The Journal of Immunology

Acute respiratory distress syndrome (ARDS) is a devastating complication of respiratory infections; however, the biological mechanisms that initiate its onset are poorly defined. Here we show that TNFRSF13B polymorphisms increase the risk of ARDS following SARS-CoV-2 infection up to 7.4-fold compared to the WT genotype. The increased risk was not due to immune-deficiency or impaired virus neutralization. On the contrary, TNFRSF13B mutant subjects mounted better antibody neutralization compared to subjects with WT TNFRSF13B. However, IgG from subjects expressing TNFRSF13B variants had less sialic acid, terminal galactose, and fucose than IgG from subjects with a WT genotype. Moreover, IgG from TNFRSF13B mutant subjects exhibited increased recruitment of complement factors. Thus, besides well-known actions governing plasma cell differentiation, TNFRSF13B impacts both affinity maturation and effector functions of IgG in ways that independently govern complement activation controlling inflammatory responses known to trigger ARDS.

Flow cytometry can establish T cell clonality by detecting a restricted expression pattern of the T cell receptor (TCR) {beta} constant region (TRBC), expressed in association with CD3. However, T cell neoplasms frequently lose surface expression of the CD3/TCR complex, posing a challenge to demonstrating T cell lineage and clonality. To address this challenge, here we present a 12-color flow cytometry panel, called cytoTCR, to characterize cytoplasmic expression of CD3/TCR complex components. We apply cytoTCR to 38 patient specimens with immunophenotypically abnormal T cell populations, demonstrating this approach can efficiently establish T cell lineage and clonality in challenging T cell neoplasms that have lost surface CD3 expression. While we show that natural killer (NK)-lineage neoplasms can express cytoplasmic CD3 at similar levels to T cells, we show that absent expression of cytoplasmic TCR components by mature lymphocytes can help confirm NK cell lineage. We demonstrate that cytoTCR can detect cytoplasmic TRBC-restriction in challenging cases of null-phenotype anaplastic large cell lymphoma, which lack surface expression of pan-T cell antigens. In cases of T-lymphoblastic leukemia, cytoTCR shows that cytoplasmic TRBC expression matches the expected developmental stage of the leukemia. Finally, we use cytoTCR to characterize atypical cCD3-CD7- T cells in a patient with a history of T-lymphoblastic leukemia as well as recent CAR-T therapy, showing that this atypical population is polytypic and represents CAR-T product rather than residual disease. Our study presents a broadly applicable flow cytometric approach to simultaneously assess T cell lineage and clonality in suspected T lineage populations with absent surface CD3 expression.

Brain-derived tau (BD-tau) is an emerging blood-based biomarker for neurodegeneration, yet there are currently limited well validated BD-tau assays available for research and clinical use. To enhance access to this vital biomarker for neurological disorders including traumatic brain injury (TBI), we developed a novel blood-based immunoassay for BD-tau on the ultra-sensitive Quanterix HD-X platform using Single Molecule Array technology. Analytical validation assessed dilution linearity, specificity, precision, detection limits, and spike recovery, each recording robust metrics in agreement with international expert recommendations. The assay demonstrated robust validation metrics, achieving between-run stability of 95% when analyzing aliquots from six independent plasma and serum samples across five analytical runs. It also showed strong dilution linearity when diluted four-fold and achieved over 90% recovery when spiked with cerebrospinal fluid. Next, we evaluated the clinical utility of the assay in cohorts of individuals with traumatic brain injury (TBI), where strong performances were recorded whether using the 2-step or 3-step assay formats ({rho}= 0.94; p < 0.0001). Furthermore, plasma BD-tau distinguished samples from TBI patients based on time from injury and severity (AUC=0.93). Plasma BD-tau differentiated between favorable and unfavorable functional outcomes in the acute-severe group. Our findings underscore the significant potential of the BD-tau assay as a biomarker for TBI in the severe phase.

Background Endometrial cancer exhibits marked molecular and immune heterogeneity that is only partially explained by established genomic biomarkers. We investigated whether T cell receptor (TCR) repertoire architecture captures complementary dimensions of antitumor immunity beyond conventional molecular classification. Methods Paired tumor and peripheral blood samples from eight patients with molecularly characterized endometrial cancer underwent TCR repertoire profiling. Diversity, clonality, and tumor blood overlap metrics were integrated with genomic variables, including tumor mutational burden (TMB), genomic instability metric (GIM), and POLE status. Principal component analysis and correlation analyses were used to identify major dimensions of repertoire organization. Composite Immune Focusing and Immune Sharing Scores were derived to summarize dominant repertoire patterns. Results The first two principal components explained 70.1% of total repertoire variance and revealed substantial heterogeneity independent of histological subtype. TMB was strongly associated with reduced repertoire diversity and increased clonal dominance, resulting in a robust association with the Immune Focusing Score ({rho} = 0.88, p = 0.004). POLE mutated tumors occupied the extreme end of this focusing continuum. In contrast, genomic instability was associated with increased tumor blood repertoire overlap and preserved diversity, reflected by a strong correlation between GIM and the Immune Sharing Score ({rho} = 0.76, p = 0.027). The two immune scores showed minimal correlation with each other ({rho} = -0.24, p = 0.57), indicating that they capture largely independent aspects of immune organization. Conclusion Integrative analysis of TCR repertoire architecture and tumor genomics identifies distinct immunogenomic states in endometrial cancer that are not fully captured by conventional molecular classification. If validated in larger cohorts, immune focusing and immune sharing metrics may provide complementary biomarkers for patient stratification and immunotherapy-oriented precision oncology

Objective: To characterize pregnancy outcomes and menstrual irregularities in Mexican women with systemic lupus erythematosus (SLE) and identify clinical factors associated with adverse pregnancy outcomes and early-onset menopause. Methods: We conducted a cross-sectional study of women with SLE enrolled in the Mexican Lupus Registry (LupusRGMX) between May 2021 and September 2024. Clinical and reproductive data were collected using standardized questionnaires. Menopause was defined as the absence of menstruation for [&ge;]12 consecutive months, and early menopause as onset before age 40. Univariable and multivariable logistic regression analyses were used to identify factors associated with pregnancy complications and early menopause. Results: A total of 210 women were included. Median age was 38 years (IQR 29-46) and median disease duration was 4 years (IQR 1-10). Among women with a history of pregnancy (47%), full-term delivery predominated (61%), while pregnancy loss occurred in 26% and preterm delivery in 13%. Pregnancy complications were reported in 9.6%, most commonly preeclampsia (6.7%). Younger maternal age was independently associated with pregnancy complications (OR 0.89, 95% CI 0.83-0.95) and adverse outcomes (OR 0.95, 95% CI 0.92-0.98). Higher disease activity was associated with complications in univariable analysis. Most pregnancies (68.3%) occurred before diagnosis. Early menopause was observed in 6.2% and independently associated with longer disease duration and older age. Conclusion: Younger maternal age was independently associated with adverse pregnancy outcomes, whereas disease activity showed an association in univariable analysis. Most pregnancies occurred prior to SLE diagnosis. Early menopause was associated with longer disease duration, suggesting impact of cumulative disease burden on ovarian function.

Vaccines to prevent infant group B streptococcus (GBS) disease are advancing, with licensure likely based on safety and immunologic endpoints rather than clinical efficacy data. This approach requires robust, generalisable serological thresholds of risk reduction (SToRRs). We combined data from six case-control studies in Europe and Africa to define SToRRs for early-onset (EOD) and late-onset (LOD) GBS disease. Across diverse epidemiological and healthcare settings, anti-capsular polysaccharide IgG concentrations were consistently higher in infants who remained disease free than in those who developed disease. Higher antibody concentrations were required to reduce the risk of EOD than LOD, and higher concentrations were required for serotype Ia than for serotype III. This study provides a quantitative framework to support correlates-based evaluation and potential licensure of maternal GBS vaccines.

Effective filtration and concentration of stool specimens is an essential pre-analytical step for reducing fecal debris and improving organism recovery using microscopy-based ova and parasite (O&P) examination. This study evaluated three commercially available fecal sedimentation-based filtration/concentration systems, ParaPak SpinCon (Meridian Bioscience), Mini Parasep SF (Apacor), and the newly-available ParadeviceReingenuity), for qualitative parasite detection and workflow logistics using conventional and artificial intelligence (AI)-assisted microscopy. Forty clinical stool specimens (20 parasite-positive and 20 parasite-negative) were processed with the 3 devices, and the resultant 120 wet mount and 120 trichrome stained smear preparations were examined using conventional microscopy. Trichrome-stained slides were also scanned at 40x magnification using a Hamamatsu NanoZoomerS360 flatbed digital slide scanner and images were analyzed using the Techcyte Fusion Human Fecal Trichrome AI algorithm. Positive and indeterminate digital findings were confirmed by conventional glass slide microscopy. Slides and digital images were reviewed in a blinded manner. Concordance was assessed among the 360 initial evaluations (microscopy and AI-assisted), and discrepant parasitology results were resolved through re-review and specimen reprocessing as needed. Final qualitative agreement across slide/image evaluations using all three concentration systems was 100%. Minor discrepancies in protozoan and white/red blood cell detection/identification were noted in 5 and 7 cases, respectively, and likely reflected sampling and observer variability. While the three concentration systems produced equivalent qualitative results, the Paradevice and Mini Parasep SF offered the most streamlined workflows. These findings support the Paradevice and Mini Parasep SF as efficient, analytically equivalent systems that are compatible with traditional and AI-assisted O&P workflows.

In February 2026, the FDA announced that a single pivotal phase 3 (P3) trial would become the new default standard for drug approval - a regulatory direction that had been legally enabled since the FDA Modernization Act of 1997. This announcement has strategic, scientific, and economic implications for drug developers, contract research organizations (CROs), and biotech investors. We argue that the expansion of this framework, originally reserved for various niche submissions, represents a paradigm change, dramatically increasing the value of rigorous early phase (P1 and P2) trial design, requiring sponsors to establish both statistical efficacy signals and mechanistic biological understanding before entering phase 3. Using a CNS indication cost model, we show that single P3 approval can reduce total development expenditure from approximately $447 million over 14 years to $297 million over 12 years - a savings of $150 million and providing two years of additional commercial runway for a modeled CNS drug. Case examples including lecanemab, omaveloxolone, and tofersen illustrate how biomarker-informed early phase strategies can establish the confirmatory evidence necessary for single-trial approval. We provide practical guidance for maximizing the value of P1 and P2 under this evolving framework.

Introduction: Synthetic oligopeptides provide a rapid and cost-efficient approach to developing antibodies and diagnostics for emerging viral variants. Methods: This study computationally and experimentally characterized a synthetic peptide analog of the SARS-CoV-2 spike subdomain 2 major disulfide loop (SD2MDL), designated S621 (CPVAIHADQLTPTWRVYSTC). Binding affinity was computationally estimated using the Heuristic Affinity Prediction Tool for Immune Complexes (HAPTIC), while experimental validation was performed using enzyme-linked immunosorbent assay (ELISA) with rabbit-derived antipeptide antibodies. Clinical diagnostic accuracy testing was done using plasma samples from RT-PCR-confirmed COVID-19 patients and pre-COVID-19 controls. Results: S621 demonstrated nanomolar binding affinity (Kdapp = 1.14 nM) and high avidity (3.67 nM), closely matching HAPTIC predictions (3.54 nM). Diagnostic evaluation yielded a sensitivity of 89.92% and specificity of 27.79%, corresponding to an overall accuracy of 71.79%. Discussion: These findings demonstrate that a single synthetic peptide derived from a conserved spike subdomain can function as a high-affinity surrogate for full-length antigens, supporting its potential application in rapid peptide-based immunodiagnostics.

Seroprevalence surveys reveal the extent of humoral immunity against pathogens such as severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), and under some circumstances represent cumulative incidence of prior infection. However, antibody waning - or seroreversion - biases these estimates by reducing assay sensitivity in a time-varying manner. Because assay sensitivity decays over time, naively using serosurveys can substantially bias estimates of SARS-CoV-2 cumulative incidence and fatality rates. The Bayesian assay-specific, time-varying sensitivity adjustment developed in this paper can reliably correct for this bias and account for the delay between infection and serosurvey. In seroprevalence studies conducted in the United States in 2020, adjusting for time-varying sensitivity increased cumulative incidence by up to 1.4-fold, with an adjustment of 1.08 for a national study. Our estimates contrast with a previously published 2-fold adjustment that did not account for assay design. This suggests that previous analyses overestimated cumulative incidence by applying seroreversion corrections that did not account for assay-specific effects, or underestimated cumulative incidence by not applying seroreversion corrections. These biases imply fatality rate underestimation and overestimation, respectively. Our model provides a framework for design-specific time-varying sensitivity corrections in seroprevalence surveys for other pathogens.

Functional ex vivo assays using live tumor tissues have demonstrated strong predictive accuracy for response to immune checkpoint inhibitors (ICIs) but are not scalable, requiring manual processing of large resections collected at academic centers. Here, an ex vivo live tumor fragment (LTF) platform was developed using standard-of-care biopsies from 228 patients with suspected malignancy collected across prospective, multicenter observational trials and biobanks. Hierarchical clustering of ICI-mediated changes in cytokine production identified two groups: responders and nonresponders. A binary classifier (elive index) using 8 cytokines achieved an AUC of 0.99 for cluster prediction. elive index correctly predicted clinical benefit in 93% (26/28) of patients (P = 3.2x10-5) and accurately identified 83% (10/12) of objective responders. Critically, elive responders were identified among biomarker-negative patients, highlighting the platform as a scalable approach that complements existing companion diagnostics and expands the population of patients identified to benefit from ICI therapy.

Introduction: Brain-predicted age, estimated from structural MRI data, is a machine-learning biomarker of biological brain aging. Greater brain age gap (BAG) indicates advanced brain aging and is associated with cognitive decline and mortality. Cardiometabolic risk factors, including elevated blood glucose, body mass index (BMI), blood pressure, and cholesterol, increase risk of cognitive impairment and dementia in aging. Their relationship with BAG in severe obesity remains poorly characterized despite increased prevalence of cardiometabolic risk factors among this population. Methods: T1-weighted MRI data from 97 adults (BMI 35-73) were used to calculate BAG using ENIGMA and Pyment brain age models. Associations between BAG and HbA1c, BMI, hypertension, and hyperlipidemia were examined using multiple linear regression and MM-estimation robust regression, adjusting for age, sex, and race. Post hoc analyses stratified models by clinical HbA1c cutoffs (normoglycemic, prediabetic, diabetic). Results: Higher HbA1c was associated with greater BAGENIGMA (B = 1.58, p = .014) and BAGPyment (B = 0.93, p = .013) in linear regression models. In robust models, HbA1c remained significantly associated with BAGENIGMA (B = 1.70, p = .002) but not BAGPyment (B = 0.71, p = .13). BMI, hypertension, and hyperlipidemia were not associated with BAG in either linear or robust models. HbA1c was associated with greater BAGENIGMA (B = 2.15, p = .01) and BAGPyment (B =1.21, p = .04) in those at or above prediabetic levels and with BAGENIGMA (B = 2.49, p = .047) in those with diabetes. Conclusions: Elevated HbA1c is associated with accelerated brain aging in individuals with severe obesity. BAG was not associated with BMI, hypertension, and hyperlipidemia, which may reflect the restricted BMI range inherent to the sample with severe obesity.

Since the U.S. 2013/14 influenza season, the CDC's FluSight Challenge has provided a platform for evaluating influenza forecasting models and fostering collaboration across institutions. The Challenge aims to improve the science and enhance the utility of infectious disease forecasts for public health decision making. We analyzed ten years of submitted forecasts (2014/15-2019/20 (influenza-like illness seasons) and 2021/22-2024/25 (hospital admissions seasons)) across a range of model types, including statistical, mechanistic, machine learning, and hybrid models. Influenza-like illness (ILI) forecasts were evaluated using the exponentiated logarithmic score (skill metric) while hospital admissions forecasts were evaluated using the log transformed relative Weighted Interval Score. Corresponding potential performance differences were assessed using Wilcoxon rank-sum tests, and associations with team participation history were evaluated using Spearman's rank correlation. Model performance varied by season, and no single model type consistently outperformed others. In ILI seasons, statistical models generally performed better than mechanistic and machine learning models, though consistent differences were not observed in more recent hospital admissions seasons. Ensemble forecasts showed better overall performance across seasons, and the CDC's FluSight ensemble ranked among the top-performing forecasts every year. We also found a positive correlation between forecast accuracy and the number of years a team participated in the Challenge, with statistically significant associations in four seasons. These findings highlight the benefits of ensemble approaches and sustained engagement in improving forecasting performance, while also underscoring the continued value of forecast evaluation before and following the COVID-19 pandemic. Insights from the FluSight Challenge can guide future infectious disease forecasting efforts and support more effective public health preparedness.

An ongoing outbreak of Bundibugyo virus disease (BVD) in the Democratic Republic of the Congo was deemed a public health emergency of international concern in May 2026. To prevent cross-border importation, many countries, including the United States, Canada, India, Thailand, and Kenya have already proposed containment strategies, and others are likely to follow suit. How well (or poorly) are screening and quarantine containment measures are likely to work? We leverage established epidemiological theory and develop a mathematical model of traveler screening and post-arrival quarantine for BVD to answer this question. We find that traveler screening via symptom screening or molecular testing will miss the majority of infected travelers, and should be complemented by post-arrival quarantine and monitoring of sufficient duration to detect those with long incubation periods. Our findings underscore the limitations of border screening and the importance of complementary measures like post-arrival quarantine to prevent local importation of BVD.

Human metapneumovirus (HMPV) causes acute lower respiratory infections, primarily affecting young children and older adults, with seasonal outbreaks peaking annually in March or April in the United States and other temperate regions in the Northern hemisphere. However, the factors driving HMPV seasonality in the United States remain poorly understood. We analyzed laboratory-confirmed HMPV cases and age-specific emergency department visits across 10 US regions, fitting an age-stratified dynamic transmission model to assess spatiotemporal patterns and investigate the influence of environmental variables and viral interference from RSV on HMPV transmission rates. We found that models incorporating climate variables into the transmission rate, including vapor pressure, precipitation, potential evapotranspiration, and minimum temperature, could not capture the timing of HMPV activity across all regions. Instead, HMPV timing was associated with RSV activity, with the HMPV transmission rate reduced in the presence of RSV. We showed that, unlike RSV, only models incorporating viral interference could reproduce the biennial pattern of HMPV observed in some regions, characterized by alternating late-small and early-large epidemics. Furthermore, our model successfully reproduced post-COVID-19 HMPV and RSV epidemics and predicted that RSV interventions are not likely to lead to a substantial increase in HMPV activity despite decreasing competition from RSV. Our work unravels the spatiotemporal dynamics of HMPV and its interaction with RSV, informing future seasonal forecasting and intervention strategies for HMPV.

Synapse loss is an early feature of neurodegeneration and may provide sensitive biomarkers for experimental medicine. Positron emission tomography (PET) with the synaptic vesicle glycoprotein 2A radioligand [11C]UCB-J shows widespread signal reduction across dementias. However, it remains unclear which aspects of synaptic integrity [11C]UCB-J PET measures. We developed a histological-imaging pipeline to quantify structurally intact synapses in post-mortem brain tissue. We applied it to six donors with the tauopathy progressive supranuclear palsy (PSP) who had ante-mortem [11C]UCB-J-PET, alongside six controls across 11 brain regions. Synapse loss in PSP was widespread but region-specific across cortical, subcortical, and brainstem regions. Greater synapse loss was associated with higher tau burden and pathology, and cortical synaptic density correlated with ante-mortem cognition. Post-mortem synaptic density correlated with in vivo [11C]UCB-J-PET signal. This study provides validation of SV2A PET as a biomarker of synaptic density and supports integration of imaging with histopathology in neurodegenerative disease research.

Background Coeliac disease (CD) diagnosis on duodenal biopsies is limited by interobserver variability. We have previously demonstrated pathologist-level performance with our artificial intelligence (AI) model for the histopathological diagnosis of adult CD, but not in paediatric practice. As paediatric CD screening programmes expand internationally, accurate and scalable diagnostic tools are needed. We investigated whether an AI model trained exclusively on adult whole-slide images (WSIs) can generalise to paediatric CD diagnosis across independent centres. Methods A training and validation dataset of 9,958 WSIs from 8,421 adult patients (961 CD) from five centres was used to develop an ensemble of multiple-instance learning models using features from a foundation model. Testing was performed on 708 consecutive paediatric patients (86 CD) from two centres (Edinburgh and Southampton) not included in training. Model calibration was assessed, and probability outputs were grouped into clinically interpretable categories. Findings In adult cross-validation, the AI model achieved an area under the receiver operating characteristic curve (AUC) of 98.7%, sensitivity of 84.9%, specificity of 99.0%, and negative predictive value (NPV) of 98.1%. On testing (paediatric) datasets, performance remained high (AUC 98.8%, sensitivity 80.2%, specificity 98.4%, NPV 97.3%). Restricting analysis to predictions outside the intermediate-probability range (predicted CD probability <10% or [&ge;]65%; 85.3% of cases) improved sensitivity to 100% and specificity to 98.7%. No misclassifications were observed among high-confidence predictions (<2% or [&ge;]85%; 66.0% of cases). The expected calibration error was 0.03. Performance improved significantly when biopsies from both duodenal sites (bulb [D1] and descending [D2/3]) were considered. Interpretation Our AI model, trained on adult biopsies, generalises to paediatric CD diagnosis across centres and scanner platforms. Well-calibrated probability outputs provide clinically interpretable measures of diagnostic confidence and could support safe identification of CD-negative biopsies within defined thresholds. These findings demonstrate the feasibility of applying adult-derived AI models in paediatric populations and reinforce the importance of multi-site (D1 & D2) biopsy sampling.

Postoperative resilience varies widely among older adults, yet the biological drivers of recovery remain unclear. We evaluated whether preoperative immune profiles, measured in plasma and through ex vivo whole blood stimulation, predict resilience to the acute stress of total knee arthroplasty. A total of 152 adults (greater or equal to 60 years) in the PRIME KNEE cohort underwent elective total knee arthroplasty and had available blood samples for measurement of 45 immune biomarkers, quantified in plasma and in whole blood stimulated ex vivo for 24 hours with lipopolysaccharide (LPS) or influenza antigen (FLU). Resilience was assessed using Expected Recovery Differential (ERD) and Resilience Trajectory (RT) across pain severity, pain interference, lower extremity physical activities of daily living (LE PADLs), and step counts. An exploratory stability selection framework using LASSO identified biomarker predictors of postoperative outcomes. Plasma and stimulated biomarkers showed broadly similar predictive performance. A shared set of biomarkers, including LBP, leptin, TNFR1, CD30, and LIF, was consistently selected across models. Immune predictors explained ~12-24% of the variance in resilience outcomes. Distinct immune signatures emerged for pain versus functional recovery: pain related predictors mapped to local inflammatory and neuroimmune pathways, whereas function related predictors reflected systemic inflammatory load and cytokine signaling. Preoperative immune biomarkers, whether measured in plasma or after ex vivo stimulation, capture meaningful variance in postoperative resilience. The divergence between pain related and function related immune signatures highlights biologically distinct pathways underlying different dimensions of recovery and supports further development of immune based perioperative risk assessment.

Background: Mucopolysaccharidosis type IIIB (MPS IIIB) is a devastating neurodegenerative lysosomal storage disorder caused by alpha-N-acetylglucosaminidase (NAGLU) deficiency. There is currently no approved therapy. We report the 3-month outcomes of a novel intracerebroventricular (ICV) gene therapy in a child with MPS IIIB. Methods: In an open-label, single-center, investigator-initiated trial (ChiCTR2600121466), a single dose of RDGT-101 (2.0E14; vg of an AAV9 vector encoding human NAGLU) was administered via ICV infusion. Primary outcomes were safety and tolerability. Secondary outcomes included serum NAGLU activity, urinary heparan sulfate (HS) excretion, and neurocognitive function. Exploratory analyses included hematological parameters. Results: The patient achieved serum NAGLU activity (17.06 nmol/mL/hour) approaching that of healthy controls (17.75 {+/-} 1.37 nmol/mL/hour) by Month 3, accompanied by a 58.4% reduction in urinary HS. Clinically, previously severe hand and toe contractures resolved, allowing for full extension. Neurocognitive improvements were observed, including clear articulation, logical conversation, and sustained eye contact. Hematological analyses revealed normalized red blood cell indices and improved iron utilization. No dose-limiting toxicities, serious adverse events, or clinically significant laboratory abnormalities were observed. Conclusions: A single ICV infusion of RDGT-101 was safe and well-tolerated in this patient with MPS IIIB. Early biochemical correction was accompanied by marked improvements in somatic, neurocognitive, and hematological parameters. These findings support further investigation of ICV AAV9 gene therapy for MPS IIIB.

Objective: Tumor-infiltrating lymphocytes (TILs) in breast cancer are one of the most important indicators of the immune response within the tumor microenvironment. They play a particularly significant prognostic and predictive role in triple-negative and HER2-positive subtypes. However, substantial inter-observer variability has been reported in TIL scoring among pathologists, which limits its reliability in clinical practice. The aim of this study was to evaluate the agreement between artificial intelligence (AI) models and pathologists in TIL scoring and to compare this agreement using different statistical approaches, thereby assessing the potential of AI integration into pathology practice. Materials and Methods: Digitized histopathological images of breast cancer cases were included in the study. Tumor regions annotated by pathologists were evaluated for both stromal TIL percentage and the proportion of stromal tumor area within each ROI, with assessments performed independently by three pathologists and two AI models. Agreement was assessed among pathologists, between pathologists and AI, and between AI models. Statistical analyses included intraclass correlation coefficient (ICC), Cohen and Fleiss kappa, correlation tests, and Bland-Altman analysis. In addition, categorical agreement was examined using different cut-off values. Results: Inter-pathologist agreement was high, with an ICC of 0.81. In contrast, the global agreement between pathologists and AI models was lower (ICC 0.41). Pairwise comparisons of pathologist-AI agreement yielded substantially lower ICC values (0.12-0.21), although this improved to 0.53 when three pathologists were assessed jointly with a single AI model. The strongest categorical agreement was observed with dichotomized TIL scores ([&le;]10% vs. >10%), whereas multi-category classifications were associated with a marked reduction in kappa values. Spearman correlation coefficients between pathologists and AI models ranged from moderate to good ({rho} = 0.48-0.81). Agreement between the two AI models themselves was moderate, with an ICC of 0.64