BACKGROUND:Postoperative recurrence remains the leading cause of treatment failure in gastric cancer despite curative resection. Current surveillance strategies do not adequately reflect time-varying recurrence risk. This study aimed to characterize the temporal dynamics and spatial patterns of recurrence using hazard function analysis to inform risk-adapted follow-up. METHODS:In this multicenter retrospective cohort study, 3278 patients who underwent curative gastrectomy between 2012 and 2020 at ten tertiary centers in China were included. Kernel-smoothed hazard functions were applied to estimate instantaneous recurrence risk over time. Disease-free survival (DFS) and overall survival (OS) were analyzed using Kaplan-Meier and Cox regression models. RESULTS:During a median follow-up of 61 months, 1630 patients (49.7%) developed recurrence. Overall recurrence risk peaked at 12.7 months postoperatively (hazard rate: 0.021). Stage-specific analysis showed an early and pronounced peak in stage III disease (13.9 months; hazard rate: 0.040), whereas stage II disease demonstrated a later, attenuated peak (17.0 months; hazard rate: 0.012). Peritoneal dissemination was the most aggressive pattern, peaking at 10.0 months and accounting for 29.1% of recurrences, while locoregional recurrence showed a delayed peak at 28.7 months (9.9%). Adjuvant chemotherapy status was associated with distinct recurrence hazard trajectories: patients without chemotherapy showed an earlier and higher peak at 10.3 months (hazard rate: 0.091), whereas those receiving chemotherapy showed a lower and later peak at 16.6 months (hazard rate: 0.037). Multivariable analysis identified pT4 stage, pN3 stage, lymphovascular invasion, and perineural invasion as independent risk factors, while adjuvant chemotherapy was protective. CONCLUSIONS:Gastric cancer recurrence follows stage-dependent, non-linear trajectories with distinct temporal and site-specific patterns. The high-risk window of 10-20 months postoperatively may represent a relevant period for intensified surveillance, particularly for peritoneal dissemination. These findings provide a data-informed basis for generating risk-adapted, time-sensitive follow-up hypotheses that require prospective validation.

Colon adenocarcinoma (COAD) exhibits substantial molecular and microenvironmental heterogeneity, limiting the reliability and clinical application of single-omics biomarkers. Post-translational modifications (PTMs) play a pivotal role in connecting genotype to phenotype, yet prognostic models informed by PTMs seldom incorporate tissue architecture alongside transcriptomic states. This study developed a PTM-based multimodal framework for prognostic stratification and mechanistic analysis in COAD. Additionally, bulk RNA-seq data, clinicopathological information, and whole-slide H&E images from TCGA-COAD were integrated with single-cell RNA-seq (GSE132465) and spatial transcriptomics (GSE225857). Differentially expressed genes related to PTMs were identified, and pathomic features derived from H&E slides were fused with transcriptomic data via an autoencoder-based latent space. Prognosis-related latent features were then selected to establish a patient-level PTM-related multimodal risk score (PTMLS). A total of 102 PTM-related differentially expressed genes were enriched in glycan biosynthesis and ubiquitin-mediated pathways. The PTMLS effectively stratified patients into high- and low-risk groups, with significant differences in overall survival in both training (HR = 3.54, 95% CI 2.67-4.70, P = 0.001) and validation cohorts (HR = 2.68, 95% CI 1.49-4.82, P = 0.006). Low-risk tumors displayed higher immune and ESTIMATE scores, lower TIDE scores, distinct immunotherapy responder profiles, and differential predicted drug sensitivity. Cross-modal interpretation identified B3GNT6 as a key contributor linked to histopathological features. Single-cell and spatial analyses localized B3GNT6-associated epithelial states and their interactions with vascular and stromal compartments. Functional experiments further revealed that B3GNT6 expression was downregulated in tumors and inhibited colorectal cancer proliferation, migration, tumor growth, and metastasis. This PTM-informed multimodal framework integrates routine pathology and transcriptomic data, providing robust prognostic stratification, clinically relevant immune and therapeutic phenotyping, and mechanistic insight supporting B3GNT6 as a potential therapeutic target in COAD.

ETHNOPHARMACOLOGICAL RELEVANCE:Banxia Xiexin Decoction (BXD) is a traditional herbal formula with a history spanning over a millennium, significantly contributing to the treatment of gastrointestinal disorders, especially gastric precancerous lesions (GPL) such as gastritis, intestinal metaplasia, and dysplasia. Although it is effective in reducing inflammation, enhancing intestinal metaplasia, and providing antitumor benefits, the exact mechanisms of its action are not yet understood. AIM OF THE STUDY:This study aims to clarify how BXD influences the gastric microbiota to enhance GPL. MATERIALS AND METHODS:A GPL model was developed using N-Methyl-N'-nitro-N-nitrosoguanidine. The pharmacological effects of BXD on GPL were evaluated using ELISA, qPCR, AB-PAS staining, immunohistochemistry, H&E staining, Western blotting, and immunofluorescence. Subsequently, the protective effects of BXD against MNNG-induced GES-1 cells were evaluated in vitro. Furthermore, network pharmacology was employed to analyze the underlying mechanisms of BXD in the treatment of GPL. 16S rRNA sequencing was further employed to examine the structure and stability of the gastric microbiota. Finally, the roles of Lactobacillus and Streptococcus in GPL were further validated through animal experiments. RESULTS:The BXD intervention decreased TNF-α, IL-1β, and IL-6 levels in both gastric tissue and serum of GPL mice, enhanced IL-10 production, ameliorated intestinal metaplasia, and suppressed Ki67 and p53 expression. BXD mitigated gastric mucosal injury by upregulating the expression of Zonula occludens-1, E-cadherin, and Occludin, as well as inhibiting the PI3K-Akt and MAPK signaling pathways. Furthermore, BXD improved the dysbiosis of the gastric microbiota in GPL mice.The gastric microbiota of mice at the phylum level predominantly consisted of Bacillota, Proteobacteria, Cyanobacteria, Bacteroidota, Actinobacteria, Desulfobacterota, and Campylobacterota.In GPL mice, dysbiosis was notably marked by an increased presence of Bacillota and Proteobacteria, with Lactobacillus and Streptococcus at the genus level being significant contributors to GPL progression. The BXD intervention restored gastric microbiota balance by enhancing Bacillota and Lactobacillus abundance while reducing Proteobacteria and Streptococcus richness. Further intervention with Lactobacillus improved gastric mucosal atrophy and intestinal metaplasia associated with GPL, leading to a reduction in the levels of TNF-α, IL-1β, and IL-6 in gastric tissues, while enhancing the production of IL-10. This treatment also promoted the expression of Zonula occludens-1, E-cadherin, and Occludin, and inhibited the activation of the PI3K-Akt and MAPK signaling pathways. In contrast, treatment with Streptococcus in GPL mice resulted in more severe gastric mucosal atrophy and intestinal metaplasia, marked by increased levels of TNF-α, IL-1β, and IL-6, reduced IL-10 production, no significant increase in the expression of Zonula occludens-1, E-cadherin, and Occludin, and enhanced activation of the PI3K-Akt and MAPK signaling pathways. CONCLUSION:BXD intervention demonstrated significant therapeutic effects in GPL mice. The underlying mechanisms are associated with the restoration of gastric microbiota dysbiosis, promotion of beneficial bacterial growth, suppression of potential pathogens, modulation of the inflammatory response, and alleviation of mucosal damage.

Immune checkpoint inhibitors (ICIs) have transformed the treatment landscape for metastatic colorectal cancer (mCRC) exhibiting DNA mismatch repair deficiency (dMMR) or microsatellite instability-high (MSI-H). However, the comparative efficacy and safety of monotherapy versus dual immunotherapy remain inadequately quantified. This meta-analysis aims to evaluate the benefits and risks of these strategies by quantitatively synthesizing data from randomized controlled trials (RCTs) to inform clinical decision-making. We systematically searched major databases (PubMed, Web of Science, Cochrane, Embase) from inception to September 2025 for relevant RCTs. Outcomes included objective response rate (ORR), progression-free survival (PFS), overall survival (OS), duration of response (DOR), and adverse events. Statistical analyses were performed using RevMan 5.3 and Stata 18.0. Six RCTs (1,460 patients) were included. Pooled analysis showed that PD-1/PD-L1 monotherapy significantly improved ORR compared with chemotherapy (OR = 1.52, 95

In locally advanced gastric cancer, a substantial number of patients relapse with liver metastasis months after an apparently curative operation, and standard tumor staging offers little warning of who is at risk. Here, we develop the Radiopathomics-Clinical Stratification Assessment (RCSA), an interpretable model that integrates three complementary sources of information: radiomic features from preoperative computed tomography, pathomic features from routine hematoxylin and eosin tumor slides, and conventional clinical features. Trained on patients from one hospital and then tested on separate internal, external, public, and prospective trial groups (NCT02555358), RCSA consistently separates high- and low-risk patients, with area under the curves between 0.862 and 0.909. Tumors it labels low-risk carry a notably more active immune environment, indicating that these patients are the ones most likely to gain from added immunotherapy. RCSA therefore turns existing hospital data into individualized guidance for postoperative follow-up and treatment. Accurate prediction of metachronous liver metastasis (MLM) after curative surgery for locally advanced gastric cancer (LAGC) remains challenging. Here, the authors develop a multimodal model to predict MLM in LAGC patients by integrating clinical, imaging, and pathology data across multi-centre cohorts, also revealing patients who could be more responsive to adjuvant immunotherapy.

Gastric cancer staging is frequently limited by the low sensitivity of routine imaging for occult peritoneal metastasis (OPM), necessitating invasive staging laparoscopy. We developed a Multimodal Model, integrating primary tumor radiomics from CT with clinical factors to non-invasively predict OPM in locally advanced gastric cancer. The model was trained and internally validated in a large cohort (n = 940) and externally validated across two independent multi-center cohorts (n = 309), an incremental cohort (n = 477), and a prospective clinical trial cohort (n = 168). In all cohorts, the model achieved robust performance (AUCs: 0.834-0.857), significantly outperforming single-modality models. Crossover validation showed AI assistance increased the average radiologist AUC from 0.735 to 0.872. Transcriptomic analysis revealed that the model’s low-risk stratification correlated with an enhanced antitumor immune microenvironment (CD8 T cells, TNFα signaling). This validated model provides a practical tool for accurate, non-invasive OPM prediction and individualized treatment planning.

Tumor biomaterials show great potential for targeted cancer therapy,yet their development and clinical translation have long been hampered by inefficient empirical trial-and-error models.These traditional methods cannot fully characterize the nonlinear relationships between a material's physicochemical properties and its complex biological effects,nor can they resolve tumor heterogeneity—the primary cause of inconsistent clinical outcomes.This review systematically explores the application of artificial intelligence(AI)across the entire development pipeline of tumor biomaterials,from early rational material design to clinical treatment optimization.We show that AI addresses key bottlenecks in the field in four core ways:it speeds up novel material discovery via generative algorithms,accurately predicts the in vivo transport and uptake of materials,enables noninvasive and precise patient stratification,and optimizes synergistic combination treatment regimens.These advances form a data-driven closed-loop framework that connects preclinical research and clinical translation,overcoming the core limitations of traditional development models.We also outline key unresolved challenges,including data standardization,model interpretability,and regulatory compliance,and highlight AI's growing role as a core driver of precision oncology and translational medicine.

Cardiac organoids have emerged as pivotal models for cardiovascular disease research and drug development due to their ability to recapitulate the in vivo cardiac microenvironment, structure, and function. Imaging technologies are core tools for deciphering their three-dimensional (3D) complexity. This review outlines the key framework of cardiac organoid imaging research, focusing on pre-imaging sample processing, mainstream imaging technologies, image analysis workflows and applications in disease mechanisms and drug screening. We also provide guidelines for technique selection based on research objectives. Currently, long-term 4D imaging of cardiac organoids is still in its infancy. Future efforts should optimize imaging strategies and advance high-resolution dynamic techniques to deepen understanding of the temporal dynamics of cardiac pathology, providing technical support for cardiovascular research.

Gastric cancer remains a leading cause of cancer-related mortality, with positive peritoneal lavage cytology (PLC) indicating early peritoneal metastasis and poor prognosis. However, conventional imaging and laparoscopy lack sufficient sensitivity for early detection. In this study, six mRNA biomarkers (NOX1, CHEK1, BUB1, NEK2, PRC1, SLCO1B3) were identified through transcriptomic analysis and integrated with clinical features to develop a Risk Stratification Assessment (RSA) model for predicting PLC-positive status. The six-mRNA panel was significantly upregulated in PLC-positive tissues and peripheral blood. The RSA model demonstrated strong diagnostic performance, with AUCs of 0.869 and 0.856 in training and validation cohorts, respectively, and 0.874 and 0.912 in peripheral blood analysis. Moreover, high-risk patients classified by the model had significantly lower 5-year overall survival rates than low-risk patients. This transcriptomics-based RSA model offers a non-invasive, accurate, and clinically meaningful tool for early identification of PLC-positive gastric cancer, potentially surpassing the sensitivity of current diagnostic methods. This study includes prospective analyses from two registered clinical trials: Chinese Clinical Trial Registry, ChiCTR1800014817 (registration date: January 8, 2018; http://www.chictr.org.cn) and ClinicalTrials.gov, NCT03718624 (registration date: October 26, 2018; https://clinicaltrials.gov/ct2/show/NCT03718624).

Bitter taste receptors (TAS2Rs) were initially regarded as sensors for potentially toxic substances, but accumulating evidence indicates that they also play an important role in the regulation of innate immunity. The aim of this study is to summarize the major immunological functions of bitter taste receptors and to clarify their relevance across different physiological systems. Current studies suggest that bitter taste receptors contribute to microbial sensing and immune defense in the respiratory tract, maintain microbial homeostasis and antimicrobial activity in the digestive system, and participate in inflammatory regulation and local immune balance in the musculoskeletal and reproductive systems. In addition, they are involved in immune-related processes in the nervous, circulatory, endocrine, and urinary systems. Collectively, these findings indicate that bitter taste receptors act as widely distributed immune-sensing molecules that link environmental stimuli to innate immune responses, providing a broader framework for understanding multisystem immune defense mechanisms.

Inflammatory bowel disease (IBD) is a chronic, heterogeneous condition characterized by recurrent intestinal inflammation and sustained mucosal barrier damage, profoundly impairing patients’ quality of life and imposing a considerable socioeconomic burden. Current therapeutic options are often constrained by low oral bioavailability, pronounced systemic toxicity, and inadequate tissue specificity, limiting their ability to achieve precise and durable efficacy. In recent years, membrane vesicle-based drug delivery systems (MV-DDSs) have shown considerable promise for precision IBD therapy owing to their excellent biocompatibility, mucosal barrier-penetrating capacity, and low immunogenicity. Building upon a systematic discussion of the roles of MV-DDSs in suppressing inflammatory signaling, modulating oxidative stress, preserving barrier integrity, reshaping the gut microbiota, and regulating programmed cell death, this review further compares the differences in key molecular targets and functional outcomes among vesicles of diverse origins and carrying distinct therapeutic payloads. These insights provide a comprehensive strategic reference and theoretical foundation for the rational design, mechanistic optimization, and clinical translation of MV-DDSs in IBD therapy.

Gastric cancer with peritoneal dissemination remains a significant clinical challenge due to its poor prognosis and difficulty in early detection. This study introduces a multimodal artificial intelligence-based risk stratification assessment (RSA) model, integrating radiomic and clinical data to predict peritoneal lavage cytology-positive (GC-CY1) in gastric cancer patients. The RSA model is trained and validated across retrospective, external, and prospective cohorts. In the training cohort, the RSA model achieved an area under the curve (AUC) of 0.866, outperforming traditional clinical and radiomic feature models. External validation cohorts confirmed its robustness, with AUC values of 0.883 and 0.823 for predicting peritoneal metastasis and recurrence, respectively. In a prospective validation involving 152 patients, the model maintained superior predictive performance (AUC = 0.835). The RSA model also demonstrated significant clinical benefits by effectively identifying high-risk patients likely to benefit from specific treatments, such as paclitaxel-based conversion therapy. These findings suggest that the RSA model offers a reliable, non-invasive diagnostic tool for gastric cancer, capable of improving early detection and treatment outcomes. Further prospective studies are warranted to explore its full clinical potential.

Early postoperative recurrence in locally advanced gastric cancer (LAGC) severely compromises patient outcomes, yet current predictive models are inadequate due to limited generalizability and insufficient validation. This study aimed to establish and robustly validate a multimodal Radiomics-Clinical Integrated Risk Stratification Assessment (RSA) model to accurately predict early recurrence. We retrospectively analyzed 2,516 LAGC patients from six hospitals across northern and southern China, divided into training, internal validation, and external validation cohorts. Radiomics features were extracted from portal venous-phase abdominal CT images using nnU-Net-based segmentation, followed by a hierarchical feature-selection framework integrating the mRMR algorithm and LASSO regression. A predictive RSA model combining clinical factors and radiomics features was developed. Model robustness was validated internally, externally, prospectively (clinical trial cohort, NCT01516944, n = 569), and further independently tested using a publicly available dataset from The Cancer Imaging Archive (TCIA, n = 41). The RSA model exhibited superior performance across all validation cohorts, achieving AUC values of 0.873 (training), 0.871–0.872 (internal validation), 0.870–0.873 (external validation), 0.857 (prospective validation), and 0.850 (TCIA validation). Decision curve analyses confirmed the RSA model provided significant clinical benefits beyond traditional models. Transcriptomic analyses revealed that low-risk patients exhibited enhanced immune infiltration and significant activation of immune-related pathways, including IL6/JAK/STAT3 and interferon signaling. Multivariate Cox regression demonstrated the RSA model independently predicted five-year overall survival across validation cohorts (HR range: 1.830–2.166, all P < 0.001), surpassing standard clinical staging. Our RSA model, integrating robust radiomics methodologies and critical clinical parameters, consistently demonstrated high accuracy and clinical applicability in diverse populations, significantly improving prediction of early postoperative recurrence in LAGC. This approach provides novel insights into tumor-immune microenvironment interactions, paving the way toward personalized postoperative management strategies.

Cold tumors, defined by insufficient immune cell infiltration and a highly immunosuppressive tumor microenvironment (TME), exhibit limited responsiveness to conventional immunotherapies. This review systematically summarizes the mechanisms of immune evasion and the therapeutic strategies for cold tumors as revealed by multi-omics technologies. By integrating genomic, transcriptomic, proteomic, metabolomic, and spatial multi-omics data, the review elucidates key immune evasion mechanisms, including activation of the WNT/β-catenin pathway, transforming growth factor-β (TGF-β)-mediated immunosuppression, metabolic reprogramming (e.g., lactate accumulation), and aberrant expression of immune checkpoint molecules. Furthermore, this review proposes multi-dimensional therapeutic strategies, such as targeting immunosuppressive pathways (e.g., programmed death-1 (PD-1)/programmed death-ligand 1 (PD-L1) inhibitors combined with TGF-β blockade), reshaping the TME through chemokine-based therapies, oncolytic viruses, and vascular normalization, and metabolic interventions (e.g., inhibition of lactate dehydrogenase A (LDHA) or glutaminase (GLS)). In addition, personalized neoantigen vaccines and engineered cell therapies (e.g., T cell receptor-engineered T (TCR-T) and natural killer (NK) cells) show promising potential. Emerging evidence also highlights the role of epigenetic regulation (e.g., histone deacetylase (HDAC) inhibitors) and N6-Methyladenosine (m6A) RNA modifications in reversing immune evasion. Despite the promising insights offered by multi-omics integration in guiding precision immunotherapy, challenges remain in clinical translation, including data heterogeneity, target-specific toxicity, and limitations in preclinical models. Future efforts should focus on coupling dynamic multi-omics technologies with intelligent therapeutic design to convert cold tumors into immunologically active ("hot") microenvironments, ultimately facilitating breakthroughs in personalized immunotherapy.

Knee osteoarthritis (KOA)is an age-related degenerative whole-joint disease characterized by poor outcomes. Wenjing Tongluo Decoction (WJTLD), a Chinese herbal remedy, has demonstrated favorable clinical effects on KOA. However, the precise mechanisms therein remain poorly defined. In this study, we employed the method of anterior cruciate ligament transection (ACLT) method to establish a rat model of KOA. Following 8 weeks of oral administration of WJTLD, the morphology of knee joint cartilage was evaluated using Safranin-O/Fast green staining, H&E staining, and micro-CT imaging. Utilizing GC-MS based untargeted metabolomics and nano-LC-QE-MS based proteomics, we identified altered metabolites and proteins associated with knee cartilage in different rat groups, which were further validated through western blotting and real-time PCR. Our findings indicate that WJTLD alleviates damage to knee joint cartilage and inhibits cartilage degradation. Proteomics data revealed that the altered proteins in OA and WJTLD treated group were related to the biological process including amoebiasis, platelet activation, ECM-receptor interaction, protein digestion and absorption, and ribosome function. Western blotting results confirmed that the expression levels of MMP8 and LDHA were significantly upregulated in the KOA group but were rescued by WJTLD treatment. According to untargeted metabolomics, the intensities of lactic acid, isoleucine, lysine, glutamate, myo-inositol, adenosine, and β-alanine were significantly elevated in the KOA group, however, these metabolites experienced a dramatic following WJTLD treatment. These results suggest that WJTLD exerts a therapeutic effect on KOA by suppressing inflammation and cartilage degradation, as well as regulating multiple pathways related to ECM degradation, amino acid metabolism, and energy metabolism, including glycolysis.

BACKGROUND:Banxia Xiexin Decoction (BXD) has been shown to exert therapeutic effects on Functional dyspepsia (FD). This study aims to investigate the therapeutic mechanisms of BXD in treating FD. METHODS:Network pharmacology was employed to explore the potential targets of BXD in the treatment of FD. Immunoinfiltration analysis assessed immune activation in FD, with the XGBoost machine learning algorithm used to predict the feature importance of key targets. Deep learning and molecular docking were employed to assess the interactions between active compounds and key targets. Finally, an FD mouse model was established, and Western blotting, immunofluorescence, immunohistochemistry, and Enzyme-linked immunosorbent assay were conducted to validate the findings. RESULTS:Through network pharmacology analysis and machine learning predictions, three key active compounds were identified. GO enrichment analysis indicated that the mechanism of BXD primarily involves biological processes related to inflammatory responses. Immunoinfiltration analysis suggested that immune activation in FD may be associated with increased mast cell presence. Seven hub genes were identified through PPI analysis, with STAT3 identified as a key feature in XGBoost predictions of FD. In vivo experiments showed that BXD inhibited p- STAT3, alleviated mast cell infiltration and mucosal barrier damage, and enhanced gastrointestinal motility. CONCLUSION:BXD may alleviate mast cell infiltration and mucosal barrier damage in FD by inhibiting the expression of p-STAT3, thereby exerting its therapeutic effects.

Background:Gastric cancer (GC) remains a leading cause of cancer-related mortality due to its late diagnosis and poor prognosis. Butyrate metabolism (BM) has demonstrated significant roles in tumor biology, but its prognostic implications in GC remain unexplored. We aimed to investigate the effect of butyrate metabolic biomarkers on the prognosis of GC. Methods:We acquired datasets from The Cancer Genome Atlas and Gene Expression Omnibus. Differential BM-related genes (BMGs) were identified using weighted gene co-expression network analysis (WGCNA). Patients were stratified into subtypes, and a prognostic model was constructed using least absolute shrinkage and selection operator (LASSO) regression. Mendelian randomization (MR) analysis was conducted using genetic variants as instrumental variables to establish causal links between BM and GC prognosis. Results:Our model demonstrated robust prognostic accuracy with an area under the receiver operating characteristic (ROC) curve of 0.716. Transcriptomic analysis identified two key BMGs, SMC2 and HSPB1, with significant implications for GC survival. However, MR analysis provided no evidence of a causal association between BM and GC. Conclusions:We identified two butyrate metabolic prognostic genes, namely, structural maintenance of chromosome 2 and heat shock protein beta-1, as the prognostic markers for GC. Furthermore, MR indicated no causal association between the butyrate metabolic pathway and GC.

BackgroundThe increasing incidence of early-stage T1 gastric cancer (GC) underscores the need for accurate preoperative risk stratification of lymph node metastasis (LNM). Current pathological assessments often misclassify patients, leading to unnecessary radical surgeries. MethodsThrough analysis of transcriptomic data from public databases and T1 GC tissues, we identified a 4-mRNA panel (SDS, TESMIN, NEB, and GRB14). We developed and validated a Risk Stratification Assessment (RSA) model combining this panel with clinical features using surgical specimens (training cohort: n = 218; validation cohort: n = 186), gastroscopic biopsies (n = 122), and liquid biopsies (training cohort: n = 147; validation cohort: n = 168). ResultsThe RSA model demonstrated excellent predictive accuracy for LNM in surgical specimens (training AUC = 0.890, validation AUC = 0.878), gastroscopic biopsies (AUC = 0.928), and liquid biopsies (training AUC = 0.873, validation AUC = 0.852). This model significantly reduced overtreatment rates from 83.9 to 44.1% in tissue specimens and from 84.4 to 56.0% in liquid biopsies. The 4-mRNA panel showed specificity for T1 GC compared to other gastrointestinal cancers (P < 0.001). ConclusionsWe developed and validated a novel liquid biopsy-based RSA model that accurately predicts LNM in T1 GC patients. This non-invasive approach could significantly reduce unnecessary surgical interventions and optimize treatment strategies for high-risk T1 GC patients.

Phosphoglycerate kinase 1 (PGK1) is traditionally recognized for its pivotal role in glycolysis. Our findings reveal that PGK1 also functions as a protein kinase phosphorylating valosin-containing protein (VCP) at S746, which subsequently reduces Beclin 1 deubiquitination and impairs autophagy. Inhibition of PGK1 initiates autophagy in T315I-mutant chronic myeloid leukemia (CML) cells, thereby enhancing their sensitivity to first-generation Tyrosine Kinase Inhibitor (TKI) imatinib and third-generation TKI ponatinib. Despite the significant clinical implications, few PGK1-targeting inhibitors have been approved for clinical use to date. Through a comprehensive high-throughput screening of ∼20,000 natural compounds, we identified flavonoid as potent inhibitors of the enzymatic activity of PGK1. Subsequent structural optimization of these flavonoid derivatives led to the development of CPU-216, a compound that binds to the GLU344 and PHE292 residues of PGK1, effectively inhibiting its enzymatic and kinase activity. Notably, CPU-216 induces autophagy via VCP and Beclin 1 in CML-T315I cells, enhancing their responsiveness to TKIs. These discoveries propose a novel therapeutic strategy for T315I-mutant CML, underscoring the potential to develop targeted treatments that leverage the kinase functions of PGK1.

Aim: Lung adenocarcinoma (LUAD), the most prevalent subtype of non-small cell lung cancer (NSCLC), presents significant clinical challenges due to its high mortality and limited therapeutic options. The molecular heterogeneity and the development of therapeutic resistance further complicate treatment, underscoring the need for a more comprehensive understanding of its cellular and molecular characteristics. This study sought to delineate novel cellular subpopulations and molecular subtypes of LUAD, identify critical biomarkers, and explore potential therapeutic targets to enhance treatment efficacy and patient prognosis. Methods: An integrative multi-omics approach was employed to incorporate single-cell RNA sequencing (scRNA-seq), bulk transcriptomic analysis, and genome-wide association study (GWAS) data from multiple LUAD patient cohorts. Advanced computational approaches, including Bayesian deconvolution and machine learning algorithms, were used to comprehensively characterize the tumor microenvironment, classify LUAD subtypes, and develop a robust prognostic model. Results: Our analysis identified eleven distinct cellular subpopulations within LUAD, with epithelial cells predominating and exhibiting high mutation frequencies in Tumor Protein 53 (TP53) and Titin (TTN) genes. Two molecular subtypes of LUAD [consensus subtype (CS)1 and CS2] were identified, each showing distinct immune landscapes and clinical outcomes. The CS2 subtype, characterized by increased immune cell infiltration, demonstrated a more favorable prognosis and higher sensitivity to immunotherapy. Furthermore, a multi-omics-driven machine learning signature (MOMLS) identified ribonucleotide reductase M1 (RRM1) as a critical biomarker associated with chemotherapy response. Based on this model, several potential therapeutic agents targeting different subtypes were proposed. Conclusion: This study presents a comprehensive multi-omics framework for understanding the molecular complexity of LUAD, providing insights into cellular heterogeneity, molecular subtypes, and potential therapeutic targets. Differential sensitivity to immunotherapy among various cellular subpopulations was identified, paving the way for future immunotherapy-focused research.