Cognitive, biomarker, and neuroimaging indices associated with traumatic encephalopathy syndrome across two independent athlete cohorts

Cognitive, biomarker, and neuroimaging indices associated with traumatic encephalopathy syndrome across two independent athlete cohorts | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article Cognitive, biomarker, and neuroimaging indices associated with traumatic encephalopathy syndrome across two independent athlete cohorts Brooke Conway Kleven, Lung-Chang Chien, Dana Surwill, Michael L. Alosco, and 16 more This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-9385305/v1 This work is licensed under a CC BY 4.0 License Status: Under Review Version 1 posted 8 You are reading this latest preprint version Abstract Background : Traumatic encephalopathy syndrome (TES) is a clinical research construct used to identify individuals at risk for chronic traumatic encephalopathy (CTE) following exposure to repetitive head impacts (RHI). Adjudication of TES relies on clinical features such as progressive cognitive impairment and neurobehavioral dysregulation. Blood-based biomarkers and structural neuroimaging abnormalities have been associated with TES but are not part of the criteria. This study evaluated whether TES identification was associated with the combined contribution of cognitive performance, blood biomarkers, and structural neuroimaging measures across two well-characterized cohorts. Methods : Participants included 158 professional fighters from the Professional Athletes Brain Health Study and 149 former American football players from The DIAGNOSE CTE Research Project. Three indices were constructed representing complementary domains: a cognitive index reflecting cohort-specific cognitive features, a blood biomarker index including plasma neurofilament light chain, glial fibrillary acidic protein, total tau, tau phosphorylated at amino acid 231, and APOE -ε4 carrier status, and an imaging index comprising volumetric MRI measures of subcortical structures, ventricles, and corpus callosum subregions. Grouped weighted quantile sum regression models were estimated within each cohort to evaluate associations between these indices and TES while adjusting for age, race, competition status, and RHI exposure. Results : Multidomain models demonstrated improved model performance compared with single-domain models in both cohorts (PABHS: AUC=0.91, PPV=0.80; DIAGNOSE CTE: AUC=0.84, PPV=0.85). Biomarker and imaging indices contributed additional information across cohorts, although imaging contributions were more prominent in fighters whereas blood biomarker associations were stronger in football players. Conclusion : TES in RHI-exposed athletes was associated with a convergent clinicobiological profile observed across two independent cohorts with distinct exposure patterns. These findings support multidomain analytic frameworks for evaluating correlated biological signals in RHI-exposed populations and may inform future studies of TES and CTE. Chronic traumatic encephalopathy Traumatic encephalopathy syndrome repetitive head impacts blood biomarkers neuroimaging neurodegeneration Figures Figure 1 Introduction The long-term effects of repetitive head impacts (RHI) raise concerns regarding contact sports, including the risk of developing chronic traumatic encephalopathy (CTE) 1 , a neurodegenerative disorder associated with RHI exposure. While the standard for CTE diagnosis is a post-mortem neuropathologic confirmation of hyperphosphorylated tau (pTau) deposits at the depths of cortical sulci and around small blood vessels, the National Institute of Neurological Disorders and Stroke (NINDS) developed consensus diagnostic criteria for traumatic encephalopathy syndrome (TES) to assist in research involving living individuals with a history of RHI and potential CTE. 2 The criteria for TES adjudication follows a stepwise process requiring an individual to have substantial exposure to RHI, core clinical features of neurobehavioral dysregulation and/or cognitive impairment, a progressive course of symptoms, and no other medical condition(s) that can fully account for the symptoms. 2 Research evaluating the validity of TES has explored contributions of various biomarkers that may support the earlier diagnosis of CTE or serve as a method to track the progression of neurodegenerative processes that may occur following RHI exposure. Blood biomarkers such as neurofilament light chain protein (NfL), which serves as a marker of axonal injury, and glial fibrillary acidic protein (GFAP), indicative of activated astroglia, have been associated with a diagnosis of TES. 3 – 5 Magnetic resonance imaging (MRI) indicates that TES is frequently associated with subcortical and medial temporal lobe atrophy, with volumetric loss observed in regions such as the thalamus, hippocampus, and corpus callosum. 6 – 8 Biomarker studies typically evaluate markers individually rather than integrating multiple biological domains into a composite framework. 3 , 6 , 15 – 18 The purpose of the current study was to determine whether TES identification is characterized by the combined contribution of cognitive performance, blood biomarkers, and structural neuroimaging measures across two independent cohorts. Because cognitive impairment is currently a core feature of the TES diagnostic criteria, the cognitive domain in the present study represents the clinical component of the syndrome. We therefore hypothesized that individuals identified as TES+ would demonstrate a convergent pattern of cognitive dysfunction, biological abnormalities, and neuroanatomical changes, and that biological domains would contribute to TES identification beyond the clinical domain alone. To evaluate this hypothesis, we applied an integrative modeling framework to quantify how individual and combined domains align with TES identification. 19 We leveraged data from two independent athlete cohorts, the Professional Athletes Brain Health Study (PABHS; consisting primarily of boxers and mixed martial arts fighters) 20 and the Diagnostics, Imaging, and Genetics Network for the Objective Study and Evaluation of Chronic Traumatic Encephalopathy (DIAGNOSE CTE; consisting of former professional and college football players) 21 to determine whether multimodal associations with TES are reproducible across athlete populations with distinct patterns of RHI exposure and demographic characteristics. Methods Details of the PABHS cohort have been previously published. 20 The PABHS is a longitudinal cohort of over 900 professional athletes with a history of RHI exposure and age/sex-matched unexposed controls. To meet the cohort’s criteria of a retired fighter, athletes must not have plans for future fights and cannot have competed in a sanctioned fight within the past two years. Active fighters must have competed in a professionally sanctioned event within the past two years and must wait at least 45 days following a fight to attend baseline and/or annual visits. For any participant with multiple visits, only the data from the most recent visit were included in the current analyses. RHI exposure was measured using the number of professional fights completed. Consistent with prior analyses from PABHS, a minimum threshold of 10 professional fights was used to operationalize substantial exposure to RHI within this cohort. This threshold was selected to represent sustained professional-level combat exposure and to exclude individuals with limited professional experience whose fight histories may not reflect consistent RHI exposure. In combat sports, the number of professional fights provides a more direct proxy for cumulative head impacts than years of participation alone, as bout frequency, rounds per fight, and intensity of exposure vary substantially across athletes. Methods for the DIAGNOSE CTE have been detailed previously. 22 Briefly, this eight-year longitudinal study enrolled 240 men ages 45–74, including 120 former professional football players (PRO), 60 former college football players (COL), and 60 unexposed asymptomatic men. The asymptomatic men were not included in this study. The PRO participants were required to have played 12 or more years of organized football, including at least three years in college and more than three years in the National Football League (NFL). COL participants were required to have played six or more years of American football, with at least three years at the college level. 13 , 23 TES Adjudication TES status was determined independently for each cohort through multidisciplinary diagnostic consensus conferences using the NINDS Consensus Diagnostic Criteria for TES. 2 To be identified as TES+ required confirmation of substantial RHI exposure, evaluation of core clinical features including cognitive impairment and/or neurobehavioral dysregulation, evidence of progressive symptom worsening, and determination that symptoms were not fully explained by other neurological or psychiatric conditions. “Substantial exposure” reflects a history of RHI from activities such as contact sports or military service, typically involving prolonged participation or roles associated with frequent head impacts. 2 All participants included in the present analyses (TES + and TES-) met cohort-specific predefined thresholds for substantial exposure as outlined above. As such, RHI exposure did not distinguish TES+ from TES- within the analytic sample and was not used to define case status in the present analyses. Cognitive impairment was uniformly defined based on reported decline from prior functioning and performance at least 1.5 standard deviations below normative expectations on formal neuropsychological testing, with impairment required in episodic memory or executive function. As the neuropsychological batteries differed between cohorts and included different numbers and types of tests, cognitive impairment was operationalized using comparable criteria across studies. In DIAGNOSE CTE, participants were assigned provisional levels of certainty for CTE pathology (suggestive, possible, or probable) based on TES criteria and supportive clinical features. The same TES adjudication framework was applied in PABHS; however, provisional levels of certainty for CTE pathology were not assigned in that cohort. Study Population Participants from both cohorts with critical missing data were excluded. To maintain comparability between cohorts, participants with TES based solely on neurobehavioral dysregulation without evidence of cognitive impairment were excluded from the present analyses. Although neurobehavioral dysregulation was considered during TES adjudication in both cohorts, the methods used to ascertain these symptoms differed between studies, including the use of certain instruments, informant, or study partner reports in DIAGNOSE CTE that were not available in PABHS. The resulting analytic sample therefore included TES-positive (TES+) participants with cognitive features and TES-negative (TES-) participants across both cohorts. Participants in both cohorts provided written informed consent prior to any study procedures at their respective study sites. Study Variables To evaluate whether TES is characterized by convergence across clinical and biological domains, we constructed three complementary predictor indices representing cognitive function, blood-based biomarkers, and volumetric MRI regions: Cognitive Index . Cognitive impairment is central to TES and required for possible or probable CTE diagnoses. The cognitive index was constructed to reflect TES-relevant cognitive features while accommodating cohort-specific neuropsychological batteries. Given that cognitive impairment contributes to TES adjudication, some measures included in the cognitive index overlap with those used to define TES classification; accordingly, the cognitive index represents the clinical domain of the syndrome rather than an independent predictor. Because the specific instruments differed across cohorts, cognitive variables were harmonized at the domain level (e.g., memory, processing speed, executive function) rather than by individual tests. Standardized performance metrics were used to represent each domain, and indices were derived and analyzed separately within each cohort. Comparisons between cohorts were therefore performed at the level of overall model patterns rather than pooled variable-level data. For PABHS, cognitive performance was derived from standardized computer-based assessments administered through CNS Vital Signs and the Cleveland Clinic Concussion App (C3 Logix), supplemented by a semantic verbal fluency task using Animal Naming. 24 , 25 The PABHS cognitive index encompassed seven variables: educational attainment as a marker of cognitive reserve, verbal memory, verbal fluency, processing speed, psychomotor speed, reaction time, and executive control expressed through speeded task performance. For DIAGNOSE CTE, cognitive performance was derived from a comprehensive neuropsychological battery spanning attention, psychomotor speed, executive function, learning and memory, language, and visuospatial ability. 26 Raw scores were converted to age, sex, and/or education-adjusted T-scores using established normative data. 13 For the present analyses, the DIAGNOSE CTE cognitive index incorporated six variables: educational attainment, verbal memory, verbal fluency, attention, executive function, and psychomotor or processing speed. Processing speed, psychomotor speed, and reaction time represent related but distinct components of cognitive function and motor-cognitive integration. Impairments in these activities have been reported in RHI-exposed cohorts, including in samples characterized using TES criteria. 6 , 14 These domains were included to capture broader cognitive variability for analytic comparison and were not used for TES adjudication. Specifically, because cognitive impairment is a core feature of TES, the cognitive index is not independent of the outcome. Therefore, sensitivity analyses excluding the cognitive index were conducted to evaluate whether biomarker and imaging domains independently associated with TES. Blood Biomarker Index . Plasma biomarkers selected as predictors in this analysis included NfL, GFAP, total tau protein, and tau protein phosphorylated at threonine 231 (pTau231). These biomarkers were selected a priori based on prior literature implicating neuroaxonal injury, glial response, and tau-related processes in RHI and neurodegenerative risk. To ensure analytic consistency, only biomarkers measured in both PABHS and DIAGNOSE CTE were included in the present index. Apolipoprotein E ( APOE ) genotype was determined for all PABHS and DIAGNOSE CTE participants at baseline to identify the presence of the APOE -ε 4 allele. Blood biomarkers were measured using Single molecule array (Simoa) technology on an HD-X platform (Quanterix, Billerica, MA, USA) at the Clinical Neurochemistry Laboratory, Sahlgrenska University Hospital, Mölndal, Sweden by laboratory technicians who were blinded to the clinical data. These samples, while analyzed on the same platform, were performed in separate batches among the cohorts. Batch-bridging using internal quality control samples ascertained comparability of the measurements. A full description of the sample collection process was previously published for both cohorts. 21 , 27 , 28 The blood biomarker index comprised a mixture of five predictors, including plasma total tau, pTau231, GFAP, NfL, and a dichotomous APOE -ε 4 carrier status. Imaging Index . High-resolution T1-weighted anatomical MRIs were performed at a single timepoint for DIAGNOSE CTE participants and at each annual visit for PABHS athletes. In accordance with the methodology described above, for participants with multiple visits, only the scan from the most recent visit was included in the present analyses to ensure temporal alignment across study variables. For PABHS, all imaging was acquired at a single site. DIAGNOSE CTE imaging was acquired across multiple sites using harmonized acquisition protocols. 22 The Siemens Skyra with a 32-channel head coil was used to acquire structural 3D T1-weighted magnetization-prepared rapid acquisition gradient echo images across both cohorts. Imaging data underwent quality control using FreeSurfer quality analysis tools (FreeSurfer 5.3 QATools, 2021) and visual inspection of cortical and subcortical segmentations. Reconstructions that did not meet predefined quality thresholds, including a minimum signal-to-noise ratio of 16, were reprocessed according to standardized procedures. No scans from either cohort were excluded in the current analysis based on quality control criteria. Fifteen regions of interest were pre-specified for the imaging index based on prior epidemiological and pathological studies. 1 , 14 , 17 , 29 , 30 These included averaged left and right subcortical gray matter volumes (thalamus, caudate, hippocampus, putamen, amygdala, pallidum), ventricular volumes (lateral ventricles, inferior lateral ventricles, third ventricle, fourth ventricle), and five anatomically distinct corpus callosum subregions (posterior, mid-posterior, central, mid-anterior, anterior). The corpus callosum was subdivided to capture regional vulnerability of interhemispheric white matter tracts to RHI. Volumetric regions were calculated using FreeSurfer’s automated full-brain segmentation process (version v.6; FreeSurfer, Boston, MA). Mean values of hemispherical volumes were calculated to create an average of left and right cerebral and ventricular regions. To ensure consistent directionality of volumetric predictors, ventricular volumes were multiplied by -1, as increased ventricular size reflects greater atrophy and should therefore correspond to lower structural integrity consistent with reductions in subcortical volumes. Imaging features were analyzed within cohort-specific models rather than pooled across cohorts. Data Analysis Initial evaluations included descriptive statistics of the study populations and comparisons between TES groups using t-tests for continuous variables and Fisher-Freeman-Halton test for categorical variables. Because predictors were measured on different scales, all variables were converted to percentile ranks prior to analysis to place predictors on a common scale and reduce the influence of extreme values during mixture component estimation. For imaging variables, volumetric measures were adjusted for intracranial volume prior to percentile transformation to account for individual differences in head size. We performed uni-index and multi-index generalized weighted quantile sum (GWQS) regressions to examine how the cognitive, blood biomarker, and imaging predictors were associated with TES individually and simultaneously within each cohort. Indices were constructed by transforming individual variables within each domain into quantiles and combining them into a single weighted index, with weights estimated through bootstrap sampling to reflect the relative contribution of each variable to the overall association with TES. The adjusted multi-index models included age, race, competition status (active vs retired; collegiate vs professional), and RHI exposure proxy (number of professional fights or years of football participation). Sex was included as a covariate in PABHS only, as all DIAGNOSE CTE participants were male. This is presented in the equation below: Logit (Y = 1) = β 0 ​+ θ 1 ​WQS 1 ​ + θ 2 ​WQS 2 ​ + θ 3 ​WQS 3 ​ + γ 1 ​Z 1 ​ + γ 2 ​Z 2 ​ + γ 3 ​Z 3 ​​ + γ 4 ​Z 4 ​ + γ 5 ​Z 5 where Y is the binary outcome of TES (negative = 0; positive = 1), (WQS 1 , WQS 2 , WQS 3 ) are the grouped cognitive, blood biomarker, and imaging indices, (θ 1 , θ 2 , θ 3 ) are the regression coefficients representing the effect of each index, β 0 is the model intercept, and (γ 1 , γ 2 , γ 3 , γ 4 , γ 5 ) are the regression slopes for the covariates (age, sex, race, RHI exposure, status). Note that the model for DIAGNOSE CTE does not include γ 5 ​Z 5 ​. Each model used 1000 bootstrap iterations to improve the stability of the estimated mixture weights, with final weights defined as the average across bootstrap samples. The estimated coefficients of each index were obtained using logistic regression with maximum likelihood estimation. Because both multi-index models suffered from quasi-complete separation, the Firth correction was utilized to resolve this modeling issue. 31 Odds ratios (ORs) were calculated as the exponentiated regression coefficients. The contributions of individual variables within each index were assessed through empirically estimated weights. A larger weight indicated a greater contribution from each variable to the corresponding index. These weights reflect the relative contribution of correlated predictors to model discrimination within the GWQS framework and should not be interpreted as evidence of biological hierarchy, causal influence, or mechanistic pathways. To aid interpretation, a threshold value (τ), defined as the reciprocal of the number of variables within each index, was used to identify variables contributing more than would be expected under equal weighting (cognitive index τ ≈ 0.17; blood biomarker index τ = 0.20; imaging index τ ≈ 0.07). To assess whether associations between biological measures and TES were independent of the cognitive index, sensitivity analyses were conducted using models including only biomarker and imaging indices. Model performance was evaluated using discrimination and classification metrics, including the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and positive predictive value (PPV). Calibration was additionally assessed using smoothed calibration plots and the Brier score. Analyses were conducted in RStudio 2025.05.0 (RStudio, PBC, Massachusetts), with statistical significance defined as p < 0.05. Results This analysis consisted of 307 participants, including 158 professional fighters (35% racial minority, 14% female) from the PABHS and 149 American football players from DIAGNOSE CTE (33% racial minority, 0% female). Among all participants, 51% (n=158) were identified as TES+ (PABHS=41%; DIAGNOSE CTE=63%). Across cohorts, TES+ participants had greater RHI exposure, lower educational attainment, and were more likely to be retired (PABHS) or former professional-level (DIAGNOSE CTE) athletes compared to TES- ( Table 1 ). All DIAGNOSE CTE participants were ≥45 years of age (p-value=0.7010), whereas 61% of participants in PABHS were younger than 45 years (p-value=0.0010). There were no significant differences between biological sex within the TES groups for the PABHS athletes (p-value=0.8470). Sex differences were not evaluated in DIAGNOSE CTE, as all participants were male. A full description of cohort characteristics and index-level classifications are provided in Table 1 . [ Table 1 ] In PABHS, the adjusted multi-index model including age, race, sex, competition status (active vs retired), and RHI exposure (number of professional fights) as covariates demonstrated excellent discrimination of TES (AUC=0.91), whereas discrimination remained strong but more modest in DIAGNOSE CTE (AUC=0.84). At the Youden-optimal threshold, the PABHS model achieved a sensitivity of 0.91 and specificity of 0.84 (PPV=0.80). In DIAGNOSE CTE, sensitivity was 0.84 and specificity was 0.73 (PPV=0.85). Calibration analyses demonstrated good agreement between predicted probabilities and observed TES status in both cohorts, with Brier scores of 0.11 in PABHS and 0.15 in DIAGNOSE CTE ( Additional file 1: Figure A1 ). PABHS In the PABHS cohort, each index was significantly associated with TES in uni-index models (Cognitive AUC=0.85/PPV=0.71; Biomarker AUC=0.62/PPV=0.69; Imaging AUC=0.81/PPV=0.78). The AUC and PPV values for the unadjusted multi-index model, however, performed better than the uni-index models. In the unadjusted multi-index model, higher values (improved scores) of the cognitive index were associated with 75% decreased odds of TES (OR=0.25; 95% confidence interval [CI]=0.16, 0.37; p-value<.0001) and higher values of the imaging index were associated with 70% lower odds of TES (OR=0.30; 95% CI=0.20, 0.43; p-value<.0001). The blood biomarker index did not significantly influence TES status in the unadjusted model. The adjusted multi-index model indicated the best fit for this dataset (AUC=0.91/PPV=0.80), as it demonstrated improved discrimination compared with individual indices. In the adjusted model, higher values of the cognitive index remained strongly associated with lower odds of TES (OR=0.25; 95% CI=0.14, 0.42; p-value<.0001) while the imaging index independently contributed additional discriminatory information, with a 65% lower odds of TES per unit increase (OR=0.35; 95% CI=0.17, 0.65; p-value=0.0015). Female sex (OR=9.36; 95% CI=2.44, 40.10; p-value=0.0014), older age (OR=4.01; 95% CI=1.41, 11.94; p-value=0.0101), and greater RHI exposure (OR=1.05; 95% CI=1.02, 1.08; p-value=0.0016) all significantly predicted the odds of TES ( Table 2 ). This estimate should be interpreted cautiously given the small percentage of female participants. Similar to the unadjusted model, the blood biomarker index did not significantly influence TES status in the adjusted model. The estimated weights for individual predictor variables in the adjusted multi-index model are shown in Figure 1 . In the cognitive index, the following predictors had a weight greater than τ=0.14: symbol digit coding (0.27), processing speed (0.26), and psychomotor speed (0.24). In the blood biomarker index, the following predictors had a weight greater than τ=0.20: NfL (0.29), APOE -ε4 allele (0.22), and GFAP (0.20). In the imaging index, the following predictors had a weight greater than τ=0.07: volumes of 3 rd ventricle (0.29), posterior corpus callosum (0.24), amygdala (0.16), and putamen (0.09). [ Table 2 ] DIAGNOSE CTE In the DIAGNOSE CTE cohort, each index demonstrated significant association with TES in uni-index models (Cognitive AUC=0.78/PPV=0.84; Biomarker AUC=0.62/PPV=0.76; Imaging AUC=0.62/PPV=0.74). The AUC and PPV values were improved in the unadjusted multi-index model. In the unadjusted multi-index model, higher values of the cognitive index were associated with 83% lower odds of TES (OR=0.17; 95% CI=0.08, 0.31; p-value<.0001) while higher values of the blood biomarker index were associated with a 107% increase in the odds of TES (OR=2.07; 95% CI=1.32, 3.38; p-value=0.0023). The imaging index did not significantly influence TES status in the unadjusted model. The adjusted multi-index model including age, race, competition status (college vs professional), and RHI exposure (years of football) as covariates indicated the best fit for this dataset (AUC=0.84; PPV=0.84). Similar to PABHS findings, analysis of the data supported our hypothesis that TES is best characterized by a multidomain profile, as integration of cognitive, biomarker, and imaging indices improved model discrimination beyond any single domain alone. In the adjusted model, the cognitive index remained strongly associated with lower odds of TES (OR=0.15; 95% CI=0.07, 0.30; p-value<.0001) while the blood biomarker index independently contributed additional discriminatory information, with a 125% increase in the odds of TES per unit increase (OR=2.25; 95% CI=1.39, 3.83; p-value=0.0015) ( Table 2 ). RHI exposure (OR=1.22; 95% CI=1.06, 1.43; p-value=0.0081) significantly predicted the odds of TES, while competition status and age did not. Similar to the unadjusted model, the imaging index did not significantly influence TES status in the adjusted model. In the cognitive index, the following predictors had a weight greater than τ=0.17: executive speed (0.37), verbal memory (0.29), and educational attainment (0.19). In the blood biomarker index, the following predictors had a weight greater than τ=0.20: APOE -ε4 allele (0.41) and total tau (0.38). In the imaging index, the following predictors had a weight greater than τ=0.07: caudate (0.17), 3 rd ventricle (0.09), lateral ventricle (0.08), 4 th ventricle (0.07), central corpus callosum (0.07), and amygdala (0.07). Summary statistics and visualization of the weighted cognitive, blood biomarker, and imaging indices by TES status are provided in Appendix ( Figure A2 and Table A3) , demonstrating distinct separation in index distributions between TES+ and TES- participants across both cohorts. Sensitivity Analysis: Biomarker and Imaging Indices Without Cognitive Index To evaluate whether associations between biological measures and TES were independent of the cognitive index, a sensitivity analysis, visualized in Table 3, was conducted including only biomarker and imaging indices. [Table 3] In PABHS, the unadjusted model demonstrated a significant association for the imaging index (OR=0.28; 95% CI=0.18, 0.42; p-value<.0001), whereas the blood biomarker index was not associated with TES status. After adjustment, both indices were significantly associated with TES (biomarker index OR=2.06; 95% CI=1.06, 4.28; p-value=0.0455; imaging index OR=0.18, 95% CI=0.09, 0.35; p-value<.0001), with strong model discrimination (AUC=0.88; PPV=0.72). In DIAGNOSE CTE, both indices were associated with TES in the unadjusted model (biomarker index OR=1.60; 95% CI=1.08, 2.42, p-value=0.0204; imaging index OR=0.60; 95% CI=0.37, 0.94; p-value=0.0277). These associations remained significant in adjusted models (biomarker index OR=1.65; 95% CI=1.01, 2.74; p-value=0.0486; imaging index OR=0.48; 95% CI=0.26, 0.84; p-value=0.0133), with modest model discrimination (AUC=0.71; PPV=0.78). Overall, results were consistent with the primary analyses. Although the adjusted multidomain model incorporating cognitive, biomarker, and imaging indices demonstrated the strongest discrimination of TES, biomarker and imaging indices remained significantly associated with TES when the cognitive index was excluded from the model. [ Figure 1 ] Discussion By integrating cognitive, imaging, and blood biomarker data from two independent cohorts of athletes exposed to RHI, this study demonstrates consistent associations between TES and combined cognitive, biomarker, and neuroimaging indices across two independent cohorts. Across both cohorts, multi-index models demonstrated improved model performance compared with single-domain models, with adjusted AUC values of 0.91 in PABHS and 0.84 in DIAGNOSE CTE. Even when the cognitive index was excluded, blood biomarker and imaging indices remained associated with TES across both cohorts, indicating that objective biological measures contribute substantial information beyond clinical features alone. Although the overall pattern of findings was consistent across cohorts, the relative contribution of biological domains differed. Structural imaging features contributed more prominently to model discrimination among professional fighters (PABHS), whereas blood biomarker signals were more strongly associated with TES among former football players (DIAGNOSE CTE). These differences may reflect variation in cumulative injury burden, age distribution, or – in cases where CTE is present – stage of neurodegenerative progression between cohorts. Together, these findings support the concept that TES reflects a broader clinicobiological phenotype rather than an exclusively symptom-defined construct. Likewise, the reproducibility of the outcomes in two groups of athletes with unique RHI exposures increases the credibility and applicability of the results. The GWQS framework complements other multivariate approaches such as the Global Statistical Test (GST) by empirically weighting correlated predictors to quantify their relative contributions to the overall association. While GST provides a robust global measure of group-level differences, GWQS adds variable-level interpretability, allowing clearer identification of the factors most strongly linked to diagnostic status. 19 This approach is promising for risk stratification, cohort enrichment, and experimental therapeutic efforts in populations exposed to RHI. Our results demonstrate the value of utilizing an integrated approach over individual measures for identifying individuals with TES who might eventually be shown to have CTE. We hypothesize that athletes with TES that incorporates fluid biomarker and imaging changes are more likely to have CTE than those without these concomitant indicators of brain pathology. Our findings reinforce previous literature suggesting that cognitive deficits associated with RHI exposure are diverse while also identifying cognitive features common to both cohorts. 6 , 17 , 32 , 33 In both cohorts’ cognitive indices, processing speed exceeded the τ threshold, indicating a prominent role in differentiating TES + and TES- status among both groups. These results are consistent with prior studies that have identified a decline in processing speed as associated with neurodegeneration related to RHI exposure. 17 Verbal memory was more heavily weighted in the retired football players than the fighters. This may reflect the more precise testing and quantification of neurocognitive assessments in the DIAGNOSE CTE cohort. The inverse association between educational attainment and TES supports cognitive reserve theory, in which higher education serves as a protective factor against clinical symptom expression despite potential underlying pathology. 34 While GWQS-derived weights do not imply causality or biological mechanism, the relative prominence of specific biomarkers across cohorts provides a descriptive framework for contextualizing existing biological literature. The biomarker indices revealed APOE -ε4 allele received the highest relative weight within the blood biomarker index in both cohorts, along with NfL and GFAP in fighters, whereas former football players showed elevated total tau. These findings are consistent with prior literature identifying these markers as associated with astroglial activation, axonal damage, and neurodegeneration in RHI-exposed populations. 1 , 35 – 37 Many of the PABHS participants are under the age of 45, which may contribute to lower tau burden than the DIAGNOSE CTE participants. Furthermore, the observed differences between cohorts may reflect not only methodological and demographic factors, but also the distinct biomechanical forces experienced in combat sports versus football. Impacts in combat sports often involve rotational acceleration and focal blows, potentially producing different injury patterns or affecting distinct brain regions compared to the linear and repetitive collisions common in football. While variability in imaging indices may partly result from differences in scanner platforms and cohort age, the potential influence of sport-specific exposure mechanics warrants further investigation. The consistent prominence of the APOE -ε 4 carrier across both cohorts underscores its potential role in the identification of TES risk. Present in approximately 60–70% of individuals with AD, APOE -ε 4 is well established as a genetic risk factor that promotes amyloid deposition and accelerates tau-related neurodegeneration. 38 APOE -ε 4 has also been linked to heightened inflammatory responses, including increased activation of astrocytes and microglia through impaired triglyceride metabolism. 39 These mechanistic pathways align with our broader findings, in which APOE -ε 4 clustered with elevated GFAP may be indicative of a neuroinflammatory response. Recent neuropathological evidence has demonstrated that APOE -ε 4 is associated with greater severity of CTE pathology, supporting a role for this allele in modifying disease progression following RHI exposure. 40 In the context of RHI, APOE -ε 4 may amplify vulnerability to both proteinopathies and inflammation, leading to degeneration and subsequent clinical decline. The current diagnostic criteria for TES depend solely on clinical domains. 2 These criteria rely on symptom reporting and neuropsychological performance, which may reflect heterogeneous processes, may not capture early biological changes, and may overlap with other neurological or psychiatric conditions. In addition, the clinical data required for TES adjudication often involve comprehensive neuropsychological evaluation and longitudinal clinical assessment, which can be resource-intensive and may limit scalability. Integrating biologic data with clinical observations represents one approach to resolving this dilemma. In contrast, the biological measures included in the multi-index model, such as blood biomarkers and structural MRI, can be obtained using standardized protocols and may offer more scalable and objective approaches to monitoring brain health. Specifically, the multi-index model developed here can serve as a stratification tool for identifying high-risk individuals or as an inclusion criterion for clinical trials investigating disease targeted therapies. The improved positive predictive values (PPV = 80 and 85%, respectively) underscore the potential to reduce heterogeneity and improve cohort stratification for such trials. This study had several strengths, including a large sample size (N = 307) and reproducibility of findings from two cohorts of professional athletes with substantial RHI exposure. Several limitations are acknowledged. Although the TES criteria applied here are currently the best available, the lack of pathological confirmation restricts our ability to validate and directly link index performance to underlying CTE pathology. The statistical modeling in this study, while valuable for identifying individual and cumulative weights, limits inferences about causality or the progression of the syndrome. The DIAGNOSE CTE cohort was restricted to individuals aged 45 years and older, whereas the PABHS cohort included many younger participants. Age differences influence both biomarker expression and neurodegenerative burden. Furthermore, some of the cognitive measures applied normative values derived from general population samples that differ from the demographic and educational profiles of the PABHS cohort. As the PABHS participants generally have lower educational attainment than the DIAGNOSE CTE participants, applying standard normative corrections may have led to misidentification or overestimation of impairment. The absence of behavioral information in the PABHS cohort may have influenced the diagnosis and characteristics of TES identified in this cohort. The p-tau measure available was p-tau231 and other p-tau species may be more informative. Our findings should be interpreted as descriptive and hypothesis-generating with respect to TES identification. Future studies should evaluate the independent contribution of neurobehavioral dysregulation to TES in cohorts where these symptoms are systematically measured. Such analyses may help clarify whether neurobehavioral features contribute meaningfully to multidomain TES profiles or represent a distinct clinical dimension of RHI exposure. Conclusion TES in RHI-exposed athletes was associated with a convergent clinicobiological profile across two independent cohorts. In both cohorts, multidomain models integrating cognitive, biomarker, and neuroimaging indices demonstrated stronger discrimination of TES than single-domain models. Biomarker and imaging indices contributed additional information across cohorts, although imaging contributions were more prominent in fighters while biomarker associations were stronger in football players. These findings support multidomain analytic frameworks for examining biological signals associated with TES and may inform future studies aimed at improving risk stratification for CTE. Declarations Ethics approval and consent to participate The current study represents a secondary analysis of a de-identified database with no contact with participants; therefore, it was submitted to and approved by the University of Nevada, Las Vegas Institutional Review Board for administrative review of exempt research (UNLV#2024-245). This study was conducted in accordance with applicable institutional policies and federal regulations, including 45 CFR 46 and the HIPAA Privacy Rule, and adhered to the principles of the Declaration of Helsinki. Consent for publication Not applicable Availability of data and materials The datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request. Competing interests JLC has provided consultation to Acadia, Acumen, ALZpath, AnnovisBio, Artery, Axsome, Biogen, Bristol-Myers Squibb, Eisai, Fosun, GAP Foundation, Hummingbird Diagnostics, IGC, Janssen, Julius Clinical, Kinoxis, Lilly, LSP/EQT, Merck, MoCA Cognition, Novo Nordisk, NSC Therapeutics, Otsuka, ReMYND, Roche, Scottish Brain Sciences, Signant Health, Simcere, Sinaptica, and T-Neuro pharmaceutical, assessment, and investment companies. HZ has served at scientific advisory boards and/or as a consultant for Abbvie, Acumen, Alamar, Alector, Alzinova, ALZpath, Amylyx, Annexon, Apellis, Artery Therapeutics, AZTherapies, Cognito Therapeutics, CogRx, Denali, Eisai, Enigma, LabCorp, Merck Sharp & Dohme, Merry Life, Nervgen, Novo Nordisk, Optoceutics, Passage Bio, Pinteon Therapeutics, Prothena, Quanterix, Red Abbey Labs, reMYND, Roche, Samumed, ScandiBio Therapeutics AB, Siemens Healthineers, Triplet Therapeutics, and Wave, has given lectures sponsored by Alzecure, BioArctic, Biogen, Cellectricon, Fujirebio, LabCorp, Lilly, Novo Nordisk, Oy Medix Biochemica AB, Roche, and WebMD, is a co-founder of Brain Biomarker Solutions in Gothenburg AB (BBS), which is a part of the GU Ventures Incubator Program, and is a shareholder of CERimmune Therapeutics (outside submitted work). The other authors report no competing interests. Funding CB has received research funding from the UFC, Top Rank Promotions, Haymon Boxing, and Bellator Promotions. MLA reports honoraria from the Michael J. Fox Foundation for services unrelated to this work, royalties from Oxford University Press Inc., and research support from Life Molecular Imaging Inc. JLC is supported by NIGMS grant P20GM109025, NIA R35AG71476, NIA R25AG083721, and NINDS R01NS139383, as well as funding from the Alzheimer’s Disease Drug Discovery Foundation, the Ted and Maria Quirk Endowment, and the Joy Chambers-Grundy Endowment. DIAGNOSE CTE was supported by NINDS U01NS093334. PABHS is supported by philanthropic grants from the Ultimate Fighting Championship, Top Rank Boxing, and Haymon Boxing. HZ is a Wallenberg Scholar and a Distinguished Professor at the Swedish Research Council supported by grants from the Swedish Research Council (#2023-00356, #2022-01018 and #2019-02397), the European Union’s Horizon Europe research and innovation programme under grant agreement No 101053962, and Swedish State Support for Clinical Research (#ALFGBG-71320). The remaining authors report no relevant funding. Authors’ contributions BCK conceived and designed the study, performed the statistical analyses with LC, and drafted the manuscript. LC contributed to study design, performed the statistical analyses, and contributed to interpretation of the results. DS, MLA, JVW, YT, CHA, MES, OP, DIK, EP, LJB, IKK, JM, EMR, RCC, RAS, and HZ contributed to data acquisition, interpretation of results, and critical revision of the manuscript for important intellectual content. CB and JLC provided senior oversight, contributed to study design and interpretation, and critically revised the manuscript. All authors reviewed and approved the final manuscript. Acknowledgements We thank the PABHS and DIAGNOSE CTE research teams at the Lou Ruvo Center for Brain Health and Boston University for their contributions to data collection and study coordination. We also thank the athletes who participated in these studies for their time and commitment to advancing understanding of the long-term effects of repetitive head impacts. References McKee AC, Stein TD, Huber BR, et al. Chronic traumatic encephalopathy (CTE): criteria for neuropathological diagnosis and relationship to repetitive head impacts. Acta Neuropathol . Apr 2023;145(4):371-394. doi:10.1007/s00401-023-02540-w Katz DI, Bernick C, Dodick DW, et al. National Institute of Neurological Disorders and Stroke Consensus Diagnostic Criteria for Traumatic Encephalopathy Syndrome. Neurology . May 4 2021;96(18):848-863. doi:10.1212/WNL.0000000000011850 Conway Kleven BD, Chien LC, Cross CL, Labus B, Bernick C. Traumatic Encephalopathy Syndrome: Head Impact Exposure and Blood Biomarkers in Professional Combat Athletes. J Head Trauma Rehabil . Feb 25 2025;doi:10.1097/htr.0000000000001048 Ashton NJ, Hye A, Rajkumar AP, et al. An update on blood-based biomarkers for non-Alzheimer neurodegenerative disorders. Nature Reviews Neurology . 2020/05/01 2020;16(5):265-284. doi:10.1038/s41582-020-0348-0 Asken BM, Tanner JA, VandeVrede L, et al. Multi-Modal Biomarkers of Repetitive Head Impacts and Traumatic Encephalopathy Syndrome: A Clinicopathological Case Series. J Neurotrauma . Sep 2022;39(17-18):1195-1213. doi:10.1089/neu.2022.0060 Ritter A, Shan G, Montes A, Randall R, Bernick C. Traumatic encephalopathy syndrome: application of new criteria to a cohort exposed to repetitive head impacts. British Journal of Sports Medicine . 2023;57:389-394. Asken BM, DeKosky ST, Clugston JR, Jaffee MS, Bauer RM. Diffusion tensor imaging (DTI) findings in adult civilian, military, and sport-related mild traumatic brain injury (mTBI): a systematic critical review. Brain imaging and behavior . 2018;12(2):585-612. Asken BM, Rabinovici GD. Identifying degenerative effects of repetitive head trauma with neuroimaging: a clinically-oriented review. Acta neuropathologica communications . 2021;9(1):96. Kinnunen KM, Cash DM, Poole T, et al. Presymptomatic atrophy in autosomal dominant Alzheimer's disease: A serial magnetic resonance imaging study. Alzheimers Dement . Jan 2018;14(1):43-53. doi:10.1016/j.jalz.2017.06.2268 Rohrer JD, Nicholas JM, Cash DM, et al. Presymptomatic cognitive and neuroanatomical changes in genetic frontotemporal dementia in the Genetic Frontotemporal dementia Initiative (GENFI) study: a cross-sectional analysis. Lancet Neurol . Mar 2015;14(3):253-62. doi:10.1016/s1474-4422(14)70324-2 Brewer JB, Magda S, Airriess C, Smith ME. Fully-automated quantification of regional brain volumes for improved detection of focal atrophy in Alzheimer disease. American Journal of Neuroradiology . 2009;30(3):578-580. Lin Y, Fu Y, Zeng Y-F, et al. Six visual rating scales as a biomarker for monitoring atrophied brain volume in Parkinson’s disease. Aging and disease . 2020;11(5):1082. Alosco ML, Barr WB, Banks SJ, et al. Neuropsychological test performance of former American football players. Alzheimers Res Ther . Jan 3 2023;15(1):1. doi:10.1186/s13195-022-01147-9 Bernick C, Banks SJ, Shin W, et al. Repeated head trauma is associated with smaller thalamic volumes and slower processing speed: the Professional Fighters' Brain Health Study. Br J Sports Med . Aug 2015;49(15):1007-11. doi:10.1136/bjsports-2014-093877 Bernick C, Shan G, Ritter A, et al. Blood biomarkers and neurodegeneration in individuals exposed to repetitive head impacts. Alzheimers Res Ther . Oct 12 2023;15(1):173. doi:10.1186/s13195-023-01310-w Conway Kleven B, Chien LC, Young DL, Cross CL, Labus B, Bernick C. Repetitive head impacts among professional fighters: a pilot study evaluating Traumatic Encephalopathy Syndrome and postural balance. Phys Sportsmed . Mar 6 2024:1-7. doi:10.1080/00913847.2024.2325331 Conway Kleven BD, Chien L-C, Labus B, et al. Longitudinal changes in regional brain volumes and cognition of professional fighters with traumatic encephalopathy syndrome. Neurology . 2023;101(11):e1118-e1126. Kleven BDC, Chien L-C, Cross CL, Labus B, Bernick C. Traumatic Encephalopathy Syndrome: Head Impact Exposure and Blood Biomarkers in Professional Combat Athletes. The Journal of Head Trauma Rehabilitation . 2025:10.1097/HTR.0000000000001048. doi:10.1097/htr.0000000000001048 Wheeler DC, Rustom S, Carli M, Whitehead TP, Ward MH, Metayer C. Assessment of Grouped Weighted Quantile Sum Regression for Modeling Chemical Mixtures and Cancer Risk. International Journal of Environmental Research and Public Health . 2021;18(2):504. Bernick C, Banks S, Phillips M, et al. Professional fighters brain health study: rationale and methods. Am J Epidemiol . Jul 15 2013;178(2):280-6. doi:10.1093/aje/kws456 Alosco ML, Mariani ML, Adler CH, et al. Developing methods to detect and diagnose chronic traumatic encephalopathy during life: rationale, design, and methodology for the DIAGNOSE CTE Research Project. Alzheimer's research & therapy . 2021;13(1):136-136. doi:10.1186/s13195-021-00872-x Alosco ML, Mariani ML, Adler CH, et al. Developing methods to detect and diagnose chronic traumatic encephalopathy during life: rationale, design, and methodology for the DIAGNOSE CTE Research Project. Alzheimers Res Ther . Aug 12 2021;13(1):136. doi:10.1186/s13195-021-00872-x Stern RA, Trujillo-Rodriguez D, Tripodis Y, et al. Amyloid PET across the cognitive spectrum in former professional and college American football players: findings from the DIAGNOSE CTE Research Project. Alzheimers Res Ther . Oct 5 2023;15(1):166. doi:10.1186/s13195-023-01315-5 Gualtieri CT, Johnson LG. Reliability and validity of a computerized neurocognitive test battery, CNS Vital Signs. Archives of Clinical Neuropsychology . 2006;21(7):623-643. doi:10.1016/j.acn.2006.05.007 Simon M, Maerlender A, Metzger K, Decoster L, Hollingworth A, Valovich McLeod T. Reliability and concurrent validity of select C3 logix test components. Developmental neuropsychology . 2017;42(7-8):446-459. Alosco ML, Mariani ML, Adler CH, et al. Developing methods to detect and diagnose chronic traumatic encephalopathy during life: rationale, design, and methodology for the DIAGNOSE CTE Research Project. Alzheimer's research & therapy . 2021;13(1):136. Bernick C, Zetterberg H, Shan G, Banks S, Blennow K. Longitudinal Performance of Plasma Neurofilament Light and Tau in Professional Fighters: The Professional Fighters Brain Health Study. J Neurotrauma . Oct 15 2018;35(20):2351-2356. doi:10.1089/neu.2017.5553 Bernick C, Shan G, Ritter A, et al. Blood biomarkers and neurodegeneration in individuals exposed to repetitive head impacts. Alzheimer's Research & Therapy . 2023;15(1):173. Bernick C, Shan G, Zetterberg H, et al. Longitudinal change in regional brain volumes with exposure to repetitive head impacts. Neurology . Jan 21 2020;94(3):e232-e240. doi:10.1212/WNL.0000000000008817 Arciniega H, Baucom ZH, Tuz-Zahra F, et al. Brain morphometry in former American football players: Findings from the DIAGNOSE CTE research project. Brain . 2024;147(10):3596-3610. D'Angelo G, Ran D. Tutorial on Firth's Logistic Regression Models for Biomarkers in Preclinical Space. Pharmaceutical Statistics . 2025; Bernick C, Shan G, Bennett L, Alberts J, Cummings J. Assessing Clinical Change in Individuals Exposed to Repetitive Head Impacts: The Repetitive Head Impact Composite Index. Front Neurol . 2021;12:605318. doi:10.3389/fneur.2021.605318 LoBue C, Bickart KC, Bieniek KF, et al. Cognitive Impairment and Dementia-Related Diagnoses in Chronic Traumatic Encephalopathy: A Systematic Review and Meta-Analysis. Neurology . Aug 26 2025;105(4):e213934. doi:10.1212/wnl.0000000000213934 Banks SJ, Obuchowski N, Shin W, et al. The protective effect of education on cognition in professional fighters. Arch Clin Neuropsychol . Feb 2014;29(1):54-9. doi:10.1093/arclin/act079 Asken BM, Yang Z, Xu H, et al. Acute effects of sport-related concussion on serum glial fibrillary acidic protein, ubiquitin C-terminal hydrolase L1, total tau, and neurofilament light measured by a multiplex assay. Journal of neurotrauma . 2020;37(13):1537-1545. Gao W, Zhang Z, Lv X, et al. Neurofilament light chain level in traumatic brain injury: a system review and meta-analysis. Medicine . 2020;99(38):e22363. Mastandrea P, Mengozzi S, Bernardini S. Systematic review and meta-analysis of observational studies evaluating glial fibrillary acidic protein (GFAP) and ubiquitin C-terminal hydrolase L1 (UCHL1) as blood biomarkers of mild acute traumatic brain injury (mTBI) or sport-related concussion (SRC) in adult subjects. Diagnosis (Berl) . Aug 20 2024;doi:10.1515/dx-2024-0078 Garcia MJ, Leadley R, Ross J, et al. Prognostic and Predictive Factors in Early Alzheimer's Disease: A Systematic Review. J Alzheimers Dis Rep . 2024;8(1):203-240. doi:10.3233/adr-230045 Stephenson RA, Sepulveda J, Johnson KR, et al. Triglyceride metabolism controls inflammation and microglial phenotypes associated with APOE 4. Cell Rep . Jul 22 2025;44(7):115961. doi:10.1016/j.celrep.2025.115961 Atherton K, Han X, Chung J, et al. Association of APOE Genotypes and Chronic Traumatic Encephalopathy. JAMA Neurology . 2022;79(8):787-796. doi:10.1001/jamaneurol.2022.1634 Jia J, Ning Y, Chen M, et al. Biomarker changes during 20 years preceding Alzheimer’s disease. New England Journal of Medicine . 2024;390(8):712-722. Thakur R, Saini AK, Taliyan R, Chaturvedi N. Neurodegenerative diseases early detection and monitoring system for point-of-care applications. Microchemical Journal . 2025;208:112280. Tables Tables 1 to 3 are available in the Supplementary Files section. Additional Declarations Competing interest reported. Competing interests JLC has provided consultation to Acadia, Acumen, ALZpath, AnnovisBio, Artery, Axsome, Biogen, Bristol-Myers Squibb, Eisai, Fosun, GAP Foundation, Hummingbird Diagnostics, IGC, Janssen, Julius Clinical, Kinoxis, Lilly, LSP/EQT, Merck, MoCA Cognition, Novo Nordisk, NSC Therapeutics, Otsuka, ReMYND, Roche, Scottish Brain Sciences, Signant Health, Simcere, Sinaptica, and T-Neuro pharmaceutical, assessment, and investment companies. HZ has served at scientific advisory boards and/or as a consultant for Abbvie, Acumen, Alamar, Alector, Alzinova, ALZpath, Amylyx, Annexon, Apellis, Artery Therapeutics, AZTherapies, Cognito Therapeutics, CogRx, Denali, Eisai, Enigma, LabCorp, Merck Sharp & Dohme, Merry Life, Nervgen, Novo Nordisk, Optoceutics, Passage Bio, Pinteon Therapeutics, Prothena, Quanterix, Red Abbey Labs, reMYND, Roche, Samumed, ScandiBio Therapeutics AB, Siemens Healthineers, Triplet Therapeutics, and Wave, has given lectures sponsored by Alzecure, BioArctic, Biogen, Cellectricon, Fujirebio, LabCorp, Lilly, Novo Nordisk, Oy Medix Biochemica AB, Roche, and WebMD, is a co-founder of Brain Biomarker Solutions in Gothenburg AB (BBS), which is a part of the GU Ventures Incubator Program, and is a shareholder of CERimmune Therapeutics (outside submitted work). The other authors report no competing interests. Funding CB has received research funding from the UFC, Top Rank Promotions, Haymon Boxing, and Bellator Promotions. MLA reports honoraria from the Michael J. Fox Foundation for services unrelated to this work, royalties from Oxford University Press Inc., and research support from Life Molecular Imaging Inc. JLC is supported by NIGMS grant P20GM109025, NIA R35AG71476, NIA R25AG083721, and NINDS R01NS139383, as well as funding from the Alzheimer’s Disease Drug Discovery Foundation, the Ted and Maria Quirk Endowment, and the Joy Chambers-Grundy Endowment. DIAGNOSE CTE was supported by NINDS U01NS093334. PABHS is supported by philanthropic grants from the Ultimate Fighting Championship, Top Rank Boxing, and Haymon Boxing. HZ is a Wallenberg Scholar and a Distinguished Professor at the Swedish Research Council supported by grants from the Swedish Research Council (#2023-00356, #2022-01018 and #2019-02397), the European Union’s Horizon Europe research and innovation programme under grant agreement No 101053962, and Swedish State Support for Clinical Research (#ALFGBG-71320). The remaining authors report no relevant funding. Supplementary Files Appendix.docx Tables.docx Cite Share Download PDF Status: Under Review Version 1 posted Reviews received at journal 12 May, 2026 Reviewers agreed at journal 24 Apr, 2026 Reviewers agreed at journal 19 Apr, 2026 Reviewers agreed at journal 18 Apr, 2026 Reviewers invited by journal 18 Apr, 2026 Editor assigned by journal 16 Apr, 2026 Submission checks completed at journal 16 Apr, 2026 First submitted to journal 11 Apr, 2026 You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-9385305","acceptedTermsAndConditions":true,"allowDirectSubmit":false,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":629138635,"identity":"a52766fa-6d91-4277-b7b7-e76feb53b56a","order_by":0,"name":"Brooke Conway Kleven","email":"data:image/png;base64,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","orcid":"","institution":"University of Nevada, Las Vegas","correspondingAuthor":true,"prefix":"","firstName":"Brooke","middleName":"Conway","lastName":"Kleven","suffix":""},{"id":629138639,"identity":"008cf08b-c98e-4e9e-8f25-6c84b8229b60","order_by":1,"name":"Lung-Chang Chien","email":"","orcid":"","institution":"University of Nevada, Las Vegas","correspondingAuthor":false,"prefix":"","firstName":"Lung-Chang","middleName":"","lastName":"Chien","suffix":""},{"id":629138642,"identity":"925554dd-3eb6-4bae-9133-00c7235f9525","order_by":2,"name":"Dana Surwill","email":"","orcid":"","institution":"University of Nevada, Las Vegas","correspondingAuthor":false,"prefix":"","firstName":"Dana","middleName":"","lastName":"Surwill","suffix":""},{"id":629138643,"identity":"6c68e4b0-07c9-40d8-aa54-9f78901b96c3","order_by":3,"name":"Michael L. Alosco","email":"","orcid":"","institution":"Boston University","correspondingAuthor":false,"prefix":"","firstName":"Michael","middleName":"L.","lastName":"Alosco","suffix":""},{"id":629138645,"identity":"919fa7e1-4916-49fb-b6b4-a7ee45a6d4d3","order_by":4,"name":"Jennifer V. Wethe","email":"","orcid":"","institution":"Mayo Clinic","correspondingAuthor":false,"prefix":"","firstName":"Jennifer","middleName":"V.","lastName":"Wethe","suffix":""},{"id":629138647,"identity":"959a9d2e-d510-4591-bd3b-0f213506df16","order_by":5,"name":"Yorghos Tripodis","email":"","orcid":"","institution":"Boston University","correspondingAuthor":false,"prefix":"","firstName":"Yorghos","middleName":"","lastName":"Tripodis","suffix":""},{"id":629138648,"identity":"fb3bae1d-a5f5-4799-8748-cfaaad412229","order_by":6,"name":"Charles H. Adler","email":"","orcid":"","institution":"Mayo Clinic","correspondingAuthor":false,"prefix":"","firstName":"Charles","middleName":"H.","lastName":"Adler","suffix":""},{"id":629138649,"identity":"cbbcb4fb-a9bc-49f8-b8c3-e3bad936aa15","order_by":7,"name":"Martha E. Shenton","email":"","orcid":"","institution":"Harvard University","correspondingAuthor":false,"prefix":"","firstName":"Martha","middleName":"E.","lastName":"Shenton","suffix":""},{"id":629138651,"identity":"e4dd17cd-dc9d-43d6-9bb8-6cce55bebeca","order_by":8,"name":"Ofer Pasternak","email":"","orcid":"","institution":"Harvard University","correspondingAuthor":false,"prefix":"","firstName":"Ofer","middleName":"","lastName":"Pasternak","suffix":""},{"id":629138653,"identity":"b8f8632f-3fe2-478a-8303-96a747f86eb5","order_by":9,"name":"Douglas I. Katz","email":"","orcid":"","institution":"Boston University","correspondingAuthor":false,"prefix":"","firstName":"Douglas","middleName":"I.","lastName":"Katz","suffix":""},{"id":629138654,"identity":"79420c5c-fd6a-4cb1-86d8-054967d205c8","order_by":10,"name":"Elaine Peskind","email":"","orcid":"","institution":"University of Washington","correspondingAuthor":false,"prefix":"","firstName":"Elaine","middleName":"","lastName":"Peskind","suffix":""},{"id":629138655,"identity":"7a1da993-e0ab-4f1a-b51f-dc2b9ac7a4cd","order_by":11,"name":"Laura J. Balcer","email":"","orcid":"","institution":"New York University","correspondingAuthor":false,"prefix":"","firstName":"Laura","middleName":"J.","lastName":"Balcer","suffix":""},{"id":629138656,"identity":"f3782b2f-c24e-4eb7-a822-1c8df43953c5","order_by":12,"name":"Inga K. Koerte","email":"","orcid":"","institution":"Ludwig-Maximilians-Universität München","correspondingAuthor":false,"prefix":"","firstName":"Inga","middleName":"K.","lastName":"Koerte","suffix":""},{"id":629138657,"identity":"132ee255-93ef-46d3-a3c3-5f93222a1d2f","order_by":13,"name":"Jesse Mez","email":"","orcid":"","institution":"Boston University","correspondingAuthor":false,"prefix":"","firstName":"Jesse","middleName":"","lastName":"Mez","suffix":""},{"id":629138658,"identity":"e3987862-89b2-4ac8-9481-742cbae84caf","order_by":14,"name":"Eric M. Reiman","email":"","orcid":"","institution":"Banner Health","correspondingAuthor":false,"prefix":"","firstName":"Eric","middleName":"M.","lastName":"Reiman","suffix":""},{"id":629138659,"identity":"050c766d-a517-497c-8320-af3ee31b05bc","order_by":15,"name":"Robert C. Cantu","email":"","orcid":"","institution":"Boston University","correspondingAuthor":false,"prefix":"","firstName":"Robert","middleName":"C.","lastName":"Cantu","suffix":""},{"id":629138660,"identity":"45a4ebb8-1b1a-4cda-b5e6-5ff2ea1940f1","order_by":16,"name":"Robert A. Stern","email":"","orcid":"","institution":"Boston University","correspondingAuthor":false,"prefix":"","firstName":"Robert","middleName":"A.","lastName":"Stern","suffix":""},{"id":629138661,"identity":"186928f8-294b-4895-861f-061ce59c7c5a","order_by":17,"name":"Henrik Zetterberg","email":"","orcid":"","institution":"University of Gothenburg","correspondingAuthor":false,"prefix":"","firstName":"Henrik","middleName":"","lastName":"Zetterberg","suffix":""},{"id":629138665,"identity":"f2dcd9dd-f885-4c2b-9822-cfb1a0d67399","order_by":18,"name":"Charles Bernick","email":"","orcid":"","institution":"Lou Ruvo Brain Institute","correspondingAuthor":false,"prefix":"","firstName":"Charles","middleName":"","lastName":"Bernick","suffix":""},{"id":629138666,"identity":"6a0c975a-13b3-4ca4-97bd-8023aec9e02c","order_by":19,"name":"Jeffrey L. Cummings","email":"","orcid":"","institution":"University of Nevada, Las Vegas","correspondingAuthor":false,"prefix":"","firstName":"Jeffrey","middleName":"L.","lastName":"Cummings","suffix":""}],"badges":[],"createdAt":"2026-04-11 06:54:02","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-9385305/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-9385305/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":108007624,"identity":"b2676ade-b096-4b9e-9b9c-a5e682255986","added_by":"auto","created_at":"2026-04-28 13:00:58","extension":"jpeg","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":233047,"visible":true,"origin":"","legend":"\u003cp\u003eEstimated weights of predictors in the a) PABHS cognitive index, b) PABHS blood biomarker index, c) PABHS imaging index, d) DIAGNOSE CTE cognitive index, e) DIAGNOSE CTE blood biomarker index, and f) DIAGNOSE CTE imaging index among the adjusted multi-index models for both cohorts.\u003c/p\u003e\n\u003cp\u003eLegend: The red lines indicate the τ values, where τ = 0.14 and 0.17 for the cognitive index (PABHS and DIAGNOSE CTE, respectively), 0.20 for the blood biomarker index, and 0.07 for the imaging index.\u003c/p\u003e","description":"","filename":"floatimage1.jpeg","url":"https://assets-eu.researchsquare.com/files/rs-9385305/v1/a44807cbadb571c54b7b4707.jpeg"},{"id":108008917,"identity":"b7ef97ff-09b8-46e4-aa81-4b49450f6394","added_by":"auto","created_at":"2026-04-28 13:08:34","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":558106,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-9385305/v1/7fd263de-f0c4-477c-bc4f-588caf061c3f.pdf"},{"id":107948172,"identity":"86ad0f3b-9f9d-490a-a946-7031f0e922f3","added_by":"auto","created_at":"2026-04-28 00:18:26","extension":"docx","order_by":1,"title":"","display":"","copyAsset":false,"role":"supplement","size":405124,"visible":true,"origin":"","legend":"","description":"","filename":"Appendix.docx","url":"https://assets-eu.researchsquare.com/files/rs-9385305/v1/c0f5e1ee317376fb0068f69b.docx"},{"id":107948174,"identity":"614cbd8f-681d-4f1e-8486-97a77828278f","added_by":"auto","created_at":"2026-04-28 00:18:26","extension":"docx","order_by":2,"title":"","display":"","copyAsset":false,"role":"supplement","size":90390,"visible":true,"origin":"","legend":"","description":"","filename":"Tables.docx","url":"https://assets-eu.researchsquare.com/files/rs-9385305/v1/b6bf91c0c68021c010b5b7ec.docx"}],"financialInterests":"Competing interest reported. Competing interests\nJLC has provided consultation to Acadia, Acumen, ALZpath, AnnovisBio, Artery, Axsome, Biogen, Bristol-Myers Squibb, Eisai, Fosun, GAP Foundation, Hummingbird Diagnostics, IGC, Janssen, Julius Clinical, Kinoxis, Lilly, LSP/EQT, Merck, MoCA Cognition, Novo Nordisk, NSC Therapeutics, Otsuka, ReMYND, Roche, Scottish Brain Sciences, Signant Health, Simcere, Sinaptica, and T-Neuro pharmaceutical, assessment, and investment companies. \nHZ has served at scientific advisory boards and/or as a consultant for Abbvie, Acumen, Alamar, Alector, Alzinova, ALZpath, Amylyx, Annexon, Apellis, Artery Therapeutics, AZTherapies, Cognito Therapeutics, CogRx, Denali, Eisai, Enigma, LabCorp, Merck Sharp \u0026 Dohme, Merry Life, Nervgen, Novo Nordisk, Optoceutics, Passage Bio, Pinteon Therapeutics, Prothena, Quanterix, Red Abbey Labs, reMYND, Roche, Samumed, ScandiBio Therapeutics AB, Siemens Healthineers, Triplet Therapeutics, and Wave, has given lectures sponsored by Alzecure, BioArctic, Biogen, Cellectricon, Fujirebio, LabCorp, Lilly, Novo Nordisk, Oy Medix Biochemica AB, Roche, and WebMD, is a co-founder of Brain Biomarker Solutions in Gothenburg AB (BBS), which is a part of the GU Ventures Incubator Program, and is a shareholder of CERimmune Therapeutics (outside submitted work). \nThe other authors report no competing interests.\nFunding\nCB has received research funding from the UFC, Top Rank Promotions, Haymon Boxing, and Bellator Promotions. MLA reports honoraria from the Michael J. Fox Foundation for services unrelated to this work, royalties from Oxford University Press Inc., and research support from Life Molecular Imaging Inc. JLC is supported by NIGMS grant P20GM109025, NIA R35AG71476, NIA R25AG083721, and NINDS R01NS139383, as well as funding from the Alzheimer’s Disease Drug Discovery Foundation, the Ted and Maria Quirk Endowment, and the Joy Chambers-Grundy Endowment. DIAGNOSE CTE was supported by NINDS U01NS093334. PABHS is supported by philanthropic grants from the Ultimate Fighting Championship, Top Rank Boxing, and Haymon Boxing. \nHZ is a Wallenberg Scholar and a Distinguished Professor at the Swedish Research Council supported by grants from the Swedish Research Council (#2023-00356, #2022-01018 and #2019-02397), the European Union’s Horizon Europe research and innovation programme under grant agreement No 101053962, and Swedish State Support for Clinical Research (#ALFGBG-71320). \nThe remaining authors report no relevant funding.","formattedTitle":"Cognitive, biomarker, and neuroimaging indices associated with traumatic encephalopathy syndrome across two independent athlete cohorts","fulltext":[{"header":"Introduction","content":"\u003cp\u003eThe long-term effects of repetitive head impacts (RHI) raise concerns regarding contact sports, including the risk of developing chronic traumatic encephalopathy (CTE)\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e\u003c/sup\u003e, a neurodegenerative disorder associated with RHI exposure. While the standard for CTE diagnosis is a post-mortem neuropathologic confirmation of hyperphosphorylated tau (pTau) deposits at the depths of cortical sulci and around small blood vessels, the National Institute of Neurological Disorders and Stroke (NINDS) developed consensus diagnostic criteria for traumatic encephalopathy syndrome (TES) to assist in research involving living individuals with a history of RHI and potential CTE.\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e The criteria for TES adjudication follows a stepwise process requiring an individual to have substantial exposure to RHI, core clinical features of neurobehavioral dysregulation and/or cognitive impairment, a progressive course of symptoms, and no other medical condition(s) that can fully account for the symptoms.\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eResearch evaluating the validity of TES has explored contributions of various biomarkers that may support the earlier diagnosis of CTE or serve as a method to track the progression of neurodegenerative processes that may occur following RHI exposure. Blood biomarkers such as neurofilament light chain protein (NfL), which serves as a marker of axonal injury, and glial fibrillary acidic protein (GFAP), indicative of activated astroglia, have been associated with a diagnosis of TES.\u003csup\u003e\u003cspan additionalcitationids=\"CR4\" citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR5\" class=\"CitationRef\"\u003e5\u003c/span\u003e\u003c/sup\u003e Magnetic resonance imaging (MRI) indicates that TES is frequently associated with subcortical and medial temporal lobe atrophy, with volumetric loss observed in regions such as the thalamus, hippocampus, and corpus callosum.\u003csup\u003e\u003cspan additionalcitationids=\"CR7\" citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR8\" class=\"CitationRef\"\u003e8\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eBiomarker studies typically evaluate markers individually rather than integrating multiple biological domains into a composite framework.\u003csup\u003e\u003cspan citationid=\"CR3\" class=\"CitationRef\"\u003e3\u003c/span\u003e,\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan additionalcitationids=\"CR16 CR17\" citationid=\"CR15\" class=\"CitationRef\"\u003e15\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR18\" class=\"CitationRef\"\u003e18\u003c/span\u003e\u003c/sup\u003e The purpose of the current study was to determine whether TES identification is characterized by the combined contribution of cognitive performance, blood biomarkers, and structural neuroimaging measures across two independent cohorts. Because cognitive impairment is currently a core feature of the TES diagnostic criteria, the cognitive domain in the present study represents the clinical component of the syndrome. We therefore hypothesized that individuals identified as TES+ would demonstrate a convergent pattern of cognitive dysfunction, biological abnormalities, and neuroanatomical changes, and that biological domains would contribute to TES identification beyond the clinical domain alone. To evaluate this hypothesis, we applied an integrative modeling framework to quantify how individual and combined domains align with TES identification.\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e We leveraged data from two independent athlete cohorts, the Professional Athletes Brain Health Study (PABHS; consisting primarily of boxers and mixed martial arts fighters)\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e and the Diagnostics, Imaging, and Genetics Network for the Objective Study and Evaluation of Chronic Traumatic Encephalopathy (DIAGNOSE CTE; consisting of former professional and college football players)\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e\u003c/sup\u003e to determine whether multimodal associations with TES are reproducible across athlete populations with distinct patterns of RHI exposure and demographic characteristics.\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003eDetails of the PABHS cohort have been previously published.\u003csup\u003e\u003cspan citationid=\"CR20\" class=\"CitationRef\"\u003e20\u003c/span\u003e\u003c/sup\u003e The PABHS is a longitudinal cohort of over 900 professional athletes with a history of RHI exposure and age/sex-matched unexposed controls. To meet the cohort\u0026rsquo;s criteria of a retired fighter, athletes must not have plans for future fights and cannot have competed in a sanctioned fight within the past two years. Active fighters must have competed in a professionally sanctioned event within the past two years and must wait at least 45 days following a fight to attend baseline and/or annual visits. For any participant with multiple visits, only the data from the most recent visit were included in the current analyses. RHI exposure was measured using the number of professional fights completed. Consistent with prior analyses from PABHS, a minimum threshold of 10 professional fights was used to operationalize substantial exposure to RHI within this cohort. This threshold was selected to represent sustained professional-level combat exposure and to exclude individuals with limited professional experience whose fight histories may not reflect consistent RHI exposure. In combat sports, the number of professional fights provides a more direct proxy for cumulative head impacts than years of participation alone, as bout frequency, rounds per fight, and intensity of exposure vary substantially across athletes.\u003c/p\u003e \u003cp\u003eMethods for the DIAGNOSE CTE have been detailed previously.\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e Briefly, this eight-year longitudinal study enrolled 240 men ages 45\u0026ndash;74, including 120 former professional football players (PRO), 60 former college football players (COL), and 60 unexposed asymptomatic men. The asymptomatic men were not included in this study. The PRO participants were required to have played 12 or more years of organized football, including at least three years in college and more than three years in the National Football League (NFL). COL participants were required to have played six or more years of American football, with at least three years at the college level.\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e,\u003cspan citationid=\"CR23\" class=\"CitationRef\"\u003e23\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cdiv id=\"Sec3\" class=\"Section2\"\u003e \u003ch2\u003eTES Adjudication\u003c/h2\u003e \u003cp\u003eTES status was determined independently for each cohort through multidisciplinary diagnostic consensus conferences using the NINDS Consensus Diagnostic Criteria for TES.\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e To be identified as TES+ required confirmation of substantial RHI exposure, evaluation of core clinical features including cognitive impairment and/or neurobehavioral dysregulation, evidence of progressive symptom worsening, and determination that symptoms were not fully explained by other neurological or psychiatric conditions. \u0026ldquo;Substantial exposure\u0026rdquo; reflects a history of RHI from activities such as contact sports or military service, typically involving prolonged participation or roles associated with frequent head impacts.\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e All participants included in the present analyses (TES\u0026thinsp;+\u0026thinsp;and TES-) met cohort-specific predefined thresholds for substantial exposure as outlined above. As such, RHI exposure did not distinguish TES+ from TES- within the analytic sample and was not used to define case status in the present analyses.\u003c/p\u003e \u003cp\u003eCognitive impairment was uniformly defined based on reported decline from prior functioning and performance at least 1.5 standard deviations below normative expectations on formal neuropsychological testing, with impairment required in episodic memory or executive function. As the neuropsychological batteries differed between cohorts and included different numbers and types of tests, cognitive impairment was operationalized using comparable criteria across studies. In DIAGNOSE CTE, participants were assigned provisional levels of certainty for CTE pathology (suggestive, possible, or probable) based on TES criteria and supportive clinical features. The same TES adjudication framework was applied in PABHS; however, provisional levels of certainty for CTE pathology were not assigned in that cohort.\u003c/p\u003e \u003c/div\u003e\n\u003ch3\u003eStudy Population\u003c/h3\u003e\n\u003cp\u003eParticipants from both cohorts with critical missing data were excluded. To maintain comparability between cohorts, participants with TES based solely on neurobehavioral dysregulation without evidence of cognitive impairment were excluded from the present analyses. Although neurobehavioral dysregulation was considered during TES adjudication in both cohorts, the methods used to ascertain these symptoms differed between studies, including the use of certain instruments, informant, or study partner reports in DIAGNOSE CTE that were not available in PABHS. The resulting analytic sample therefore included TES-positive (TES+) participants with cognitive features and TES-negative (TES-) participants across both cohorts. Participants in both cohorts provided written informed consent prior to any study procedures at their respective study sites.\u003c/p\u003e\n\u003ch3\u003eStudy Variables\u003c/h3\u003e\n\u003cp\u003eTo evaluate whether TES is characterized by convergence across clinical and biological domains, we constructed three complementary predictor indices representing cognitive function, blood-based biomarkers, and volumetric MRI regions:\u003c/p\u003e \u003cp\u003e \u003cem\u003eCognitive Index\u003c/em\u003e. Cognitive impairment is central to TES and required for possible or probable CTE diagnoses. The cognitive index was constructed to reflect TES-relevant cognitive features while accommodating cohort-specific neuropsychological batteries. Given that cognitive impairment contributes to TES adjudication, some measures included in the cognitive index overlap with those used to define TES classification; accordingly, the cognitive index represents the clinical domain of the syndrome rather than an independent predictor. Because the specific instruments differed across cohorts, cognitive variables were harmonized at the domain level (e.g., memory, processing speed, executive function) rather than by individual tests. Standardized performance metrics were used to represent each domain, and indices were derived and analyzed separately within each cohort. Comparisons between cohorts were therefore performed at the level of overall model patterns rather than pooled variable-level data.\u003c/p\u003e \u003cp\u003eFor PABHS, cognitive performance was derived from standardized computer-based assessments administered through CNS Vital Signs and the Cleveland Clinic Concussion App (C3 Logix), supplemented by a semantic verbal fluency task using Animal Naming.\u003csup\u003e\u003cspan citationid=\"CR24\" class=\"CitationRef\"\u003e24\u003c/span\u003e,\u003cspan citationid=\"CR25\" class=\"CitationRef\"\u003e25\u003c/span\u003e\u003c/sup\u003e The PABHS cognitive index encompassed seven variables: educational attainment as a marker of cognitive reserve, verbal memory, verbal fluency, processing speed, psychomotor speed, reaction time, and executive control expressed through speeded task performance.\u003c/p\u003e \u003cp\u003eFor DIAGNOSE CTE, cognitive performance was derived from a comprehensive neuropsychological battery spanning attention, psychomotor speed, executive function, learning and memory, language, and visuospatial ability.\u003csup\u003e\u003cspan citationid=\"CR26\" class=\"CitationRef\"\u003e26\u003c/span\u003e\u003c/sup\u003e Raw scores were converted to age, sex, and/or education-adjusted T-scores using established normative data.\u003csup\u003e\u003cspan citationid=\"CR13\" class=\"CitationRef\"\u003e13\u003c/span\u003e\u003c/sup\u003e For the present analyses, the DIAGNOSE CTE cognitive index incorporated six variables: educational attainment, verbal memory, verbal fluency, attention, executive function, and psychomotor or processing speed.\u003c/p\u003e \u003cp\u003eProcessing speed, psychomotor speed, and reaction time represent related but distinct components of cognitive function and motor-cognitive integration. Impairments in these activities have been reported in RHI-exposed cohorts, including in samples characterized using TES criteria.\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e\u003c/sup\u003e These domains were included to capture broader cognitive variability for analytic comparison and were not used for TES adjudication. Specifically, because cognitive impairment is a core feature of TES, the cognitive index is not independent of the outcome. Therefore, sensitivity analyses excluding the cognitive index were conducted to evaluate whether biomarker and imaging domains independently associated with TES.\u003c/p\u003e \u003cp\u003e \u003cem\u003eBlood Biomarker Index\u003c/em\u003e. Plasma biomarkers selected as predictors in this analysis included NfL, GFAP, total tau protein, and tau protein phosphorylated at threonine 231 (pTau231). These biomarkers were selected a priori based on prior literature implicating neuroaxonal injury, glial response, and tau-related processes in RHI and neurodegenerative risk. To ensure analytic consistency, only biomarkers measured in both PABHS and DIAGNOSE CTE were included in the present index. Apolipoprotein E (\u003cem\u003eAPOE\u003c/em\u003e) genotype was determined for all PABHS and DIAGNOSE CTE participants at baseline to identify the presence of the \u003cem\u003eAPOE\u003c/em\u003e-ε\u003cem\u003e4\u003c/em\u003e allele.\u003c/p\u003e \u003cp\u003eBlood biomarkers were measured using Single molecule array (Simoa) technology on an HD-X platform (Quanterix, Billerica, MA, USA) at the Clinical Neurochemistry Laboratory, Sahlgrenska University Hospital, M\u0026ouml;lndal, Sweden by laboratory technicians who were blinded to the clinical data. These samples, while analyzed on the same platform, were performed in separate batches among the cohorts. Batch-bridging using internal quality control samples ascertained comparability of the measurements. A full description of the sample collection process was previously published for both cohorts.\u003csup\u003e\u003cspan citationid=\"CR21\" class=\"CitationRef\"\u003e21\u003c/span\u003e,\u003cspan citationid=\"CR27\" class=\"CitationRef\"\u003e27\u003c/span\u003e,\u003cspan citationid=\"CR28\" class=\"CitationRef\"\u003e28\u003c/span\u003e\u003c/sup\u003e The blood biomarker index comprised a mixture of five predictors, including plasma total tau, pTau231, GFAP, NfL, and a dichotomous \u003cem\u003eAPOE\u003c/em\u003e-ε\u003cem\u003e4\u003c/em\u003e carrier status.\u003c/p\u003e \u003cp\u003e\u003cem\u003eImaging Index\u003c/em\u003e. High-resolution T1-weighted anatomical MRIs were performed at a single timepoint for DIAGNOSE CTE participants and at each annual visit for PABHS athletes. In accordance with the methodology described above, for participants with multiple visits, only the scan from the most recent visit was included in the present analyses to ensure temporal alignment across study variables. For PABHS, all imaging was acquired at a single site. DIAGNOSE CTE imaging was acquired across multiple sites using harmonized acquisition protocols.\u003csup\u003e\u003cspan citationid=\"CR22\" class=\"CitationRef\"\u003e22\u003c/span\u003e\u003c/sup\u003e The Siemens Skyra with a 32-channel head coil was used to acquire structural 3D T1-weighted magnetization-prepared rapid acquisition gradient echo images across both cohorts. Imaging data underwent quality control using FreeSurfer quality analysis tools (FreeSurfer 5.3 QATools, 2021) and visual inspection of cortical and subcortical segmentations. Reconstructions that did not meet predefined quality thresholds, including a minimum signal-to-noise ratio of 16, were reprocessed according to standardized procedures. No scans from either cohort were excluded in the current analysis based on quality control criteria.\u003c/p\u003e \u003cp\u003eFifteen regions of interest were pre-specified for the imaging index based on prior epidemiological and pathological studies.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan citationid=\"CR14\" class=\"CitationRef\"\u003e14\u003c/span\u003e,\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR29\" class=\"CitationRef\"\u003e29\u003c/span\u003e,\u003cspan citationid=\"CR30\" class=\"CitationRef\"\u003e30\u003c/span\u003e\u003c/sup\u003e These included averaged left and right subcortical gray matter volumes (thalamus, caudate, hippocampus, putamen, amygdala, pallidum), ventricular volumes (lateral ventricles, inferior lateral ventricles, third ventricle, fourth ventricle), and five anatomically distinct corpus callosum subregions (posterior, mid-posterior, central, mid-anterior, anterior). The corpus callosum was subdivided to capture regional vulnerability of interhemispheric white matter tracts to RHI. Volumetric regions were calculated using FreeSurfer\u0026rsquo;s automated full-brain segmentation process (version v.6; FreeSurfer, Boston, MA). Mean values of hemispherical volumes were calculated to create an average of left and right cerebral and ventricular regions. To ensure consistent directionality of volumetric predictors, ventricular volumes were multiplied by -1, as increased ventricular size reflects greater atrophy and should therefore correspond to lower structural integrity consistent with reductions in subcortical volumes. Imaging features were analyzed within cohort-specific models rather than pooled across cohorts.\u003c/p\u003e \u003cdiv id=\"Sec6\" class=\"Section2\"\u003e \u003ch2\u003eData Analysis\u003c/h2\u003e \u003cp\u003eInitial evaluations included descriptive statistics of the study populations and comparisons between TES groups using t-tests for continuous variables and Fisher-Freeman-Halton test for categorical variables. Because predictors were measured on different scales, all variables were converted to percentile ranks prior to analysis to place predictors on a common scale and reduce the influence of extreme values during mixture component estimation. For imaging variables, volumetric measures were adjusted for intracranial volume prior to percentile transformation to account for individual differences in head size.\u003c/p\u003e \u003cp\u003eWe performed uni-index and multi-index generalized weighted quantile sum (GWQS) regressions to examine how the cognitive, blood biomarker, and imaging predictors were associated with TES individually and simultaneously within each cohort. Indices were constructed by transforming individual variables within each domain into quantiles and combining them into a single weighted index, with weights estimated through bootstrap sampling to reflect the relative contribution of each variable to the overall association with TES. The adjusted multi-index models included age, race, competition status (active vs retired; collegiate vs professional), and RHI exposure proxy (number of professional fights or years of football participation). Sex was included as a covariate in PABHS only, as all DIAGNOSE CTE participants were male. This is presented in the equation below:\u003c/p\u003e \u003cp\u003eLogit (Y\u0026thinsp;=\u0026thinsp;1) = β\u003csub\u003e0\u003c/sub\u003e ​+ θ\u003csub\u003e1\u003c/sub\u003e​WQS\u003csub\u003e1\u003c/sub\u003e​ + θ\u003csub\u003e2\u003c/sub\u003e​WQS\u003csub\u003e2\u003c/sub\u003e​ + θ\u003csub\u003e3\u003c/sub\u003e​WQS\u003csub\u003e3\u003c/sub\u003e​ + γ\u003csub\u003e1\u003c/sub\u003e​Z\u003csub\u003e1\u003c/sub\u003e​ + γ\u003csub\u003e2\u003c/sub\u003e​Z\u003csub\u003e2\u003c/sub\u003e​ + γ\u003csub\u003e3\u003c/sub\u003e​Z\u003csub\u003e3\u003c/sub\u003e​​ + γ\u003csub\u003e4\u003c/sub\u003e​Z\u003csub\u003e4\u003c/sub\u003e​ + γ\u003csub\u003e5\u003c/sub\u003e​Z\u003csub\u003e5\u003c/sub\u003e\u003c/p\u003e \u003cp\u003ewhere Y is the binary outcome of TES (negative\u0026thinsp;=\u0026thinsp;0; positive\u0026thinsp;=\u0026thinsp;1), (WQS\u003csub\u003e1\u003c/sub\u003e, WQS\u003csub\u003e2\u003c/sub\u003e, WQS\u003csub\u003e3\u003c/sub\u003e) are the grouped cognitive, blood biomarker, and imaging indices, (θ\u003csub\u003e1\u003c/sub\u003e, θ\u003csub\u003e2\u003c/sub\u003e, θ\u003csub\u003e3\u003c/sub\u003e) are the regression coefficients representing the effect of each index, β\u003csub\u003e0\u003c/sub\u003e is the model intercept, and (γ\u003csub\u003e1\u003c/sub\u003e, γ\u003csub\u003e2\u003c/sub\u003e, γ\u003csub\u003e3\u003c/sub\u003e, γ\u003csub\u003e4\u003c/sub\u003e, γ\u003csub\u003e5\u003c/sub\u003e) are the regression slopes for the covariates (age, sex, race, RHI exposure, status). Note that the model for DIAGNOSE CTE does not include γ\u003csub\u003e5\u003c/sub\u003e​Z\u003csub\u003e5\u003c/sub\u003e​.\u003c/p\u003e \u003cp\u003eEach model used 1000 bootstrap iterations to improve the stability of the estimated mixture weights, with final weights defined as the average across bootstrap samples. The estimated coefficients of each index were obtained using logistic regression with maximum likelihood estimation. Because both multi-index models suffered from quasi-complete separation, the Firth correction was utilized to resolve this modeling issue.\u003csup\u003e\u003cspan citationid=\"CR31\" class=\"CitationRef\"\u003e31\u003c/span\u003e\u003c/sup\u003e Odds ratios (ORs) were calculated as the exponentiated regression coefficients. The contributions of individual variables within each index were assessed through empirically estimated weights. A larger weight indicated a greater contribution from each variable to the corresponding index. These weights reflect the relative contribution of correlated predictors to model discrimination within the GWQS framework and should not be interpreted as evidence of biological hierarchy, causal influence, or mechanistic pathways. To aid interpretation, a threshold value (τ), defined as the reciprocal of the number of variables within each index, was used to identify variables contributing more than would be expected under equal weighting (cognitive index τ\u0026thinsp;\u0026asymp;\u0026thinsp;0.17; blood biomarker index τ\u0026thinsp;=\u0026thinsp;0.20; imaging index τ\u0026thinsp;\u0026asymp;\u0026thinsp;0.07). To assess whether associations between biological measures and TES were independent of the cognitive index, sensitivity analyses were conducted using models including only biomarker and imaging indices. Model performance was evaluated using discrimination and classification metrics, including the area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and positive predictive value (PPV). Calibration was additionally assessed using smoothed calibration plots and the Brier score. Analyses were conducted in RStudio 2025.05.0 (RStudio, PBC, Massachusetts), with statistical significance defined as p\u0026thinsp;\u0026lt;\u0026thinsp;0.05.\u003c/p\u003e \u003c/div\u003e"},{"header":"Results","content":"\u003cp\u003eThis analysis consisted of 307 participants,\u0026nbsp;including 158 professional fighters (35% racial minority, 14% female) from the PABHS and 149 American football players from DIAGNOSE CTE (33% racial minority, 0% female). Among all participants, 51% (n=158) were identified as TES+ (PABHS=41%; DIAGNOSE CTE=63%). Across cohorts, TES+ participants had greater RHI exposure, lower educational attainment, and were more likely to be retired (PABHS) or former professional-level (DIAGNOSE CTE) athletes compared to TES- (\u003cstrong\u003eTable 1\u003c/strong\u003e). All DIAGNOSE CTE participants were \u0026ge;45 years of age (p-value=0.7010), whereas 61% of participants in PABHS were younger than 45 years (p-value=0.0010). There were no significant differences between biological sex within the TES groups for the PABHS athletes (p-value=0.8470). Sex differences were not evaluated in DIAGNOSE CTE, as all participants were male. A full description of cohort characteristics and index-level classifications are provided in \u003cstrong\u003eTable 1\u003c/strong\u003e.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e[\u003cstrong\u003eTable 1\u003c/strong\u003e]\u003c/p\u003e\n\u003cp\u003eIn PABHS, the adjusted multi-index model including\u0026nbsp;age, race, sex, competition status (active vs retired), and RHI exposure (number of professional fights) as covariates demonstrated excellent discrimination of TES (AUC=0.91), whereas discrimination remained strong but more modest in DIAGNOSE CTE (AUC=0.84). At the Youden-optimal threshold, the PABHS model achieved a sensitivity of 0.91 and specificity of 0.84 (PPV=0.80). In DIAGNOSE CTE, sensitivity was 0.84 and specificity was 0.73 (PPV=0.85). Calibration analyses demonstrated good agreement between predicted probabilities and observed TES status in both cohorts, with Brier scores of 0.11 in PABHS and 0.15 in DIAGNOSE CTE (\u003cstrong\u003eAdditional file 1: Figure A1\u003c/strong\u003e).\u003c/p\u003e\n\u003cp\u003e\u003cem\u003ePABHS\u0026nbsp;\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eIn the PABHS cohort, each index was significantly associated with TES in uni-index models (Cognitive AUC=0.85/PPV=0.71; Biomarker AUC=0.62/PPV=0.69; Imaging AUC=0.81/PPV=0.78). The AUC and PPV values for the unadjusted multi-index model, however, performed better than the uni-index models. In the unadjusted multi-index model, higher values (improved scores) of the cognitive index were associated with 75% decreased odds of TES (OR=0.25; 95% confidence interval [CI]=0.16, 0.37; p-value\u0026lt;.0001) and higher values of the imaging index were associated with 70% lower odds of TES (OR=0.30; 95% CI=0.20, 0.43; p-value\u0026lt;.0001). The blood biomarker index did not significantly influence TES\u0026nbsp;status\u0026nbsp;in the unadjusted model.\u003c/p\u003e\n\u003cp\u003eThe adjusted multi-index model indicated the best fit for this dataset (AUC=0.91/PPV=0.80), as it demonstrated improved discrimination compared with individual indices. In the adjusted model, higher values of the cognitive index remained strongly associated with lower odds of TES (OR=0.25; 95% CI=0.14, 0.42; p-value\u0026lt;.0001) while the imaging index independently contributed additional discriminatory information, with a 65% lower odds of TES per unit increase (OR=0.35; 95% CI=0.17, 0.65; p-value=0.0015). Female sex (OR=9.36; 95% CI=2.44, 40.10; p-value=0.0014), older age (OR=4.01; 95% CI=1.41, 11.94; p-value=0.0101), and greater RHI exposure (OR=1.05; 95% CI=1.02, 1.08; p-value=0.0016) all significantly predicted the odds of TES (\u003cstrong\u003eTable 2\u003c/strong\u003e). This estimate should be interpreted cautiously given the small percentage of female participants. Similar to the unadjusted model, the blood biomarker index did not significantly influence TES\u0026nbsp;status in the adjusted model.\u003c/p\u003e\n\u003cp\u003eThe estimated weights for individual predictor variables in the adjusted multi-index model are shown in \u003cstrong\u003eFigure 1\u003c/strong\u003e.\u0026nbsp;In the cognitive index, the following predictors had a weight greater than \u0026tau;=0.14: symbol digit coding (0.27), processing speed (0.26), and psychomotor speed (0.24). In the blood biomarker index, the following predictors had a weight greater than \u0026tau;=0.20: NfL (0.29), \u003cem\u003eAPOE\u003c/em\u003e-\u0026epsilon;4 allele (0.22), and GFAP (0.20). In the imaging index, the following predictors had a weight greater than \u0026tau;=0.07: volumes of 3\u003csup\u003erd\u003c/sup\u003e ventricle (0.29), posterior corpus callosum (0.24), amygdala (0.16), and putamen (0.09).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e[\u003cstrong\u003eTable 2\u003c/strong\u003e]\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eDIAGNOSE CTE\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eIn the DIAGNOSE CTE cohort, each index demonstrated significant association with TES in uni-index models (Cognitive AUC=0.78/PPV=0.84; Biomarker AUC=0.62/PPV=0.76; Imaging AUC=0.62/PPV=0.74). The\u0026nbsp;AUC and PPV values were improved in the unadjusted multi-index model. In the unadjusted multi-index model, higher values of the cognitive index were associated with 83% lower odds of TES\u0026nbsp;(OR=0.17; 95% CI=0.08, 0.31; p-value\u0026lt;.0001) while higher values of the blood biomarker index were associated with a 107% increase in the odds of TES (OR=2.07; 95% CI=1.32, 3.38; p-value=0.0023). The imaging index did not significantly influence TES\u0026nbsp;status\u0026nbsp;in the unadjusted model.\u003c/p\u003e\n\u003cp\u003eThe adjusted multi-index model including\u0026nbsp;age, race, competition status (college vs professional), and RHI exposure (years of football) as covariates indicated the best fit for this dataset (AUC=0.84; PPV=0.84). Similar to PABHS findings, analysis of the data supported our hypothesis that TES is best characterized by a multidomain profile, as integration of cognitive, biomarker, and imaging indices improved model discrimination beyond any single domain alone. In the adjusted model, the cognitive index remained strongly associated with lower odds of TES (OR=0.15; 95% CI=0.07, 0.30; p-value\u0026lt;.0001) while the blood biomarker index independently contributed additional discriminatory information, with a 125% increase in the odds of TES per unit increase (OR=2.25; 95% CI=1.39, 3.83; p-value=0.0015) (\u003cstrong\u003eTable 2\u003c/strong\u003e). RHI exposure (OR=1.22; 95% CI=1.06, 1.43; p-value=0.0081) significantly predicted the odds of TES, while competition status and age did not. Similar to the unadjusted model, the imaging index did not significantly influence TES status in the adjusted model.\u003c/p\u003e\n\u003cp\u003eIn the cognitive index, the following\u0026nbsp;predictors had a weight greater than \u0026tau;=0.17: executive speed (0.37), verbal memory (0.29), and educational attainment (0.19). In the blood biomarker index, the following predictors had a weight greater than \u0026tau;=0.20: \u003cem\u003eAPOE\u003c/em\u003e-\u0026epsilon;4 allele (0.41) and total tau (0.38). In the imaging index, the following predictors had a weight greater than \u0026tau;=0.07: caudate (0.17), 3\u003csup\u003erd\u003c/sup\u003e ventricle (0.09), lateral ventricle (0.08), 4\u003csup\u003eth\u003c/sup\u003e ventricle (0.07), central corpus callosum (0.07), and amygdala (0.07). Summary statistics and visualization of the weighted cognitive, blood biomarker, and imaging indices by TES status are provided in Appendix (\u003cstrong\u003eFigure A2\u003c/strong\u003e and \u003cstrong\u003eTable A3)\u003c/strong\u003e, demonstrating distinct separation in index distributions between TES+ and TES- participants across both cohorts.\u003c/p\u003e\n\u003cp\u003e\u003cem\u003eSensitivity Analysis: Biomarker and Imaging Indices Without Cognitive Index\u003c/em\u003e\u003c/p\u003e\n\u003cp\u003eTo evaluate whether associations between biological measures and TES were independent of the cognitive index, a sensitivity analysis, visualized in Table 3, was conducted including only biomarker and imaging indices.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003e[Table 3]\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eIn PABHS, the unadjusted model demonstrated a significant association for the imaging index (OR=0.28; 95% CI=0.18, 0.42; p-value\u0026lt;.0001), whereas the blood biomarker index was not associated with TES status. After adjustment, both indices were significantly associated with TES (biomarker index OR=2.06; 95% CI=1.06, 4.28; p-value=0.0455; imaging index OR=0.18, 95% CI=0.09, 0.35; p-value\u0026lt;.0001), with strong model discrimination (AUC=0.88; PPV=0.72).\u003c/p\u003e\n\u003cp\u003eIn DIAGNOSE CTE, both indices were associated with TES in the unadjusted model (biomarker index OR=1.60; 95% CI=1.08, 2.42, p-value=0.0204; imaging index OR=0.60; 95% CI=0.37, 0.94; p-value=0.0277). These associations remained significant in adjusted models (biomarker index OR=1.65; 95% CI=1.01, 2.74; p-value=0.0486; imaging index OR=0.48; 95% CI=0.26, 0.84; p-value=0.0133), with modest model discrimination (AUC=0.71; PPV=0.78). Overall, results were consistent with the primary analyses. Although the adjusted multidomain model incorporating cognitive, biomarker, and imaging indices demonstrated the strongest discrimination of TES, biomarker and imaging indices remained significantly associated with TES when the cognitive index was excluded from the model.\u003c/p\u003e\n\u003cp\u003e[\u003cstrong\u003eFigure 1\u003c/strong\u003e]\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eBy integrating cognitive, imaging, and blood biomarker data from two independent cohorts of athletes exposed to RHI, this study demonstrates consistent associations between TES and combined cognitive, biomarker, and neuroimaging indices across two independent cohorts. Across both cohorts, multi-index models demonstrated improved model performance compared with single-domain models, with adjusted AUC values of 0.91 in PABHS and 0.84 in DIAGNOSE CTE. Even when the cognitive index was excluded, blood biomarker and imaging indices remained associated with TES across both cohorts, indicating that objective biological measures contribute substantial information beyond clinical features alone.\u003c/p\u003e \u003cp\u003eAlthough the overall pattern of findings was consistent across cohorts, the relative contribution of biological domains differed. Structural imaging features contributed more prominently to model discrimination among professional fighters (PABHS), whereas blood biomarker signals were more strongly associated with TES among former football players (DIAGNOSE CTE). These differences may reflect variation in cumulative injury burden, age distribution, or \u0026ndash; in cases where CTE is present \u0026ndash; stage of neurodegenerative progression between cohorts. Together, these findings support the concept that TES reflects a broader clinicobiological phenotype rather than an exclusively symptom-defined construct. Likewise, the reproducibility of the outcomes in two groups of athletes with unique RHI exposures increases the credibility and applicability of the results. The GWQS framework complements other multivariate approaches such as the Global Statistical Test (GST) by empirically weighting correlated predictors to quantify their relative contributions to the overall association. While GST provides a robust global measure of group-level differences, GWQS adds variable-level interpretability, allowing clearer identification of the factors most strongly linked to diagnostic status.\u003csup\u003e\u003cspan citationid=\"CR19\" class=\"CitationRef\"\u003e19\u003c/span\u003e\u003c/sup\u003e This approach is promising for risk stratification, cohort enrichment, and experimental therapeutic efforts in populations exposed to RHI.\u003c/p\u003e \u003cp\u003eOur results demonstrate the value of utilizing an integrated approach over individual measures for identifying individuals with TES who might eventually be shown to have CTE. We hypothesize that athletes with TES that incorporates fluid biomarker and imaging changes are more likely to have CTE than those without these concomitant indicators of brain pathology.\u003c/p\u003e \u003cp\u003eOur findings reinforce previous literature suggesting that cognitive deficits associated with RHI exposure are diverse while also identifying cognitive features common to both cohorts.\u003csup\u003e\u003cspan citationid=\"CR6\" class=\"CitationRef\"\u003e6\u003c/span\u003e,\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e,\u003cspan citationid=\"CR32\" class=\"CitationRef\"\u003e32\u003c/span\u003e,\u003cspan citationid=\"CR33\" class=\"CitationRef\"\u003e33\u003c/span\u003e\u003c/sup\u003e In both cohorts\u0026rsquo; cognitive indices, processing speed exceeded the τ threshold, indicating a prominent role in differentiating TES\u0026thinsp;+\u0026thinsp;and TES- status among both groups. These results are consistent with prior studies that have identified a decline in processing speed as associated with neurodegeneration related to RHI exposure.\u003csup\u003e\u003cspan citationid=\"CR17\" class=\"CitationRef\"\u003e17\u003c/span\u003e\u003c/sup\u003e Verbal memory was more heavily weighted in the retired football players than the fighters. This may reflect the more precise testing and quantification of neurocognitive assessments in the DIAGNOSE CTE cohort. The inverse association between educational attainment and TES supports cognitive reserve theory, in which higher education serves as a protective factor against clinical symptom expression despite potential underlying pathology.\u003csup\u003e\u003cspan citationid=\"CR34\" class=\"CitationRef\"\u003e34\u003c/span\u003e\u003c/sup\u003e\u003c/p\u003e \u003cp\u003eWhile GWQS-derived weights do not imply causality or biological mechanism, the relative prominence of specific biomarkers across cohorts provides a descriptive framework for contextualizing existing biological literature. The biomarker indices revealed \u003cem\u003eAPOE\u003c/em\u003e-ε4 allele received the highest relative weight within the blood biomarker index in both cohorts, along with NfL and GFAP in fighters, whereas former football players showed elevated total tau. These findings are consistent with prior literature identifying these markers as associated with astroglial activation, axonal damage, and neurodegeneration in RHI-exposed populations.\u003csup\u003e\u003cspan citationid=\"CR1\" class=\"CitationRef\"\u003e1\u003c/span\u003e,\u003cspan additionalcitationids=\"CR36\" citationid=\"CR35\" class=\"CitationRef\"\u003e35\u003c/span\u003e\u0026ndash;\u003cspan citationid=\"CR37\" class=\"CitationRef\"\u003e37\u003c/span\u003e\u003c/sup\u003e Many of the PABHS participants are under the age of 45, which may contribute to lower tau burden than the DIAGNOSE CTE participants. Furthermore, the observed differences between cohorts may reflect not only methodological and demographic factors, but also the distinct biomechanical forces experienced in combat sports versus football. Impacts in combat sports often involve rotational acceleration and focal blows, potentially producing different injury patterns or affecting distinct brain regions compared to the linear and repetitive collisions common in football. While variability in imaging indices may partly result from differences in scanner platforms and cohort age, the potential influence of sport-specific exposure mechanics warrants further investigation.\u003c/p\u003e \u003cp\u003eThe consistent prominence of the \u003cem\u003eAPOE\u003c/em\u003e-ε\u003cem\u003e4\u003c/em\u003e carrier across both cohorts underscores its potential role in the identification of TES risk. Present in approximately 60\u0026ndash;70% of individuals with AD, \u003cem\u003eAPOE\u003c/em\u003e-ε\u003cem\u003e4\u003c/em\u003e is well established as a genetic risk factor that promotes amyloid deposition and accelerates tau-related neurodegeneration.\u003csup\u003e\u003cspan citationid=\"CR38\" class=\"CitationRef\"\u003e38\u003c/span\u003e\u003c/sup\u003e \u003cem\u003eAPOE\u003c/em\u003e-ε\u003cem\u003e4\u003c/em\u003e has also been linked to heightened inflammatory responses, including increased activation of astrocytes and microglia through impaired triglyceride metabolism.\u003csup\u003e\u003cspan citationid=\"CR39\" class=\"CitationRef\"\u003e39\u003c/span\u003e\u003c/sup\u003e These mechanistic pathways align with our broader findings, in which \u003cem\u003eAPOE\u003c/em\u003e-ε\u003cem\u003e4\u003c/em\u003e clustered with elevated GFAP may be indicative of a neuroinflammatory response. Recent neuropathological evidence has demonstrated that \u003cem\u003eAPOE\u003c/em\u003e-ε\u003cem\u003e4\u003c/em\u003e is associated with greater severity of CTE pathology, supporting a role for this allele in modifying disease progression following RHI exposure.\u003csup\u003e\u003cspan citationid=\"CR40\" class=\"CitationRef\"\u003e40\u003c/span\u003e\u003c/sup\u003e In the context of RHI, \u003cem\u003eAPOE\u003c/em\u003e-ε\u003cem\u003e4\u003c/em\u003e may amplify vulnerability to both proteinopathies and inflammation, leading to degeneration and subsequent clinical decline.\u003c/p\u003e \u003cp\u003eThe current diagnostic criteria for TES depend solely on clinical domains.\u003csup\u003e\u003cspan citationid=\"CR2\" class=\"CitationRef\"\u003e2\u003c/span\u003e\u003c/sup\u003e These criteria rely on symptom reporting and neuropsychological performance, which may reflect heterogeneous processes, may not capture early biological changes, and may overlap with other neurological or psychiatric conditions. In addition, the clinical data required for TES adjudication often involve comprehensive neuropsychological evaluation and longitudinal clinical assessment, which can be resource-intensive and may limit scalability. Integrating biologic data with clinical observations represents one approach to resolving this dilemma. In contrast, the biological measures included in the multi-index model, such as blood biomarkers and structural MRI, can be obtained using standardized protocols and may offer more scalable and objective approaches to monitoring brain health. Specifically, the multi-index model developed here can serve as a stratification tool for identifying high-risk individuals or as an inclusion criterion for clinical trials investigating disease targeted therapies. The improved positive predictive values (PPV\u0026thinsp;=\u0026thinsp;80 and 85%, respectively) underscore the potential to reduce heterogeneity and improve cohort stratification for such trials.\u003c/p\u003e \u003cp\u003eThis study had several strengths, including a large sample size (N\u0026thinsp;=\u0026thinsp;307) and reproducibility of findings from two cohorts of professional athletes with substantial RHI exposure. Several limitations are acknowledged. Although the TES criteria applied here are currently the best available, the lack of pathological confirmation restricts our ability to validate and directly link index performance to underlying CTE pathology. The statistical modeling in this study, while valuable for identifying individual and cumulative weights, limits inferences about causality or the progression of the syndrome. The DIAGNOSE CTE cohort was restricted to individuals aged 45 years and older, whereas the PABHS cohort included many younger participants. Age differences influence both biomarker expression and neurodegenerative burden. Furthermore, some of the cognitive measures applied normative values derived from general population samples that differ from the demographic and educational profiles of the PABHS cohort. As the PABHS participants generally have lower educational attainment than the DIAGNOSE CTE participants, applying standard normative corrections may have led to misidentification or overestimation of impairment. The absence of behavioral information in the PABHS cohort may have influenced the diagnosis and characteristics of TES identified in this cohort. The p-tau measure available was p-tau231 and other p-tau species may be more informative. Our findings should be interpreted as descriptive and hypothesis-generating with respect to TES identification. Future studies should evaluate the independent contribution of neurobehavioral dysregulation to TES in cohorts where these symptoms are systematically measured. Such analyses may help clarify whether neurobehavioral features contribute meaningfully to multidomain TES profiles or represent a distinct clinical dimension of RHI exposure.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eTES in RHI-exposed athletes was associated with a convergent clinicobiological profile across two independent cohorts. In both cohorts, multidomain models integrating cognitive, biomarker, and neuroimaging indices demonstrated stronger discrimination of TES than single-domain models. Biomarker and imaging indices contributed additional information across cohorts, although imaging contributions were more prominent in fighters while biomarker associations were stronger in football players. These findings support multidomain analytic frameworks for examining biological signals associated with TES and may inform future studies aimed at improving risk stratification for CTE.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003cstrong\u003eEthics approval and consent to participate\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe current study represents a secondary analysis of a de-identified database with no contact with participants; therefore, it was submitted to and approved by the University of Nevada, Las Vegas Institutional Review Board for administrative review of exempt research (UNLV#2024-245). This study was conducted in accordance with applicable institutional policies and federal regulations, including 45 CFR 46 and the HIPAA Privacy Rule, and adhered to the principles of the Declaration of Helsinki.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConsent for publication\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eNot applicable\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAvailability of data and materials\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eThe datasets used and/or analyzed during the current study are available from the corresponding author upon reasonable request.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eCompeting interests\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eJLC has provided consultation to Acadia, Acumen, ALZpath, AnnovisBio, Artery, Axsome, Biogen, Bristol-Myers Squibb, Eisai, Fosun, GAP Foundation, Hummingbird Diagnostics, IGC, Janssen, Julius Clinical, Kinoxis, Lilly, LSP/EQT, Merck, MoCA Cognition, Novo Nordisk, NSC Therapeutics, Otsuka, ReMYND, Roche, Scottish Brain Sciences, Signant Health, Simcere, Sinaptica, and T-Neuro pharmaceutical, assessment, and investment companies.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eHZ has served at scientific advisory boards and/or as a consultant for Abbvie, Acumen, Alamar, Alector, Alzinova, ALZpath, Amylyx, Annexon, Apellis, Artery Therapeutics, AZTherapies, Cognito Therapeutics, CogRx, Denali, Eisai, Enigma, LabCorp, Merck Sharp \u0026amp; Dohme, Merry Life, Nervgen, Novo Nordisk, Optoceutics, Passage Bio, Pinteon Therapeutics, Prothena, Quanterix, Red Abbey Labs, reMYND, Roche, Samumed, ScandiBio Therapeutics AB, Siemens Healthineers, Triplet Therapeutics, and Wave, has given lectures sponsored by Alzecure, BioArctic, Biogen, Cellectricon, Fujirebio, LabCorp, Lilly, Novo Nordisk, Oy Medix Biochemica AB, Roche, and WebMD, is a co-founder of Brain Biomarker Solutions in Gothenburg AB (BBS), which is a part of the GU Ventures Incubator Program, and is a shareholder of CERimmune Therapeutics (outside submitted work).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe other authors report no competing interests.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eFunding\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eCB has received research funding from the UFC, Top Rank Promotions, Haymon Boxing, and Bellator Promotions. MLA reports honoraria from the Michael J. Fox Foundation for services unrelated to this work, royalties from Oxford University Press Inc., and research support from Life Molecular Imaging Inc. JLC is supported by NIGMS grant P20GM109025, NIA R35AG71476, NIA R25AG083721, and NINDS R01NS139383, as well as funding from the Alzheimer\u0026rsquo;s Disease Drug Discovery Foundation, the Ted and Maria Quirk Endowment, and the Joy Chambers-Grundy Endowment. DIAGNOSE CTE was supported by NINDS U01NS093334. PABHS is supported by philanthropic grants from the Ultimate Fighting Championship, Top Rank Boxing, and Haymon Boxing.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eHZ is a Wallenberg Scholar and a Distinguished Professor at the Swedish Research Council supported by grants from the Swedish Research Council (#2023-00356, #2022-01018 and #2019-02397), the European Union\u0026rsquo;s Horizon Europe research and innovation programme under grant agreement No 101053962, and Swedish State Support for Clinical Research (#ALFGBG-71320).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003eThe remaining authors report no relevant funding.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAuthors\u0026rsquo; contributions\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eBCK conceived and designed the study, performed the statistical analyses with LC, and drafted the manuscript. LC contributed to study design, performed the statistical analyses, and contributed to interpretation of the results. DS, MLA, JVW, YT, CHA, MES, OP, DIK, EP, LJB, IKK, JM, EMR, RCC, RAS, and HZ contributed to data acquisition, interpretation of results, and critical revision of the manuscript for important intellectual content. CB and JLC provided senior oversight, contributed to study design and interpretation, and critically revised the manuscript. All authors reviewed and approved the final manuscript.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eAcknowledgements\u003c/strong\u003e\u003c/p\u003e\n\u003cp\u003eWe thank the PABHS and DIAGNOSE CTE research teams at the Lou Ruvo Center for Brain Health and Boston University for their contributions to data collection and study coordination. We also thank the athletes who participated in these studies for their time and commitment to advancing understanding of the long-term effects of repetitive head impacts.\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eMcKee AC, Stein TD, Huber BR, et al. Chronic traumatic encephalopathy (CTE): criteria for neuropathological diagnosis and relationship to repetitive head impacts. \u003cem\u003eActa Neuropathol\u003c/em\u003e. Apr 2023;145(4):371-394. doi:10.1007/s00401-023-02540-w\u003c/li\u003e\n\u003cli\u003eKatz DI, Bernick C, Dodick DW, et al. National Institute of Neurological Disorders and Stroke Consensus Diagnostic Criteria for Traumatic Encephalopathy Syndrome. \u003cem\u003eNeurology\u003c/em\u003e. May 4 2021;96(18):848-863. doi:10.1212/WNL.0000000000011850\u003c/li\u003e\n\u003cli\u003eConway Kleven BD, Chien LC, Cross CL, Labus B, Bernick C. Traumatic Encephalopathy Syndrome: Head Impact Exposure and Blood Biomarkers in Professional Combat Athletes. \u003cem\u003eJ Head Trauma Rehabil\u003c/em\u003e. Feb 25 2025;doi:10.1097/htr.0000000000001048\u003c/li\u003e\n\u003cli\u003eAshton NJ, Hye A, Rajkumar AP, et al. An update on blood-based biomarkers for non-Alzheimer neurodegenerative disorders. \u003cem\u003eNature Reviews Neurology\u003c/em\u003e. 2020/05/01 2020;16(5):265-284. doi:10.1038/s41582-020-0348-0\u003c/li\u003e\n\u003cli\u003eAsken BM, Tanner JA, VandeVrede L, et al. Multi-Modal Biomarkers of Repetitive Head Impacts and Traumatic Encephalopathy Syndrome: A Clinicopathological Case Series. \u003cem\u003eJ Neurotrauma\u003c/em\u003e. Sep 2022;39(17-18):1195-1213. doi:10.1089/neu.2022.0060\u003c/li\u003e\n\u003cli\u003eRitter A, Shan G, Montes A, Randall R, Bernick C. Traumatic encephalopathy syndrome: application of new criteria to a cohort exposed to repetitive head impacts. \u003cem\u003eBritish Journal of Sports Medicine\u003c/em\u003e. 2023;57:389-394. \u003c/li\u003e\n\u003cli\u003eAsken BM, DeKosky ST, Clugston JR, Jaffee MS, Bauer RM. Diffusion tensor imaging (DTI) findings in adult civilian, military, and sport-related mild traumatic brain injury (mTBI): a systematic critical review. \u003cem\u003eBrain imaging and behavior\u003c/em\u003e. 2018;12(2):585-612. \u003c/li\u003e\n\u003cli\u003eAsken BM, Rabinovici GD. Identifying degenerative effects of repetitive head trauma with neuroimaging: a clinically-oriented review. \u003cem\u003eActa neuropathologica communications\u003c/em\u003e. 2021;9(1):96. \u003c/li\u003e\n\u003cli\u003eKinnunen KM, Cash DM, Poole T, et al. Presymptomatic atrophy in autosomal dominant Alzheimer\u0026apos;s disease: A serial magnetic resonance imaging study. \u003cem\u003eAlzheimers Dement\u003c/em\u003e. Jan 2018;14(1):43-53. doi:10.1016/j.jalz.2017.06.2268\u003c/li\u003e\n\u003cli\u003eRohrer JD, Nicholas JM, Cash DM, et al. Presymptomatic cognitive and neuroanatomical changes in genetic frontotemporal dementia in the Genetic Frontotemporal dementia Initiative (GENFI) study: a cross-sectional analysis. \u003cem\u003eLancet Neurol\u003c/em\u003e. Mar 2015;14(3):253-62. doi:10.1016/s1474-4422(14)70324-2\u003c/li\u003e\n\u003cli\u003eBrewer JB, Magda S, Airriess C, Smith ME. Fully-automated quantification of regional brain volumes for improved detection of focal atrophy in Alzheimer disease. \u003cem\u003eAmerican Journal of Neuroradiology\u003c/em\u003e. 2009;30(3):578-580. \u003c/li\u003e\n\u003cli\u003eLin Y, Fu Y, Zeng Y-F, et al. Six visual rating scales as a biomarker for monitoring atrophied brain volume in Parkinson\u0026rsquo;s disease. \u003cem\u003eAging and disease\u003c/em\u003e. 2020;11(5):1082. \u003c/li\u003e\n\u003cli\u003eAlosco ML, Barr WB, Banks SJ, et al. Neuropsychological test performance of former American football players. \u003cem\u003eAlzheimers Res Ther\u003c/em\u003e. Jan 3 2023;15(1):1. doi:10.1186/s13195-022-01147-9\u003c/li\u003e\n\u003cli\u003eBernick C, Banks SJ, Shin W, et al. Repeated head trauma is associated with smaller thalamic volumes and slower processing speed: the Professional Fighters\u0026apos; Brain Health Study. \u003cem\u003eBr J Sports Med\u003c/em\u003e. Aug 2015;49(15):1007-11. doi:10.1136/bjsports-2014-093877\u003c/li\u003e\n\u003cli\u003eBernick C, Shan G, Ritter A, et al. Blood biomarkers and neurodegeneration in individuals exposed to repetitive head impacts. \u003cem\u003eAlzheimers Res Ther\u003c/em\u003e. Oct 12 2023;15(1):173. doi:10.1186/s13195-023-01310-w\u003c/li\u003e\n\u003cli\u003eConway Kleven B, Chien LC, Young DL, Cross CL, Labus B, Bernick C. Repetitive head impacts among professional fighters: a pilot study evaluating Traumatic Encephalopathy Syndrome and postural balance. \u003cem\u003ePhys Sportsmed\u003c/em\u003e. Mar 6 2024:1-7. doi:10.1080/00913847.2024.2325331\u003c/li\u003e\n\u003cli\u003eConway Kleven BD, Chien L-C, Labus B, et al. Longitudinal changes in regional brain volumes and cognition of professional fighters with traumatic encephalopathy syndrome. \u003cem\u003eNeurology\u003c/em\u003e. 2023;101(11):e1118-e1126. \u003c/li\u003e\n\u003cli\u003eKleven BDC, Chien L-C, Cross CL, Labus B, Bernick C. Traumatic Encephalopathy Syndrome: Head Impact Exposure and Blood Biomarkers in Professional Combat Athletes. \u003cem\u003eThe Journal of Head Trauma Rehabilitation\u003c/em\u003e. 2025:10.1097/HTR.0000000000001048. doi:10.1097/htr.0000000000001048\u003c/li\u003e\n\u003cli\u003eWheeler DC, Rustom S, Carli M, Whitehead TP, Ward MH, Metayer C. Assessment of Grouped Weighted Quantile Sum Regression for Modeling Chemical Mixtures and Cancer Risk. \u003cem\u003eInternational Journal of Environmental Research and Public Health\u003c/em\u003e. 2021;18(2):504. \u003c/li\u003e\n\u003cli\u003eBernick C, Banks S, Phillips M, et al. Professional fighters brain health study: rationale and methods. \u003cem\u003eAm J Epidemiol\u003c/em\u003e. Jul 15 2013;178(2):280-6. doi:10.1093/aje/kws456\u003c/li\u003e\n\u003cli\u003eAlosco ML, Mariani ML, Adler CH, et al. Developing methods to detect and diagnose chronic traumatic encephalopathy during life: rationale, design, and methodology for the DIAGNOSE CTE Research Project. \u003cem\u003eAlzheimer\u0026apos;s research \u0026amp; therapy\u003c/em\u003e. 2021;13(1):136-136. doi:10.1186/s13195-021-00872-x\u003c/li\u003e\n\u003cli\u003eAlosco ML, Mariani ML, Adler CH, et al. Developing methods to detect and diagnose chronic traumatic encephalopathy during life: rationale, design, and methodology for the DIAGNOSE CTE Research Project. \u003cem\u003eAlzheimers Res Ther\u003c/em\u003e. Aug 12 2021;13(1):136. doi:10.1186/s13195-021-00872-x\u003c/li\u003e\n\u003cli\u003eStern RA, Trujillo-Rodriguez D, Tripodis Y, et al. Amyloid PET across the cognitive spectrum in former professional and college American football players: findings from the DIAGNOSE CTE Research Project. \u003cem\u003eAlzheimers Res Ther\u003c/em\u003e. Oct 5 2023;15(1):166. doi:10.1186/s13195-023-01315-5\u003c/li\u003e\n\u003cli\u003eGualtieri CT, Johnson LG. Reliability and validity of a computerized neurocognitive test battery, CNS Vital Signs. \u003cem\u003eArchives of Clinical Neuropsychology\u003c/em\u003e. 2006;21(7):623-643. doi:10.1016/j.acn.2006.05.007\u003c/li\u003e\n\u003cli\u003eSimon M, Maerlender A, Metzger K, Decoster L, Hollingworth A, Valovich McLeod T. Reliability and concurrent validity of select C3 logix test components. \u003cem\u003eDevelopmental neuropsychology\u003c/em\u003e. 2017;42(7-8):446-459. \u003c/li\u003e\n\u003cli\u003eAlosco ML, Mariani ML, Adler CH, et al. Developing methods to detect and diagnose chronic traumatic encephalopathy during life: rationale, design, and methodology for the DIAGNOSE CTE Research Project. \u003cem\u003eAlzheimer\u0026apos;s research \u0026amp; therapy\u003c/em\u003e. 2021;13(1):136. \u003c/li\u003e\n\u003cli\u003eBernick C, Zetterberg H, Shan G, Banks S, Blennow K. Longitudinal Performance of Plasma Neurofilament Light and Tau in Professional Fighters: The Professional Fighters Brain Health Study. \u003cem\u003eJ Neurotrauma\u003c/em\u003e. Oct 15 2018;35(20):2351-2356. doi:10.1089/neu.2017.5553\u003c/li\u003e\n\u003cli\u003eBernick C, Shan G, Ritter A, et al. Blood biomarkers and neurodegeneration in individuals exposed to repetitive head impacts. \u003cem\u003eAlzheimer\u0026apos;s Research \u0026amp; Therapy\u003c/em\u003e. 2023;15(1):173. \u003c/li\u003e\n\u003cli\u003eBernick C, Shan G, Zetterberg H, et al. Longitudinal change in regional brain volumes with exposure to repetitive head impacts. \u003cem\u003eNeurology\u003c/em\u003e. Jan 21 2020;94(3):e232-e240. doi:10.1212/WNL.0000000000008817\u003c/li\u003e\n\u003cli\u003eArciniega H, Baucom ZH, Tuz-Zahra F, et al. Brain morphometry in former American football players: Findings from the DIAGNOSE CTE research project. \u003cem\u003eBrain\u003c/em\u003e. 2024;147(10):3596-3610. \u003c/li\u003e\n\u003cli\u003eD\u0026apos;Angelo G, Ran D. Tutorial on Firth\u0026apos;s Logistic Regression Models for Biomarkers in Preclinical Space. \u003cem\u003ePharmaceutical Statistics\u003c/em\u003e. 2025;\u003c/li\u003e\n\u003cli\u003eBernick C, Shan G, Bennett L, Alberts J, Cummings J. Assessing Clinical Change in Individuals Exposed to Repetitive Head Impacts: The Repetitive Head Impact Composite Index. \u003cem\u003eFront Neurol\u003c/em\u003e. 2021;12:605318. doi:10.3389/fneur.2021.605318\u003c/li\u003e\n\u003cli\u003eLoBue C, Bickart KC, Bieniek KF, et al. Cognitive Impairment and Dementia-Related Diagnoses in Chronic Traumatic Encephalopathy: A Systematic Review and Meta-Analysis. \u003cem\u003eNeurology\u003c/em\u003e. Aug 26 2025;105(4):e213934. doi:10.1212/wnl.0000000000213934\u003c/li\u003e\n\u003cli\u003eBanks SJ, Obuchowski N, Shin W, et al. The protective effect of education on cognition in professional fighters. \u003cem\u003eArch Clin Neuropsychol\u003c/em\u003e. Feb 2014;29(1):54-9. doi:10.1093/arclin/act079\u003c/li\u003e\n\u003cli\u003eAsken BM, Yang Z, Xu H, et al. Acute effects of sport-related concussion on serum glial fibrillary acidic protein, ubiquitin C-terminal hydrolase L1, total tau, and neurofilament light measured by a multiplex assay. \u003cem\u003eJournal of neurotrauma\u003c/em\u003e. 2020;37(13):1537-1545. \u003c/li\u003e\n\u003cli\u003eGao W, Zhang Z, Lv X, et al. Neurofilament light chain level in traumatic brain injury: a system review and meta-analysis. \u003cem\u003eMedicine\u003c/em\u003e. 2020;99(38):e22363. \u003c/li\u003e\n\u003cli\u003eMastandrea P, Mengozzi S, Bernardini S. Systematic review and meta-analysis of observational studies evaluating glial fibrillary acidic protein (GFAP) and ubiquitin C-terminal hydrolase L1 (UCHL1) as blood biomarkers of mild acute traumatic brain injury (mTBI) or sport-related concussion (SRC) in adult subjects. \u003cem\u003eDiagnosis (Berl)\u003c/em\u003e. Aug 20 2024;doi:10.1515/dx-2024-0078\u003c/li\u003e\n\u003cli\u003eGarcia MJ, Leadley R, Ross J, et al. Prognostic and Predictive Factors in Early Alzheimer\u0026apos;s Disease: A Systematic Review. \u003cem\u003eJ Alzheimers Dis Rep\u003c/em\u003e. 2024;8(1):203-240. doi:10.3233/adr-230045\u003c/li\u003e\n\u003cli\u003eStephenson RA, Sepulveda J, Johnson KR, et al. Triglyceride metabolism controls inflammation and microglial phenotypes associated with \u003cem\u003eAPOE\u003c/em\u003e4. \u003cem\u003eCell Rep\u003c/em\u003e. Jul 22 2025;44(7):115961. doi:10.1016/j.celrep.2025.115961\u003c/li\u003e\n\u003cli\u003eAtherton K, Han X, Chung J, et al. Association of \u003cem\u003eAPOE\u003c/em\u003e Genotypes and Chronic Traumatic Encephalopathy. \u003cem\u003eJAMA Neurology\u003c/em\u003e. 2022;79(8):787-796. doi:10.1001/jamaneurol.2022.1634\u003c/li\u003e\n\u003cli\u003eJia J, Ning Y, Chen M, et al. Biomarker changes during 20 years preceding Alzheimer\u0026rsquo;s disease. \u003cem\u003eNew England Journal of Medicine\u003c/em\u003e. 2024;390(8):712-722. \u003c/li\u003e\n\u003cli\u003eThakur R, Saini AK, Taliyan R, Chaturvedi N. Neurodegenerative diseases early detection and monitoring system for point-of-care applications. \u003cem\u003eMicrochemical Journal\u003c/em\u003e. 2025;208:112280. \u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTables 1 to 3 are available in the Supplementary Files section.\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":false,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"alzheimers-research-and-therapy","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"azrt","sideBox":"Learn more about [Alzheimer's Research and Therapy](http://alzres.biomedcentral.com/)","snPcode":"13195","submissionUrl":"https://submission.nature.com/new-submission/13195/3","title":"Alzheimer's Research \u0026 Therapy","twitterHandle":"@AlzheimersRes","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true},"keywords":"Chronic traumatic encephalopathy, Traumatic encephalopathy syndrome, repetitive head impacts, blood biomarkers, neuroimaging, neurodegeneration","lastPublishedDoi":"10.21203/rs.3.rs-9385305/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-9385305/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e: Traumatic encephalopathy syndrome (TES) is a clinical research construct used to identify individuals at risk for chronic traumatic encephalopathy (CTE) following exposure to repetitive head impacts (RHI). Adjudication of TES relies on clinical features such as progressive cognitive impairment and neurobehavioral dysregulation. Blood-based biomarkers and structural neuroimaging abnormalities have been associated with TES but are not part of the criteria. This study evaluated whether TES identification was associated with the combined contribution of cognitive performance, blood biomarkers, and structural neuroimaging measures across two well-characterized cohorts.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e: Participants included 158 professional fighters from the Professional Athletes Brain Health Study and 149 former American football players from The DIAGNOSE CTE Research Project. Three indices were constructed representing complementary domains: a cognitive index reflecting cohort-specific cognitive features, a blood biomarker index including plasma neurofilament light chain, glial fibrillary acidic protein, total tau, tau phosphorylated at amino acid 231, and \u003cem\u003eAPOE\u003c/em\u003e-ε4 carrier status, and an imaging index comprising volumetric MRI measures of subcortical structures, ventricles, and corpus callosum subregions. Grouped weighted quantile sum regression models were estimated within each cohort to evaluate associations between these indices and TES while adjusting for age, race, competition status, and RHI exposure.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e: Multidomain models demonstrated improved model performance compared with single-domain models in both cohorts (PABHS: AUC=0.91, PPV=0.80; DIAGNOSE CTE: AUC=0.84, PPV=0.85). Biomarker and imaging indices contributed additional information across cohorts, although imaging contributions were more prominent in fighters whereas blood biomarker associations were stronger in football players.\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion\u003c/strong\u003e: TES in RHI-exposed athletes was associated with a convergent clinicobiological profile observed across two independent cohorts with distinct exposure patterns. These findings support multidomain analytic frameworks for evaluating correlated biological signals in RHI-exposed populations and may inform future studies of TES and CTE.\u003c/p\u003e","manuscriptTitle":"Cognitive, biomarker, and neuroimaging indices associated with traumatic encephalopathy syndrome across two independent athlete cohorts","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2026-04-28 00:18:22","doi":"10.21203/rs.3.rs-9385305/v1","editorialEvents":[{"type":"communityComments","content":0},{"type":"editorInvitedReview","content":"","date":"2026-05-12T15:22:26+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"124066079295440537429063439873946988015","date":"2026-04-24T16:22:34+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"109958989024432413897804252397518694061","date":"2026-04-20T03:37:30+00:00","index":"hide","fulltext":""},{"type":"reviewerAgreed","content":"193313211626533737299391326866621347786","date":"2026-04-18T15:17:51+00:00","index":"hide","fulltext":""},{"type":"reviewersInvited","content":"","date":"2026-04-18T10:27:45+00:00","index":"","fulltext":""},{"type":"editorAssigned","content":"","date":"2026-04-16T07:59:58+00:00","index":"","fulltext":""},{"type":"checksComplete","content":"","date":"2026-04-16T07:59:17+00:00","index":"","fulltext":""},{"type":"submitted","content":"Alzheimer's Research \u0026 Therapy","date":"2026-04-11T06:47:55+00:00","index":"","fulltext":""}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"alzheimers-research-and-therapy","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":false,"externalIdentity":"azrt","sideBox":"Learn more about [Alzheimer's Research and Therapy](http://alzres.biomedcentral.com/)","snPcode":"13195","submissionUrl":"https://submission.nature.com/new-submission/13195/3","title":"Alzheimer's Research \u0026 Therapy","twitterHandle":"@AlzheimersRes","acdcEnabled":true,"dfaEnabled":true,"editorialSystem":"em","reportingPortfolio":"BMC/SO AJ","inReviewEnabled":true,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"17eabfbd-4f2b-4884-8b4d-8f8e3515fe8c","owner":[],"postedDate":"April 28th, 2026","published":true,"recentEditorialEvents":[{"type":"editorInvitedReview","content":"","date":"2026-05-12T15:22:26+00:00","index":29,"fulltext":""}],"rejectedJournal":[],"revision":"","amendment":"","status":"under-review","subjectAreas":[],"tags":[],"updatedAt":"2026-04-28T00:18:22+00:00","versionOfRecord":[],"versionCreatedAt":"2026-04-28 00:18:22","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-9385305","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-9385305","identity":"rs-9385305","version":["v1"]},"buildId":"XKTyCvWXoU3ODBz1xrDgd","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}