibrillary acidic protein levels in Alzheimer’s disease. Acta Neurol Scand. 2003;107(5):318–23. 40. Englund E, Brun A. Frontal lobe degeneration of non-Alzheimer type. IV. White matter changes. Arch Gerontol Geriatr. 1987;6(3):235–43. 41. Arnold SE, Han LY, Clark CM, Grossman M, Trojanowski JQ. Quantitative neurohistological features of frontotemporal degeneration. Neurobiol Aging. 2000;21(6):913–9. Ready to submit your research ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations • maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions