Background: The development of coronary artery disease (CAD) and plaque destabilization are caused by endothelial dysfunction, inflammation and hypoxia. Recent studies suggest that circulating microRNAs (miRNAs) constitute promising non-invasive biomarkers for monitoring these pathophysiological mechanisms.
Methods: In this exploratory study, we profiled circulating miRNAs in patients with stable versus unstable CAD and investigated the associations with plaque instability, comorbidities and sex-specific modulation.
Results: Different expression patterns were observed. miR-320-3p was nominally reduced in unstable CAD, consistent with its role in endothelial integrity and blood coagulation, but this reduction was not statistically significant after correction. Sex-specific increases in miR-21 isoforms and miR-210 were observed in women with unstable CAD, suggesting hormonal and inflammatory influences. Different associations of miR-103a-5p and miR-146b-5p with metabolic stress, as well as an accumulation of inflammatory, metabolic and hypoxia-responsive miRNAs, underscore the presence of coordinated regulatory networks.
Conclusions: This pilot study identified circulating miRNA signatures associated with plaque instability and sex-dependent vascular modulation. Although these results are preliminary, they provide a mechanistic and translational framework for integrating circulating miRNAs into multimodal risk stratification, thus complementing imaging techniques and conventional biomarkers. Larger longitudinal studies are needed to confirm predictive value and clinical utility.
Chinese Tuina, a fundamental manual therapy technique in traditional Chinese medicine (TCM), treats diseases by applying precisely regulated mechanical forces to the human body. Although its clinical benefits have been extensively documented across a range of conditions—from musculoskeletal pain to neurological disorders—the molecular mechanisms underlying its systemic effects remain partially understood. Meanwhile, exosomes, ubiquitous nanoscale extracellular vesicles, have emerged as key mediators of intercellular communication. These vesicles transfer proteins, lipids and nucleic acids to coordinate bodily homeostasis and responses to stimuli. This review integrates these two fields, proposing that the mechanical forces generated by Chinese Tuina (pressure: 50–300 kPa; shear force: 10–50 dyn/cm2; tension: 5%–20% strain; typical session duration: 15–45 min) constitute an efficient, non-invasive physiological stimulus capable of systemically regulating the production and release of functional exosomes. We systematically dissect the complete pathway initiating from the perception of mechanical forces at the site of manipulation, followed by exosome release from target cells, distribution via the circulatory system, and finally, the multi-dimensional impacts on distant organs. By defining Chinese Tuina as a “mechanical stimulator of the exosome network,” this review aims to establish a novel biological paradigm for manual therapy, providing profound insights into understanding its mechanisms of action and advancing the development of precision rehabilitation medicine.
Sepsis is a complex and life-threatening syndrome resulting from infection and characterized by dysregulated host immune responses. T cells play a central role in orchestrating immune defence against pathogens, yet their function undergoes profound alterations during sepsis. The impact of sepsis on T cell function varies depending on the causative pathogen. T cells exhibit an initial hyperactivation phase in bacterial sepsis, followed by a state of exhaustion, characterised by reduced cytokine production, impaired proliferation, and metabolic dysfunction. Such disorders are associated with disruptions in T cell receptor signalling, upregulation of immune checkpoint molecules such as PD-1 and CTLA4, and mitochondrial damage. T cell dysfunction is often linked to immunosuppression in viral sepsis, with an inhibition of antiviral responses and induction of immune tolerance. The distinct immune evasion strategies impair T cell-mediated pathogen clearance in viral sepsis, while altered T cell subsets are observed in fungal and parasitic sepsis. Such immune dysregulations exacerbate sepsis-induced immune suppression, increase susceptibility to secondary infections, and worsen clinical outcomes. Elucidating the pathogen-specific pathways that underlie T cell dysfunction in sepsis is crucial for the development of precise immunotherapies. These insights could inform the design of therapeutic strategies aimed at restoring T cell function and improving the prognosis of septic patients.
Objective: This review delineates the evolution of attention architectures in automated tumor segmentation across three pivotal paradigms: classic Pre-Transformer attention, dominant Transformer-based self-attention, and emerging Mamba-based state space models.
Methods: We synthesize their functional roles—from channel enhancement to long-range dependency modelingand critically assess hybrid structures designed for multimodal clinical oncology scenarios.
Findings/Results: The review establishes attention mechanisms as a foundational pillar for the next generation of intelligent segmentation tools, highlighting their potential in feature enhancement and localization.
Conclusion: Interrogating current challenges and charting future trajectories, these mechanisms are poised to profoundly impact the landscape of precision medicine.
Background: Astrocytes, far beyond passive support cells, are critical regulators in brain homeostasis, intricately regulating synaptic plasticity, neurovascular integrity, metabolic support, and neuroimmune communication. Their dysregulation is increasingly implicated in the pathophysiology of a wide range of neuropsychiatric and neurodegenerative diseases.
Methods: This review synthesizes fundamental research exploring how astrocytic reactivity disrupts homeostatic functions. It evaluates current mechanisms of astrocyte-mediated circuit dysfunction, specifically focusing on tripartite synapse and the secretion of immunomodulatory factors. The analysis further explores the utility of emerging technologies, including single-cell transcriptomics and in vivo astrocyte-specific manipulation tools.
Results: The synthesis identifies that astrocytic reactivity leads to a profound disruption of the glutamate/GABA balance, contributing to impaired synaptic signaling. A central finding is that altered astrocyte–neuron–microglia crosstalk contributes to circuit dysfunction. Furthermore, A significant challenge remains in the characterization of heterogeneous, context-dependent astrocyte states and their specific contributions to pathogenesis.
Conclusion: A deeper mechanistic understanding of astrocyte biology is paramount for unveiling novel disease aetiologies and unlocking the potential for innovative astrocyte-targeted therapeutic strategies.
Purpose: Sex differences in the prevalence of gliomas have been reported; however, whether there are sex differences in the development and immunotherapy of gliomas remains poorly understood. This study aims to explore sex differences in genomics, immune microenvironment, and immunotherapy response in adult grade 4 gliomas.
Methods: The clinical and pathological data of 238 patients with adult grade 4 gliomas in our cohort, data of a large tumour immunotherapy cohort (a total of 1512 patients), and several glioma-related cohorts (a total of 2272 patients) were collected. To comprehensively explore sex-based differences in grade 4 gliomas, we performed a retrospective analysis, assessed immune-cell infiltration, and conducted a meta-analysis. Further characterisation was achieved through immunohistochemistry, Kaplan–Meier survival analysis, Gene Ontology enrichment and Kyoto Encyclopaedia of Genes and Genomes pathway analyses to investigate underlying biological mechanisms.
Results: We found that female grade 4 patients receiving immune checkpoint inhibitor treatment presented a poorer prognosis than male patients, while those receiving non-immunotherapy exhibited a better prognosis than male patients. Compared to male grade 4 gliomas, female patients had higher levels of tumour-infiltrating monocyte-macrophages but lower levels of circulating monocytes. In addition, there were also differences in genomics between female and male grade 4 gliomas.
Conclusion: There were sex differences in overall survival, immune microenvironment, and genomics in patients with grade 4 gliomas.
Background: The rapid adoption of ChatGPT in healthcare has generated extensive literature. However, systematic analysis of this emerging field using artificial intelligence (AI)-powered tools remains challenging due to the volume and diversity of publications and the risk of AI-generated hallucinations that compromise factual accuracy.
Objective: We developed ChatCM-RAG, a deep learning pipeline integrating BERTopic with transformer-based retrieval-augmented generation to analyse ChatGPT applications in the Medicine literature.
Methods: We processed 904 peer-reviewed articles (2022–2025) using a multi-stage pipeline: BERTopic for topic modelling with UMAP dimensionality reduction and HDBSCAN clustering, Facebook AI Similarity Search for semantic retrieval, and transformer models (T5/GPT) for answer generation. The system was evaluated using representative medical queries across retrieval accuracy, generation quality, and system efficiency metrics.
Results: ChatCM-RAG identified four distinct topic clusters: general medical AI applications (46.2%), performance evaluation (13.9%), clinical applications and patient care (12.6%), and chatbot implementations (6.0%), with 21.2% unclustered documents. To reduce hallucination and ensure citation authenticity, the generation module is constrained to a curated corpus of 904 PubMed-indexed documents and only permits PMID citations that exist in the retrieval set. In our pilot evaluation on eight representative medical queries, we observed 0 fabricated PMIDs and 0 non-existent citations in generated answers. The system achieved an average response time of 1.73 s, an answer quality score of 0.81, and demonstrated topic-aware retrieval with a 0.90 relevance score, where 73% of retrieved documents originated from topically appropriate clusters. The ChatCM-RAG model has been open-sourced at https://huggingface.co/fc28/ChatCM-RAG. Additional reproducible analyses were performed using the released dataset (n = 904) and code to quantify topic interpretability and retrieval robustness without requiring external LLM calls. Using an 80/20 held-out title-query evaluation, topic-filtered lexical retrieval (TF‑IDF within predicted cluster) improved Cluster-agreement@5 from 0.599 ± 0.309 (global TF‑IDF) to 0.862 ± 0.345. Top TF-IDF terms were identified for each cluster, and retrieval performance was assessed using a held-out evaluation.
Conclusions: ChatCM-RAG effectively synthesises large-scale medical literature, revealing that ChatGPT applications in medicine are dominated by exploratory studies with an emerging focus on clinical decision support. The open-source pipeline provides researchers with powerful tools for understanding AI integration in traditional medicine.
Background: Periodontitis is a chronic inflammatory disease initiated by microbial dysbiosis and dysregulated host immune responses, resulting in progressive destruction of the tooth-supporting periodontal tissues. Increasing evidence recognises periodontitis as a systemic inflammatory condition associated with metabolic, cardiovascular, immune-mediated, and neuroinflammatory disorders. In this context, extracellular vesicles (EVs) have emerged as critical mediators of intercellular communication, facilitating molecular crosstalk between periodontal pathogens, immune cells, and resident periodontal tissues.
Objective: This review aims to synthesise current evidence on the biogenesis, microbial and cellular sources, and functional roles of EVs in periodontal inflammation, tissue destruction, and systemic inflammatory interactions, while evaluating their emerging diagnostic and therapeutic potential in periodontology.
Methods: A narrative synthesis of experimental, mechanistic, and translational studies was conducted to examine the role of EVs in periodontal disease. Evidence from studies investigating vesicle-associated molecular cargo, signalling pathways, and their involvement in local periodontal pathology and systemic inflammatory responses was critically integrated.
Key findings: Both bacterial-derived and host-derived EVs transport diverse bioactive cargo, including lipopolysaccharides, cytokines, proteolytic enzymes, and regulatory nucleic acids. These vesicles modulate inflammatory signalling pathways, contribute to extracellular matrix degradation, influence immune dysregulation, and promote alveolar bone remodelling. Emerging evidence suggests that periodontal EVs may disseminate systemically, contributing to chronic low-grade inflammation and influencing pathological processes in distant organs. Vesicles detected in saliva, gingival crevicular fluid, and circulation also demonstrate promise as minimally invasive biomarkers for disease detection and monitoring.
Conclusion: EVs represent key regulators of periodontal pathogenesis and systemic inflammatory crosstalk. Their diagnostic and therapeutic potential offers promising opportunities for precision approaches in periodontology; however, challenges including vesicle heterogeneity, methodological standardisation, and regulatory considerations remain barriers to clinical translation.
Background: Cholangiocarcinoma (CCA) is the second most common liver malignancy after hepatocellular carcinoma and is difficult to treat with high rates of metastasis. The isocitrate dehydrogenase-1 (IDH1) inhibitor ivosidenib has emerged as a promising second-line therapy for CCA patients with IDH1 mutations.
Methods: We conducted a systematic review and meta-analysis to examine ivosidenib's impact on patient survival in advanced-stage CCA patients, following the Preferred Reporting Items for Systematic Reviews and Meta-analyses guidelines.
Results: A total of 317 patients across four studies were analysed in this review, with 256 ivosidenib-treated patients and 61 placebo-treated patients. Almost all patients have intrahepatic CCA. Ivosidenib was found to improve advanced CCA patient outcomes in median progression-free survival (PFS) (Range: 2.7-4.4 months), median overall survival (OS) (Range: 10.3–15 months), objective response rate (ORR) (Range: 2.4%–18.2%), and DCR (Range: 50.0%–63.6%), compared to outcomes in placebo-treated advanced CCA (PFS = 1.4 months, OS = 7.5 months, ORR = 0% and DCR = 27.9%). Using individual-level data and a random effects model, patient survival was calculated with a hazard ratio of 0.29 (95% confidence interval 0.21–0.42, p = 5.6 × 10-12). Common adverse events include nausea, vomiting, and abdominal pain, but QT prolongation was observed in between 5.5% and 9.1% of patients.
Conclusion: Our analysis suggests that pooled ivosidenib-treated patients with advanced CCA possess better PFS and OS compared to placebo-treated patients. However, studies combining ivosidenib with checkpoint blockade immunotherapy drugs or other targeted therapies have yet to be conducted.
Background: The discovery of translatable circular RNAs has fundamentally expanded the functional landscape of non-coding transcripts in colorectal cancer.
Objective: This review aims to synthesize current evidence on circRNA-encoded proteins and micropeptides in colorectal cancer, assess their functional dichotomy as oncogenic drivers or tumor suppressors, and evaluate the therapeutic opportunities and translational barriers in targeting these molecules.
Key Findings: A growing number of circRNA-encoded proteins and micropeptides exert distinct biological activities independent of their linear host genes. Specific subsets function as either oncogenic drivers or tumor suppressors, converging on key pathways such as Hippo-YES-associated protein (Hippo-YAP), Nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB), and Mitogen-activated protein kinase (MAPK) signaling, while also contributing to metabolic reprogramming and therapy resistance. This functional dichotomy establishes a dual therapeutic framework: inhibiting oncogenic circRNA translation products while restoring tumor-suppressive ones.
Conclusion: Significant translational barriers remain, particularly in achieving efficient and tumor-selective in vivo delivery and systematically defining the full circRNA-encoded proteome. Addressing these challenges will require innovations in delivery platforms, functional screening technologies, and mechanistic characterization. circRNA-derived proteins are emerging as uniquely specific therapeutic targets owing to their cancer-restricted expression patterns and non-canonical regulation, positioning them to meaningfully contribute to precision oncology in colorectal cancer.