Alzheimer’s disease (AD) is the leading cause of dementia worldwide, characterized by progressive cognitive decline and neurodegeneration. Although amyloid-β plaques and hyperphosphorylated tau are major pathological hallmarks, increasing evidence highlights persistent neuroinflammation as a critical contributor to disease progression. Activated microglia and astrocytes, inflammatory cytokines, inflammasome activation, and dysregulated immune responses promote neuronal damage and synaptic dysfunction. Similar inflammatory mechanisms are observed in neurological disorders such as multiple sclerosis (MS) and brain tumors, suggesting shared molecular pathways linking neuroinflammation, neurodegeneration, and tumor-associated immune responses. Proteomics, transcriptomics, and multi-omics approaches provide powerful strategies to characterize disease-associated molecular signatures, signaling pathways, biomarkers, and therapeutic targets. This review summarizes key neuroinflammatory mechanisms in AD, including NF-κB, NLRP3, PI3K/Akt, MAPK, cGAS–STING, mitochondrial, lysosomal, calcium, insulin, and blood–brain barrier pathways. Recent advances in proteomic and transcriptomic profiling and their integration with genomic, epigenomic, and metabolomic data are discussed for understanding disease heterogeneity and identifying clinically relevant biomarkers. Current and emerging therapeutic strategies, including anti-Aβ antibodies, anti-inflammatory agents, selective inhibitors, dual-target compounds, PROTACs, and protein–protein interaction modulators, are also evaluated. Finally, multi-omics-based precision medicine and artificial intelligence-assisted analysis may facilitate biomarker-guided patient stratification and development of effective disease-modifying therapies.
Extrachromosomal DNA (ecDNA) is a structurally distinct form of focal amplification that can increase oncogene dosage, reorganize enhancer–promoter interactions, and accelerate tumor evolution through unequal segregation. This review presents a single, evidence-based sequence: the structure of ecDNA confers an evolutionary capacity; selection then acts upon the genes and regulatory elements that are carried by ecDNA; the clones that have been selected may subsequently develop treatment-adaptive or immune-associated traits; and clinical application can only proceed once it has been demonstrated that ecDNA provides information that goes beyond what is seen in conventional amplification. The strongest evidence comes from unequal segregation, regulatory rewiring in specific models, reversible copy-number adaptation during targeted therapies, and replication-stress vulnerabilities in preclinical systems. In contrast, associations with reduced antigen presentation, altered interferon signaling, immune-cell exclusion, chemotherapy resistance, radiotherapy response, and immunotherapy failure remain context-dependent and not universally causal. When discussing the cGAS–STING pathway, intact nuclear ecDNA is differentiated from micronuclear DNA and damaged cytosolic fragments. Therapeutic strategies are classified based on direct preclinical evidence for ecDNA specificity, ecDNA-associated evidence, indirect biological reasoning, or a conceptual hypothesis; no strategy is presented as an established treatment directed at ecDNA. Detection methods are arranged from amplification screening to breakpoint-resolved inference to independent physical confirmation. Clinically useful driver biomarkers are separated from candidate ecDNA biomarkers that require prospective validation. The main conclusion is that ecDNA is an established driver of tumor evolution but has not yet been broadly validated as an immune-evasion mechanism, a predictive biomarker, or a routine therapeutic target.
Endothelial cell-specific molecule-1 (ESM1, also known as endocan), is a sulfated proteoglycan primarily secreted by vascular endothelial cells, although it can also be expressed by tumor cells and stromal cells. Since its overexpression in multiple tumors was systematically confirmed in 2006, ESM1 has become a research hotspot as a tumor biomarker and therapeutic target. This review systematically elaborates on the role of ESM1 in tumors at three levels. First, we analyze the multilayered regulatory mechanisms underlying its upregulation, including transcription factor activation, growth factor and inflammatory signaling, epigenetic modifications, and non-coding RNA regulation, as well as induction by the metabolic microenvironment and pathogen infection. Second, we clarify the dual biological functions of ESM1. In most tumors, ESM1 drives malignant progression through five major dimensions: angiogenesis, malignant behavior and epithelial–mesenchymal transition, metabolic reprogramming, therapeutic resistance, and immune remodeling. Limited evidence suggests that ESM1 may exert antitumor effects in specific contexts, such as prostate cancer, gastric cancer, colon cancer, and non-glycosylated forms of ESM1; however, such evidence is far less abundant than that supporting its pro-tumorigenic role, and its determinants: tumor type, cell source, glycosylation status, disease stage, and experimental model remain unclear. Finally, we evaluate its clinical translational value. As a diagnostic biomarker and prognostic indicator, ESM1 can be detected in various body fluids and is associated with tumor invasion, metastasis, and survival. However, current diagnostic and prognostic evidence is largely retrospective, single-center, and based on non-standardized detection methods. Circulating ESM1 is also confounded by infection, inflammation, tissue damage and bone marrow regeneration, and its cross-cancer diagnostic value has not yet been confirmed. As a therapeutic target, anti-ESM1 strategies show promising prospects for reversing drug resistance and combination therapy, but prospective clinical validation, safety assessment, and evidence for the feasibility of tumor-selective targeting are still lacking. In addition, ESM1 of endothelial origin and ESM1 of tumor or stromal origin may differ in functional interpretation and biomarker interpretation, yet most existing studies have not rigorously distinguished between these sources. In summary, ESM1 is a key molecular node integrating tumor angiogenesis, immune evasion, and therapeutic resistance, and its clinical translational potential as a circulating biomarker and targeted therapy warrants further exploration.
Boron neutron capture therapy (BNCT) is a binary, tumor-selective radiotherapy modality that combines preferential delivery of boron-10 to tumor tissue with neutron irradiation to generate high-linear energy transfer particles in situ. The field has undergone a defining transition from reactor-based neutron sources to accelerator-based BNCT (AB-BNCT), enabling hospital-based deployment, broader clinical translation, and renewed global interest in indications with substantial unmet needs, particularly recurrent head and neck cancers and gliomas. However, AB-BNCT is intrinsically a drug–device regimen, requiring coordinated evaluation of the boron agent, neutron irradiation system, and dedicated treatment planning software, supported by end-to-end quality management, commissioning and routine quality assurance, and postmarket surveillance. This review summarizes the system-engineering foundations and current global landscape of AB-BNCT. We outline the therapeutic principle and key subsystems, including accelerator neutron sources, beam shaping assembly, patient positioning and setup, beam/dose monitoring, treatment control, and Monte Carlo–based BNCT treatment planning. We synthesize clinical evidence from reactor-era feasibility studies, accelerator-era registration programs, and emerging real-world evidence, including the Japanese Phase II registration study reporting an objective response rate of 71.4% in patients with unresectable recurrent or locally advanced head and neck cancer and subsequent nationwide postmarketing evidence. We also compare regulatory classification and approval pathways across the European Union, United States, Japan, and China. Finally, we discuss regulatory considerations relevant to the evaluation of BNCT devices in China. Continued advances in boron agents, accelerator engineering, and standardization, together with accumulated regulatory experience, may support broader clinical implementation of AB-BNCT. Further accumulation of evidence integrating beam physics, dosimetry, and clinical outcomes may facilitate cross-jurisdictional understanding and the continued clinical development of AB-BNCT.
ObjectiveThis meta-analysis investigated the effect of virtual reality–assisted relaxation therapy (VR-ART) on the reduction of psychological symptoms and fatigue in breast cancer survivors.MethodsThis review followed PRISMA 2020 reporting guidelines. We searched 9 bibliographic databases, plus ClinicalTrials.gov, WHO ICTRP, and GreyGuide, for records published from database inception through September 2026, limited to Chinese- and English-language publications. Backward and forward citation chasing of included studies and related reviews was completed via Web of Science to identify additional eligible records. Parallel-group RCTs evaluating VR-assisted relaxation therapy on psychological symptoms (anxiety, depression, psychological distress) and cancer-related fatigue among breast cancer survivors were included. Two reviewers independently assessed trial risk of bias using the Cochrane RoB 2 tool.ResultsTen RCTs with 995 breast cancer survivors were included. Pooled estimates showed VR-ART yielded statistically significant reductions in anxiety (SMD = −1.94, 95% CI [−2.49, −1.40]; P < 0.001), depressive symptoms (SMD = −1.11, 95% CI [−1.36, −0.87]; P < 0.001), and psychological distress (MD = −1.33, 95% CI [−1.99, −0.67]; P = 0.005) versus controls. effect sizes varied across treatment stages; heterogeneity for anxiety and psychological distress disappeared among trials in active treatment stages; heterogeneity for anxiety and psychological distress disappeared among trials in active treatment stages (anxiety: P = 0.91, I² = 0%; psychological distress: P = 0.79, I² = 0%). Combining sensitivity analysis and existing research, we hypothesized that the combination of chemotherapy regimen and VR interactivity might be associated with variability in effect sizes for psychological distress (P = 0.96, I² = 0%); this observation was exploratory and hypothesis-generating only. Fatigue was synthesized qualitatively rather than pooled quantitatively due to high heterogeneity, with studies reporting reduced self-reported fatigue.ConclusionVR-ART was associated with statistically significant reductions in anxiety, depression, psychological distress, and cancer-related fatigue among breast cancer survivors. Based on findings from exploratory subgroup analyses, we speculated that intervention timing, VR interactivity, and concurrent chemotherapy status might influence the magnitude of intervention effects. Formal testing remains absent, and further research is required to develop individualized intervention strategies and optimize treatment responses.Systematic review registrationhttps://www.crd.york.ac.uk/PROSPERO/, identifier CRD420251114925.
