Traumatic brain injury (TBI) affects over one million Americans each year and may lead to long-term debilitating behavioral symptoms particularly in groups with greater health disparities (Center for Disease Control and Prevention, National Center for Injury Prevention and Control CDC, 2021). Sex differences in TBI behavioral outcomes have been suggested, but possible molecular underpinnings are unclear. Current TBI treatment guidelines remain largely based on the studies with predominantly male participants (Biegon 2021). Exploring TBI outcomes and potential candidate TBI biomarkers based on sex differences might lead to development of precision medicine therapies.

In a recent review, studies with mild to moderate TBI were more likely to use social-behavioral measures and women reported worse behavioral outcomes than men; however, for severe TBI, studies were more likely to use survival and community integration as outcome measures and women fared better than men (Gupte et al., 2019). In a review of sex differences in mild TBI, researchers found that females have a proportionally higher incidence of concussions in sports; however, comparisons of symptom prevalence between the sexes have had mixed results (Merritt et al., 2019). Other reviews highlight the paucity of studies evaluating sex differences in blood-based biomarkers in TBI (Strathmann et al., 2014; Gupte et al., 2019).

Commonly measured biomarkers of TBI include total tau, neurofilament light chain (NfL), glial fibrillary acidic protein (GFAP), and ubiquitin C-terminal hydrolase-L1 (UCH-L1) (Kawata et al., 2016; Shahim et al., 2020a). In brain tissue of rats, researchers reported that both sexes had upregulation in gene expressions of GFAP, myelin basic protein, and tau following TBI, and while higher levels of GFAP was observed in female rats, in male rats it was significantly decreased (Wright et al., 2017). No sex differences in the biomarkers were noted in separate murine study post severe closed head injury (Rubenstein et al., 2017a). A recent study that examined 20 female and 20 male athletes found that plasma tau levels were significantly higher in females 6 h and 3 days after a concussive event than in males (Mondello et al., 2020). Accumulating evidence suggests that, not only may sex differences exist, but an exploration of outcomes and associations with biomarkers are paramount to furthering the current state of the science.

The aims of this project are to examine sex differences in postconcussive, depressive, and post-traumatic stress disorder (PTSD) symptoms, levels of circulating GFAP, tau, NfL, and the relationship between symptoms and biomarkers as well as health-related quality of life (HRQOL).

To our knowledge, this study is the first to examine levels of circulating biomarkers and symptom scores in the context of sex differences in chronic TBI. In this cross-sectional convenience sample analysis, we found that levels of circulating plasma GFAP and tau were significantly higher in the female group compared to the male group. NfL levels did not differ significantly between groups. Behavioral symptoms of depression, NSI total, NSI-vestibular, NSI-somatosensory, and NSI- affective scores were higher in the female group as well. However, HRQOL of role limitations due to emotional problems, vitality, emotional well-being, social functioning, and pain were higher in the male group indicating better HRQOL. In addition, bivariate correlation results showed positive associations between symptoms, HRQOL scores and levels of GFAP, NfL, and tau.

We observed that levels of GFAP and tau were significantly higher in females. GFAP is an intermediate filament protein found in astrocytes of central nervous system and a marker of reactive astrocytes seen in neurodegenerative disease (Hol and Pekny 2015) and in TBI (Huang et al., 2015; McMahon et al., 2015). Reactive astrocytes can initiate pro-and anti-inflammatory immune mediators, communicate with innate and adaptive immune systems as well as neurons, glia, and vascular cells, and regulate blood-brain barrier in response to TBI (Burda et al., 2016). A recent study suggested that GFAP is a promising biomarker in distinguishing patients with acute mTBI from controls especially when combined with tau and NfL (Gill et al., 2018a). Another study reported that GFAP combined with UCHL-1 has excellent negative predictive value and sensitivity in predicting acute intracranial injuries detected by head computed tomography scans (Bazarian et al., 2018). In subacute and chronic TBI, serum GFAP levels have been shown to be elevated; however in contrast to acute mTBI, these levels were not sensitive in determining injury severity or outcomes (Shahim et al., 2020b). Few clinical studies have examined circulating biomarkers following TBI in relation to sex differences; however, several preclinical investigations have identified sexual dimorphic outcomes. Similar to our study, in mTBI, brain expression of GFAP was found to be elevated in female rats while being significantly lower in male rats (Wright et al., 2017). However, there were no sex differences in repetitive mTBI (Wright et al., 2017) or severe TBI in preclinical models (Rubenstein et al., 2017b).

In addition to GFAP, we also observed sex differences in tau following TBI. Tau is a microtubule-associated protein located on neuronal axons and its dysmetabolism plays a role in a number of neurodegenerative conditions such as Alzheimer’s disease (AD), corticobasal degeneration, as well as TBI (Weingarten et al., 1975; Olivera et al., 2015; Gill et al., 2018b; Didonna 2020). Plasma phosphorylated tau (p-tau) and p-tau-t-tau ratios have been demonstrated to distinguish patients with acute and chronic TBI from healthy controls (Rubenstein et al., 2017b). In chronic mTBI, exosomal tau and p-tau have been shown to be elevated in repetitive mTBI (Kenney et al., 2018). Our findings show that tau is significantly elevated in the female group compared to the male group after controlling for significant covariates. This is similar to findings in sports related concussions, where plasma tau was significantly higher in female athletes up to 3 days after initial injury (Mondello et al., 2020). Early tau depositions in the brain were higher in women measured by positron emission tomography in AD trajectory (Buckley et al., 2019). Higher levels of p-tau were also observed in women versus men among apolipoprotein E e4 (APOE4) carriers (Sundermann et al., 2020). Similarly, in a separate cohort, women had higher tau and p-tau concentrations in early stages of mild cognitive impairment (MCI) and subjective cognitive decline, while among non APOE e4 carriers, women had higher concentrations at the later stages of AD, dementia, and MCI (BabapourMofrad et al., 2020).

While we observed significant changes in GFAP and tau, NfL was not statistically significant between sexes in this report. NfL is 68 kDa subunit of the neurofilaments located on the neuronal cytoplasm which is released in response to CNS axonal damage due to neuroinflammation, neurodegeneration, and/or traumatic or vascular injury (Zetterberg 2016; Gaetani et al., 2019). Serum NfL has been shown to distinguish patients with mTBI, moderate or severe TBI for months and even years after injury (Shahim et al., 2020b). Cerebrospinal fluid NfL and CSF/serum albumin ratio were found to be significantly higher in male aging cohorts of AD (Skillbäck et al., 2021). These studies indicate there are sex differences in the levels of circulating biomarkers and additional studies are needed to corroborate these results.

Another interesting finding is that the female group had higher neurobehavioral and depressive symptom scores and significantly lower scores of HRQOL in role limitations due to emotional problems, vitality, emotional well-being, social functioning, and pain. We did not observe these trends in the male group. These results are consistent with previous reports of women having higher depressive symptoms and stress after mid-to moderate TBI (Bay et al., 2009) and higher likelihood of poor postconcussive symptom scores in mTBI (Bazarian et al., 2010). Our bivariate correlation analysis also showed that tau was positively associated with NSI total, NSI-cognitive, and marginally with NSI-somatosensory, while being negatively correlated with physical and social functioning of HRQOL. NfL also negatively correlated with physical functioning in the female group, indicating that as the concentrations of both tau and NfL increased the score for physical HRQOL functioning decreased. In our previous studies, we found that tau concentrations correlate significantly with postconcussive, post-traumatic, and depressive symptoms (Kenney et al., 2018; Pattinson et al., 2020) and exosomal NfL positively correlated with chronic postconcussive symptoms, PTSD, and depressive symptoms in predominantly male TBI groups (Guedes et al., 2020). In the current study, we also observed higher concentrations of GFAP to be associated with better emotional well-being HRQOL in the female group. While this result reached statistical significance, it is contradictory to the hypothesis that higher GFAP would be associated with lower HRQOL.

A possible mechanism underlying sex differences in TBI might include hormone levels. Several studies implicate the role of hormonal fluctuations during menstrual cycle in symptomatic presentations between sexes (Wunderle et al., 2014; Brown et al., 2015; Sundermann et al., 2020). For example, a study looking at the menstrual cycle phase and levels of progesterone concluded that women during the luteal phase of menstruation have significantly lower QOL metrics compared to women injured during the follicular phase (Wunderle et al., 2014). Lower testosterone levels resulted in higher p-tau concentrations in female APOE4 carriers, suggesting higher testosterone may be inversely related to p-tau levels in AD disease (Sundermann et al., 2020). Although we controlled for age (median age was 42 for women and 38 for men) in our analysis, hormonal fluctuations related to menopause have been documented to affect TBI outcomes. In large retrospective studies, peri- and post-menopausal women (age >50 years of age) had better outcomes and survival rate than men (Davis et al., 2006; Berry et al., 2009).

Another important factor to be considered is the role of gender as a fluid non-binary variable that may impact TBI experiences (Giordano et al., 2020). Societal gender expectations and individual vulnerabilities may affect symptoms and outcomes, coping skills, resources, and ultimately reintegration to society after TBI (Mollayeva et al., 2018). Currently, evidence is lacking on the effects of sex and gender in TBI research (Mollayeva et al., 2018) and often transgender groups are too small to be included in analysis (Giordano et al., 2020). Future clinical studies need to incorporate objective measures to further understand the impact of gender within sex on TBI outcomes.

This study has several limitations. First, our female participants in particular are of limited number. Future studies with larger sample sizes are warranted to corroborate our findings. Second, the injury mechanism of TBI (i.e., blast vs. non-blast TBI) may show a different effect on GFAP and total tau concentration. Future studies should investigate these biomarkers in relation to each injury type. Third, a cross-sectional design with self-reported TBI and symptoms makes it challenging to infer a causal relationship between the biomarkers and behavioral outcomes. Fourth, we did not measure hormonal differences or evaluated menstrual cycle phases. These may be important factors to include in future explorations on sex differences in TBI.

Lastly, the significance of the findings observed in GFAP and Tau between female and male participants in this study are affected by the lack of current clinical reference values for these biomarkers. Though GFAP and UCHL-1 have been shown to have high sensitivity for detection of acute TBI (Bazarian et al., 2018), the establishment of reference values in sex differences in TBI is still novel and clinical reference values are needed in the field. Reference ranges have been established for other biomarkers such as S100 calcium binding protein B (S100B) and neuron-specific enolase (NSE) and have been found to have sex-dependent differences. Specifically, higher s100B was observed in females and higher NSE in males (Hajduková et al., 2015). Several studies published blood-based reference ranges for GFAP and NfL; however, these ranges appear to differ between the studies (van Geel et al., 2002; Mayer et al., 2013). The utility of biomarkers, though promising in diagnostic and prognostic applications is limited by the lack of sex-based reference values. Additional studies are warranted to compare or refute clinical significance.

Despite these study limitations, this analysis shows robust sex differences in the levels of biomarkers and symptom burden in TBI after controlling for significant covariates (age, race, military status, BMI, and education). Our findings indicate that elevated biomarkers may be related to poor behavioral TBI outcomes and may impact overall HRQOL. Future explorations stratified by sex, gender, and age are paramount in delineating individual factors and identifying potential target biomarkers in treatment guidelines.