Tertiary lymphoid structures (TLSs) have been associated with favorable immunotherapy responses and prognosis in various cancers. Despite their significance, their quantification using multiplex immunohistochemistry (mIHC) staining of T and B lymphocytes remains labor-intensive, limiting its clinical utility. To address this challenge, we curated a dataset from matched mIHC and H&E whole-slide images (WSIs) and developed a deep learning model for automated segmentation of TLSs. The model achieved Dice coefficients of 0.91 on the internal test set and 0.866 on the external validation set, along with intersection over union (IoU) scores of 0.819 and 0.787, respectively. The TLS ratio, defined as the segmented TLS area over the total tissue area, correlated with B lymphocyte levels and the expression of CXCL13, a chemokine associated with TLS formation, in 6140 patients spanning 16 tumor types from The Cancer Genome Atlas (TCGA). The prognostic models for overall survival indicated that the inclusion of the TLS ratio with TNM staging significantly enhanced the models’ discriminative ability, outperforming the traditional models that solely incorporated TNM staging, in 10 out of 15 TCGA tumor types. Furthermore, when applied to biopsied treatment-naïve tumor samples, higher TLS ratios predicted a positive immunotherapy response across multiple cohorts, including specific therapies for esophageal squamous cell carcinoma, non-small cell lung cancer, and stomach adenocarcinoma. In conclusion, our deep learning-based approach offers an automated and reproducible method for TLS segmentation and quantification, highlighting its potential in predicting immunotherapy response and informing cancer prognosis.

Anxiety disorders are widespread across the world. Understanding the global burden of anxiety disorders in the past and future can provide important references for optimizing prevention and control strategies in healthcare systems. The estimation of anxiety disorders in the Global Burden of Disease Study 2021 using systematic review was conducted to describe incidence, prevalence and disability adjusted life years (DALYs) in 204 countries and regions from 1990 to 2021. Trends were examined using joinpoint regression analysis and correlation between Sociodemographic Index (SDI) and temporal trends was assessed. In 2021, the global age-standardised prevalence rate (ASPR), age-standardised incidence rate (ASIR), and age-standardised DALYs rates for anxiety disorders were 4421.9 (95 % uncertainty intervals [UIs] 3768.3 to 5182.1), 678.3(565.2 to 832.4), and 524.3(363.1 to 716.3) per 100,000 population, which were 18 %, 20.6 %, and 18.2 % higher than in 1990, respectively. In 2021, Brazil (9007.4), Portugal (9712.4), and Paraguay (8390) had the highest ASPR of anxiety disorders. Brazil (53.2 %), Peru (38.4 %), and Bolivia (42.2 %) showed the largest increases in ASPR across the study period. In women, the global DALY rate of anxiety disorders increased up to age 20-44 years and then decreased with advancing age, whereas for men the rate the same as in women. Regionally, an overall undulating pattern association was found between SDI and the age standardised DALYs rate of anxiety disorders. The global female ASIR slightly increased from 1990 to 2019 (Annual percent change [APC]: 0.02 %). However, from 2019 to 2021, it saw a substantial rise (APC: 10.78 %). With the development of society, the disease burden of anxiety disorder will become increasingly severe. It is very important to study the accurate epidemiological data on anxiety disorder for clinical diagnosis and treatment and health policy.

Recent advances in cancer immunotherapy have highlighted the potential of neoantigen-based vaccines. However, the design of such vaccines is hindered by the possibility of weak binding affinity between the peptides and the patient’s specific human leukocyte antigen (HLA) alleles, which may not elicit a robust adaptive immune response. Triggering cross-immunity by utilizing peptide mutations that have enhanced binding affinity to target HLA molecules, while preserving their homology with the original one, can be a promising avenue for neoantigen vaccine design. In this study, we introduced UltraMutate, a novel algorithm that combines Reinforcement Learning and Monte Carlo Tree Search, which identifies peptide mutations that not only exhibit enhanced binding affinities to target HLA molecules but also retains a high degree of homology with the original neoantigen. UltraMutate outperformed existing state-of-the-art methods in identifying affinity-enhancing mutations in an independent test set consisting of 3660 peptide–HLA pairs. UltraMutate further showed its applicability in the design of peptide vaccines for Human Papillomavirus and Human Cytomegalovirus, demonstrating its potential as a promising tool in the advancement of personalized immunotherapy.

Deep learning (DL) models for medical image analysis are majorly bottlenecked by the lack of well‐annotated datasets. Bronchoalveolar lavage (BAL) is a minimally invasive procedure to diagnose lung cancer, but BAL cytology suffers from low sensitivity. The success of DL in BAL cytology is rare due to the rarity of exfoliated tumor cells (ETCs) and their subtle morphological differences from normal cells. Single‐cell DNA sequencing (scDNA‐Seq) is utilized as an objective ground truth of ETC annotation for generating an unbiased, accurately annotated dataset comprising 580 ETCs and 1106 benign cells from BAL cytology slides. A DL model is developed, to distinguish ETC from benign cells in BAL fluid, achieving an Area Under the Curve of 0.997 and 0.956 for detecting large‐ and small‐sized ETCs, respectively. The model is applied in a discovery cohort (n = 156) to establish BAL‐based cytopathologic diagnostic model for lung cancer. The model is evaluated in a validation cohort (n = 158), and yielded 47.6% sensitivity and 97.7% specificity in lung cancer diagnosis, outperforming cytology with improved sensitivity (47.6% vs 19.0%). In an external validation cohort (n = 141), the model achieved 60.0% sensitivity and 92.5% specificity in lung cancer diagnosis.

Background: Non-small cell lung cancer (NSCLC) remains one of the leading causes of cancer-related mortality worldwide over the past decade. Single-cell sequencing loses spatial location information and potential cell–cell interactions, making it difficult to interpret molecular features or biological phenomena. Tertiary lymphoid structures (TLSs) inherently require such spatial immune cell distribution information. Although associations between TLS and response to immune checkpoint inhibitors (ICIs) or chemotherapy have been reported, the relationship between TLS and neoadjuvant therapy (ICI combined with chemotherapy) remains unclear. Methods: We performed spatial transcriptomics on NSCLC samples (including one lung squamous cell carcinoma (LUSC) and one lung adenocarcinoma (LUAD)). Multiplex immunohistochemistry (mIHC) was used to identify the TLS, while immunohistochemistry staining (IHC) was used to identify the TLS status and cell characteristics. We evaluated the associations between (mature) TLS density, area proportion and patients’ responses in 66 patients. Results: Heterogeneity of immune cells in NSCLC was found. Gene ontology analysis and cell score comparison identified TLS with activated B and T cells inside, while plasma cells and macrophages were mainly distributed outside TLS. Four genes from antigen-presenting machinery (TAP1, TAP2, B2M, TAPBP) were more highly expressed inside TLS than outside them. Also, TLS exhibited heterogeneity, with both mature and immature TLS. Mature TLS showed an average area of 62,387.43 μm², while the immature TLS showed 51,189.90 μm². The Spearman correlation coefficient of B-cell number and mTLS area showed r = 0.900. TLS density and mature TLS (mTLS) density in the tumor bed were 1.95 ± 0.95 TLS/10 mm² (mean ± SD, n = 34) and 1.13 ± 0.77 mTLS/10 mm², significantly higher than that in the non-responder group (1.18 ± 1.15 TLS/10 mm², 0.70 ± 0.90 mTLS/10 mm², mean ± SD, n = 32) separately. B cells belonging to TLS had a significantly higher density (71.32 ± 55.71 cells/mm², mean ± SD, n = 34) in the responder group than the non-responder group (61.33 ± 111.95 cells/mm², mean ± SD, n = 32) normalized to the tumor bed area. Conclusions: Spatial transcriptomics reveals immune cell heterogeneity and distribution patterns in the NSCLC tumor bed, with activated B and T cells localized inside and plasma cells/macrophages outside. Antigen-presenting machinery (APM)-related genes were highly expressed in TLS accompanied by a high expression of upstream and downstream genes of MHC class I. mTLS have a larger area by mainly containing more B cells. The responder group had a significantly higher (mature) TLS density and larger (mature) TLS area proportion compared with the non-responder group, suggesting their potential function in anti-tumor effect in neoadjuvant treatment.