Comparison of Serum Neurofilament Light Chain and Tau Protein Levels in Cases with Autism Spectrum Disorder and Their Healthy Siblings and Healthy Controls

Abstract

Objective

There is a growing interest among clinicians and researchers in identifying potential biomarkers associated with autism. Neurofilament light chain (NfL) and Tau protein, which are proteins associated with neurodegeneration and neuroaxonal degeneration, are particularly promising potential biomarker candidates in this field.

Methods

In this study, we compared serum NfL (sNfL) and serum Tau (sTau) levels in Autism spectrum disorder (ASD) patients, their healthy siblings (HS), and healthy controls (HC), aimed to investigate their relationship with ASD severity. Our study included 43 ASD-diagnosed participants, 43 HS participants and 42 HC participants. Clinical characteristics of the participants were assesed by Kiddie Schedule for Affective Disorders and Schizophrenia, Childhood Autism Rating Scale, Aberrant Behavior Checklist, and Strengths and Difficulties Questionnaire. Serum samples were subjected to analysis via enzyme-linked immunosorbent assay to quantitatively measure the levels of NfL and Tau protein.

Results

sNfL levels in the ASD group were significantly higher than both of the control groups. Regarding sTau levels, no significant difference was found between study and control groups. In addition, NfL and Tau levels were not significantly correlated with ASD symptom severity.

Conclusion

Our findings may indicate that the sNfl levels associated with neuroaxonal damage may constitue a potential clinical biomarker rather than being an endophenotype phenomena.

INTRODUCTION

Autism spectrum disorder (ASD) is a neurodevelopmental disorder characterized by significant deficits in social-communicative skills and restricted, repetitive behaviors and interests, typically starting in early childhood [1]. The prevalence of ASD in community has increased over the years, with the current prevalence reported to be 1−2% [2,3]. ASD is considered a multifactorial disorder resulting from the interaction of neurological, immunological, environmental, and genetic factors, although its exact cause is still unknown [4].

In recent years, there has been a growing interest in studying neurodegeneration and neuroaxonal damage in ASD. Neuroaxonal damage and loss are the pathological susbtrates of many acute and chronic neurological disorders that result in permanent disability [5]. Neurodegeneration, on the other hand, is an irreversible process leading to neuronal damage and death, commonly observed in aging and neurodegenerative diseases [6]. However, debates continue regarding the presence of neuroaxonal damage and neurodegeneration in autism. Studies in this field often rely on brain imaging and postmortem examinations, with ongoing investigations into biomarker levels in serum [7,8].

Numerous studies have reported abnormalities and alterations in various white matter regions, including the corpus callosum, cingulum bundle, and regions involving the temporal and frontal lobes in children and adults with autism [9-12]. Proton magnetic spectroscopy imaging studies by Hardan et al. [13] found significantly lower N-acetylaspartate/creatine (NAA/Cr) ratios in the frontal white matter regions of ASD patients compared to controls. This study suggested that the low NAA/Cr ratios might reflect disruption in neuroaxonal white matter tissue and shed light on potential alterations in myelin and/or axonal development in ASD.

Suzuki et al. [14] conducted a positron emission tomography imaging study, revealing increased microglial activation in the cerebellum, midbrain, pons, fusiform gyrus, anterior cingulate, and orbitofrontal cortex in individuals diagnosed with ASD compared to controls. Several studies have reported a significant reduction in the number of Purkinje cells in the cerebellum of individuals with ASD, resulting in a decrease in neurons in various brain regions compared to neurotypical indivi-duals [15-21]. Furthermore, studies have indicated fewer neurons in the amygdala of individuals diagnosed with ASD compared to controls [22]. Courchesne et al. [23] observed evidence of progressive, age-related degeneration in the brains of individuals with ASD. However, there are also authors who argue that there is no evidence of neurodegeneration in autism [24]. Some postmortem studies have shown increased oxidative stress markers in affected brain regions of individuals diagnosed with ASD, suggesting accelerated cell death under oxidative stress conditions [25-31].

Neurofilaments are specifically expressed in neurons and are major cytoskeletal proteins. Their gene expression and protein phosphorylation levels directly influence axon diameter, myelination, and conduction velocity. Neurofilaments consist of three distinct subunits based on molecular weight: neurofilament light chain (NfL), neurofilament medium chain, and neurofilament heavy chain [32-36]. The light subunit of neurofilaments, NfL, serves as a sensitive indicator of neuronal damage and neurodegeneration [37]. Levels of neurofilament proteins in cerebrospinal fluid (CSF) and/or blood increase in response to axonal damage in various neurological disorders [5]. Their levels have been found to rise in conditions such as inflammation [38], cognitive decline, and neurodegeneration [39], traumatic injuries [40], and cerebrovascular diseases [41,42], reflecting the extent of axonal damage.

Tau protein plays a crucial role in stabilizing axonal microtubules and maintaining axonal transport, particularly in neurons that locate in central nervous system [43]. When hyperphosphorylated, tau proteins tend to aggregate, which results in intraneuronal accumulation, disrupted axonal transport, synaptic miscommunication, and neurodegeneration [44,45]. In the presence of axonal damage, Tau can translocate from cells to extracellular spaces and eventually into CSF [46]. Tau levels in serum are considered to be a reliable biomarker candidate in neurological disorders, particularly including Alzheimer’s disease [47].

On the other hand, early diagnosis and intervention are crucial in ASD. Biologic research on the etiology, diagnosis, monitoring, and treatment of ASD relies, in part, on biomarker studies. Currently, there are no specific biomarkers for ASD diagnosis, and it is primarily a clinical diagnosis. Although the field of biomarker research in ASD holds promise, it is still in its early stages [48]. In recent years, biomarker studies related to NfL and Tau, particularly in neurological diseases, have gained significant attention. NfL and Tau have been included as potential biomarkers associated with neuroaxonal damage and neurodegeneration in numerous studies of neurological disorders. However, there are limited studies on NfL and Tau in relation to ASD. The aim of this study was to compare serum NfL and Tau levels between autistic patients, their unaffected siblings, and healthy controls (HC). Our hypothesis is that children with ASD may have higher levels of NfL and Tau, potential biomarkers associated with neuroaxonal damage and neurodegeneration, compared to healthy siblings (HS) and controls. We also expect that HS may have higher levels of these biomarkers compared to HC. This study is, to the best of our knowledge, the first investigation of serum NfL (sNfL) and serum Tau (sTau) in both individuals with ASD and their HS. We also think that including HS in our study will shed light on more comprehensive endophenotype studies in the future.

METHODS

Participants

This study was conducted at the Istanbul University Istanbul Medical Faculty Child and Adolescent Out-patient Clinic between October 2021 and March 2022. A total of 94 ASD patients aged between 2−12 years and their HS of similar age were invited to participate, along with 88 HC who were recruited for the study. Exclusion criteria for the ASD patients included those who declined to participate [7], those who refused to give blood samples [22], individuals with neurological diseases [5], regular use of psychiatric medications [13], and in the case of the HS group, those with neurodevelopmental diseases [4]. Children in the HC group who declined to participate [20], refused to give blood samples [18], had neurodevelopmental diseases [3], or were on regular medication [5] were also excluded from the study. In summary, a total of 43 ASD patients, 43 HS, and 42 age and gender-matched HC participated in the study.

ASD diagnosis was established based on the Diagnostic and Statistical Manual of Mental Disorders, 5th edition criteria. Each patient’s ASD diagnosis was confirmed through evaluation by two experienced child and adolescent psychiatry experts. Exclusion criteria for the case group included the presence of schizophrenia, bipolar disorders, metabolic, genetic, neurologic, or gastrointestinal disorders, regular use of medications for chronic illnesses and any other psychiatric disorders, intake of supplements such as vitamins and fish oil during the last month, active presence of infection or history of infection in the last month, and the presence of comorbid neurodevelopmental disorders except for attention deficit hyperactivity disorder (ADHD) and intellectual disabilities. For both control groups, in addition to the above exclusion criteria, individuals with ADHD and intellectual disabilities were also excluded from the study.

The study was approved by the Istanbul Faculty of Medicine Clinical Research Ethics Committee on December 20, 2022 with the number code 2021/2086. All participants and their parents were thoroughly informed about the study, and written consent was obtained from the parents.

Diagnosis and Symptom Assessment

Psychiatric diagnoses for all participants were assessed using the Kiddie Schedule for Affective Disorders and Schizophrenia Present and Lifetime Version. Participants were provided with a questionnaire consisting of questions related to sociodemographic and clinical information. To assess the severity of ASD symptoms in the case group, the Childhood Autism Rating Scale (CARS) was used, and for evaluating attention deficit, hyperactivity, and behavioral problems accompanying autism, the Strengths and Difficulties Questionnaire (SDQ) and Aberrant Behavior Checklist (ABC) were employed. The reliability and validity of CARS, SDQ, and ABC have been established for the Turkish population [49-51].

Collection of Samples and Measurement of NfL and Tau

Before the blood collection procedure, families were informed about the blood sampling process, and participants and their families were treated kindly. Venous blood samples were collected from participants after fasting for 8−12 hours and were aliquoted into gel tubes, left to clot, and then centrifuged at 3000 × g for 15 minutes. The obtained serum samples were stored at −80°C until analyzed. NfL levels were measured using ELISA kits (MyBioSource, Inc.; catalog number: MBS8802206), and Tau levels were measured using Invitrogen ProQuantum Immunoassay (catalog number: A46738) according to the manufacturers’ instructions.

Statistical Analysis

The data collected from the study were analyzed using the SPSS 26.0 (IBM Co.) software package. Descriptive statistics such as percentages, arithmetic means, and standard deviations were used. Relationships between qualitative data were evaluated using the Pearson chi-squared (χ²) test. The normality of the distribution of continuous variables was assessed using the Kolmogorov-Smirnov test. Non-parametric Kruskal-Wallis test and Spearman correlation analysis were used for data that did not show a normal distribution. For normally distributed variables, one-way analysis of variance was used for comparing three independent groups, and Pearson correlation analysis was used for correlation analysis. Age, sex, and body mass index (BMI) percentile parameters, thought to affect biochemical parameters, were used as covariates. Logarithmic transformation was applied since serum NfL and Tau levels did not follow a normal distribution. The serum log-NfL and log-Tau levels of the case and control groups were compared using the multivariate analysis of covariance (MANCOVA) test. Bonferroni tests with multiple test corrections were conducted for post-hoc tests. The statistical significance level was set at p < 0.05.

RESULTS

The study sample consisted of a total of 128 cases, including 43 children with ASD, HS of children with ASD, and 42 healthy children without ASD. The case group consisted of 8 girls and 35 boys, HS consisted of 19 girls and 24 boys, and the HC consisted of 7 girls and 35 boys. The mean age of the case group was 72.20 ± 28.60 months, the mean age of the HS was 84.83 ± 39.32 months, and the mean age of HC was 73.85 ± 27.67 months. In terms of average age, the case group and HC were similar (p = 0.149).The BMI was 16.20 ± 3.00 on average in the case group, 16.88 ± 3.21 in the HS, and 16.71 ± 2.67 on average in HC. All three groups were statistically similar in terms of BMI (p = 0.275). Sociodemo-graphic and clinical data of ASD, HS and HC group are given in Table 1. There were no significant difference between the groups in terms of age, sex and BMI (Table 1).

Table 1.

Sociodemographic data and scale scores

Variable	Case group (n = 43)	Healthy siblings (n = 43)	Healthy controls (n = 42)	pvalue
Sex (girl/boy)	8/35	19/24	7/35
Age (mo)	72.20 ± 28.60	84.83 ± 39.32	73.85 ± 27.67	0.149a
BMI (kg/m2)	16.20 ± 3.00	16.88 ± 3.21	16.71 ± 2.67	0.275b
Maternal age (yr)	35.77 ± 5.14	35.77 ± 5.14	36.60 ± 6.16	0.885b
Paternal age (yr)	39.07 ± 4.37	39.07 ± 4.37	40.74 ± 6.73	0.277a
Special education (yes)	29 (67.4)	1 (2.3)	0 (0)	< 0.001c
Birth complication (yes)	14 (32.6)	6 (13.9)	4 (9.5)	0.015c
Incubator story (yes)	16 (37.2)	8 (18.6)	6 (14.3)	0.029c
CARS total points	38.20 ± 8.03	-	-
ABC total score	59.88 ± 23.37	-	-
SDQ total score	17.74 ± 4.44	-	-

Values are presented as mean ± standard deviation or number (%).

BMI, body mass index; CARS, Childhood Autism Rating Scale; ABC, Aberrant Behavior Checklist; SDQ, Strengths and Difficulties Questionnaire; -, not available.

^aAnalysis of variance, ^bKruskal-Wallis test, ^cPearson chi-square test.

The mean sNfL levels in the case group were 80.5 ± 128.8; in the HS group the levels were 31.4 ± 25.9 and in HC group the levels were 28.4 ± 19.9, respectively. MANCOVA method and post-hoc pairwise comparisons with the Bonferroni correction showed that sNfL levels were sigificantly higher in the case group compared to HS and HC (p = 0.031 and p = 0.008, respectivly). No significant difference was found between HS and HC (p = not significant [NS]). Results are shown in Table 2.

Table 2.

Comparison of neurofilament light chain and Tau protein levels between case group and healthy groups

Variable	Case group (n = 43)	Healthy siblings (n = 43)	Healthy controls (n = 42)	MANCOVAa				Post-hoc comparisonsa,b
				F	pvalue	ηp2		I vs. II	I vs. III	II vs. III
NfL (pg/ml)	80.5 ± 128.8	31.4 ± 25.9	28.4 ± 19.9	5.564	0.005	0.084		0.031	0.008	NS
Tau (pg/ml)	145.5 ± 599.5	23.7 ± 105.7	91.1 ± 436.1	1.709	0.185	0.027		0.216	NS	0.637

Values are presented as mean ± standard deviation.

Covariants: age, sex and body mass index.

NfL, neurofilament light chain; MANCOVA, multivariate analysis of covariance; NS, not significant.

^aLogarithmic transformation was performed before MANCOVA. ^a,bPost-hoc comparisons Bonferroni.

In the sTau case group, the mean was calculated as 145.5 ± 599.5; in HS 23.7 ± 105.7 and in HC. 91.1 ± 436.1 respectively. When sTau levels in the case group were compared with HS and HC, no significant difference was found between the groups (p = 0.216 and p = NS, respectively). In addition, there was no significant difference in sTau between HS and HC (p = 0.637) (Table 2).

No significant correlations were found between sNfL and sTau levels according to the CARS, SDQ and ABC scores (NfL: p = 0.465, p = 0.588, p = 0.227; Tau: p = 0.167, p = 0.445, p = 0.337, respectively) (Table 3).

Table 3.

Correlation analysis of NfL and Tau with CARS total score, ABC total score, SDQ total score

	Correlation coefficient	pvaluea
NfL–CARS total score	0.134	0.465
NfL–ABC total score	0.100	0.588
NfL–SDQ total score	0.202	0.227
Tau–CARS total score	−0.250	0.167
Tau–ABC total score	0.140	0.445
Tau–SDQ total score	0.175	0.337

NfL, neurofilament light chain; CARS, Childhood Autism Rating Scale; ABC, Aberrant Behavior Checklist; SDQ, Strengths and Difficulties Questionnaire.

^aSpearman correlation analysis.

DISCUSSION

In our study, the levels of sNfL and sTau in cases diagnosed with ASD were compared with HS of ASD cases and HC, and the relationship of these biomarker candidates with the severity of ASD and additional clinical symptoms was evaluated.

Furthermore, in the literature, there have been studies examining the relationship of these two biomarker candidates in neurodegenerative diseases and psychiatric disorders. Some studies have examined the relationship between ASD and NfL and Tau, but to our knowledge, no study has yet been found that includes HS of ASD cases.

In the results of our study, NfL serum levels were found to be significantly higher in the case group compared to both control groups (p = 0.031, p = 0.008), while there was no significant difference in Tau serum levels (p = 0.216, p = NS), respectively.

A study conducted by He et al. [52] included 166 Chinese children, 83 individuals with autism and 83 typically developing children. In the study by He et al. [52], the mean age and sex distribution of the case group with ASD showed similar characteristics with the case group in our study. In this study, mean sNfL concentrations were found to be significantly higher in ASD than in the control group. Furthermore, increases in serum NfL levels were found to be correlated with both the increase in the risk of ASD and the increase in the severity of ASD in this study. The data from this study is not compatible with our study in terms of the correlation between the severity of ASD and sNfL levels. However, they are consistent in terms of the correlation between sNfL levels among the two control groups and the ASD case group. The size of the sample selection in the two studies, the ethnic group and the differences between the individuals who applied the scales, the clinical diagnosis of ASD may explain the different correlations between ASD severity and NfL. In a study conducted by Simone et al. [53], sNfL and glial fibrillary acidic protein levels were found to be significantly higher in children with ASD compared to HC. In this study, it was also noted that these biomarkers could shed light on both early diagnosis and etiological aspects in ASD [53].

NfL protein has been suggested as a diagnostic, monitoring, and prognostic biomarker for various conditions affecting the nervous system, including multiple sclerosis (MS) [54], amyotrophic lateral sclerosis [55], Huntington’s disease [56], Alzheimer’s disease [57], Parkinson’s disease [58], cognitive decline [39], spinal cord injury [59], and ischemic stroke [41]. These studies have noted the involvement of neuroaxonal damage and neurodegenerative processes in the etiopathogenesis of these conditions, and serum or CSF NfL levels were found to be significantly elevated in these patient groups. Given that our study found significantly elevated sNfL levels in ASD cases, we believe that neuroaxonal damage and neurodegeneration may be involved in the etiopathogenesis of ASD.

Some diffusion tensor imaging (DTI) studies in ASD cases have shown differences in functional anisotropy (FA) trajectories and involvements in the white matter and white matter abnormalities have been confirmed in DTI tractography [9-12,60]. Similarly, in studies with MS patients, diffusion measurements have demonstrated microstructural pathways involved in white matter pathology and potential predictive FA values at the onset of the disease and increased radial diffusion values with the progression of MS, primarily considered as an indicator of underlying white matter pathology independent of local lesion load [61]. Considering studies on serum and CSF NfL levels in MS patients, simultaneous NfL studies that will be conducted in ASD cases with imaging studies may be important for diagnosis and follow-up and can shed light on the presence of axonal damage. In addition, in a study by Hardan et al. [13], it was suggested that low NAA/Cr ratios might reflect damage to the neuroaxonal white matter tissue and indicated a change in myelin and/or axonal development in ASD, shedding light on the neurobiological basis of white matter abnormalities in ASD. Considering that our study found higher serum NfL levels, we believe that white matter damage may be involved in the etiology of ASD.

Data on sNfL levels in the pediatric group have been increasing in recent years. There are studies investigating NfL levels in spinal muscular atrophy, pediatric MS, febrile and epileptic seizures, childhood cerebral adrenoleukodystrophy, and ASD groups [52,62-66]. In these studies, parameters such as NfL levels in control groups and healthy children, disease severity and activity, disease prognosis, and response to treatment have been examined. In a study by Reinert et al. [63], it was emphasized that NfL levels decrease with age, but in our study, no significant correlation was found between age and NfL serum levels. From this perspective, longitudinal follow-up studies on ASD may be needed. Considering the neurodevelopmental heterogeneity in fetal ASD, degradation products in early ages may not be predictive of biomarkers in later ages. Although serum levels were examined in these studies, as in our study, there are also data on epilepsy and encephalopathy patient groups regarding NfL/albumin and Tau/albumin ratios in CSF in the pediatric population [67]. In future research, our data on serum levels may contribute to the examination of CSF and serum correlations with these two biomarker candidates in studies investigating NfL and Tau CSF measurements or serum/CSF albumin ratios.

In a study by Kadak et al. [68], mean sTau concentrations were found to be significantly higher in the control group compared to the ASD case group. Their preliminary study suggests that low levels of serum α-synuclein and tau may be implicated in the relationship between synaptic activity and autism, a-synuclein and tau aggregation may lead neuronal or synaptic dysfunction or neurodegeneration. In a study by Ayaydin et al. [69], sTau concentrations were found to be significantly higher in the ASD case group compared to the control group in which they spaculate that risk gene proteins associated with ASD are found in synaptic mechanisms and different tau levels have been reported in patients with ASD and controls, tau may also be an important factor in the pathogenesis of ASD. In our study, similar to the results of Ayaydin et al. [69], sTau levels were found to be higher in the ASD group compared to control groups but no statistically significant difference was detected. Kadak et al. [68] noted that low sTau levels in the case group may be associated with damaged transport mechanisms from the CSF to serum in the cases in the sample. According to these findings, subgroups of ASD patients. Measurement values, units of measurement, sensitivity of the kits used, laboratory analyses, the limited number of participants, and the homogeneous distribution of the data in this context may partially explain the discrepancies in these data. We also suggest that these data may be reviewed in the light of subgroup studies involving genetic syndromes with synaptic disorders in autism and may shed light on new studies. A study by Tai et al. [70]. demonstrated that genetic reduction of the Tau protein prevented behavioral symptoms of autism in two mouse models simulating different causes, supporting the hypothesis that Tau protein may facilitate the development of some ASD cases by reducing phosphatase and tensin homolog (PTEN) activity through the PI3K/Akt/mTOR pathway. Furthermore, this study suggested an activating role for Tau protein in the pathogenesis of autism and described reducing Tau protein as a potential therapeutic strategy for some disorders causing this condition. Although our study showed abnormal distributions in Tau levels, the higher serum levels in the ASD group suggest that Tau may play a role in some cases. In homogeneous ASD subgroups showing syndromic characteristics, an increase in serum or CSF studies may contribute to etiopathogenetic processes. Additionally, we believe that a multidisciplinary approach in cases of ASD with known or suspected similar mutations or existing genetic syndromes will contribute to treatment strategies in subgroups and shed light on the relationship between ASD and Tau. Ultimately, both genomic, proteomic and imaging studies are needed for the Tau and ASD relationship.

There are limited studies on NfL and Tau regarding psychiatric disorders. Some studies suggest a relationship between psychotic symptoms and Tau pathology but the results have been contradictory [71-73]. In a study of adolescents with early-onset psychosis, it was found that these patients had lower plasma Tau concentrations than HC and that Tau levels in psychotic patients were associated with the microtubule-associated Tau gene and brain measurements [74]. However, our study did not include patients with psychotic symptoms or a control group, so further studies on cases of psychosis in children and adolescents are important. In addition, in a study [75], sTau levels were found to be higher in patients diagnosed with ADHD compared to the control group, but our study did not find a significant relationship between NfL and Tau levels and comorbid ADHD in ASD cases. Future studies are needed to understand the role of NfL and Tau in children with ADHD who do not have a diagnosis of ASD. Some studies on mood disorders also indicate increased studies, and there is some data associated with neurodegeneration and neuroaxonal damage. In addition, NfL and Tau studies have been conducted with participants diagnosed with bipolar disorder and major depressive disorder and these studies can guide research in cases of comorbid mood disorders with ASD based on our data [76-81].

In our study, the lack of a significant correlation between HC and HS raises the idea that larger studies are needed to use NfL and Tau as an endophenotypic feature.

Our study has both strengths and limitations. To our knowledge, the relationship between NfL and Tau levels and their connection to ASD has been investigated in the literature, but there has been no study that includes siblings of individuals with ASD. This aspect of our study aims to gather data on individuals with genetic predisposition, evaluating NfL and Tau as potential biomarkers and endophenotypes. Additionally, the limited availability of data on NfL and Tau serum levels in the pediatric population suggests that our study can shed light on future research in this area. The scarcity of biomarker studies related to NfL and Tau in neurodevelopmental and psychiatric disorders is a strong point of our study. Furthermore, we applied exclusion criteria for conditions that could affect NfL and Tau levels, and individuals using medication were not included in the study. The inclusion of scales for assessing the severity of the condition and the presence of semi-structured interviews can also be considered strengths. On the other hand, there are limitations to our study. One limitation is that, while increased levels of NfL were observed in the cases with ASD, the specifity of NfL to the cases with ASD, as a broader marker of neurodevelopmental or neurodegenerative conditions, requires further clarification. Also our study was conducted with small sample size and may not fully capture the spectrum of variablity within the ASD population. This limitation constrains the generalizability of our findings and warrants caution when extrapolating the results to the larger ASD population. Another limitation is that we did not analyze NfL and Tau levels in CSF. Post-mortem sample analysis and neuroimaging techniques, which could help recognize neurodegenerative processes, were not part of our study. The presence of other comorbidities in some cases with ASD can be considered another limitation of the study, since it may affect NfL and Tau levels. Moreover, the wide age range of participants introduce another variability that may influence biomarker levels independent of ASD pathology. The absence of age stratification could mask age-specific patterns in biomarker expression, suggesting that future studies could benefit from a stratified design. Finally, this study is cross-sectional therefore, unable to capture changes over time so there is a need for longitudinal studies that track changes in NfL and tau protein as biological markers of ASD over time.

In conclusion, considering the data of our study, it suggests that NfL has a role in the etiopathogenesis of ASD. Additionally, NfL can be used to monitor and manage symptoms when it is an important indicator of autism severity. For this, more studies with larger samples are needed. In addition, longitudinal studies on patient groups using medication and receiving special training may be a candidate biomarker to be used in prognosis, just like in some neurological studies. In cases where it is a useful tool for early diagnosis, interventions can be started early and known treatment methods can be helpful. In ASD, there are only a few studies on biomarkers for diagnosis, and no studies on follow-up and treatment have been found about NfL. Moreover, the increases in the articles, especially those accompanied by imaging studies, may accompany each other in terms of the role of neurodegeneration and neuroaxonal damage in the role of NfL in ASD. In addition, we speculate that more studies should be done in large samples to investigate the role of Tau in some ASD cases.

Funding Statement

Funding This research was financially supported by the Scientific Research Projects Coordination Unit of Istanbul University (code: TTU-2022-38387).