Military Medical Research最新文献

Target trial emulation (TTE) has demonstrated popularity because of its ability to improve the reliability of causal inference from observational data. Nevertheless, knowledge about the current use, potential challenges, and insights of target trials in oncology is scarce. A total of 90 TTE studies in cancer areas were identified through systematic reviews in PubMed and Embase. Among the 54 applications in cancer treatment, registry databases (44.4%) and overall survival (OS, 63.0%) were predominantly used as data sources and primary endpoints, respectively. Approximately 30 (55.6%) of the included TTE cases were associated with immortal time bias, and 21 (38.9%) were associated with prevalent user bias. Among the 21 trials from 13 studies that aimed to calibrate the results from preexisting randomized controlled trials (RCTs), only 42.9% met both statistical agreement and estimate agreement. The availability of fit-for-purpose data sources and uncertainty about result concordance were the main hurdles limiting the quantity and quality of TTE in oncology areas. Promoting regulatory acceptance by initiating special projects could be crucial for the expanded application of real-world data (RWD) using TTE. Potential solutions, such as the integration of electronic medical records at the regional or country level, linkage with insurance claims databases, the modernization of eligibility criteria, the use of OS as the primary endpoint, and other best practices, were recommended for improving the feasibility and quality of oncology TTE.

Supplementary information: The online version contains supplementary material available at 10.1186/s40779-026-00685-9.

Background: Sepsis and rheumatoid arthritis (RA) are distinct yet mechanistically related conditions commonly driven by dysregulated inflammatory responses. Here, we explored the counterintuitive hypothesis that an epitope from a deleterious anti-tetranectin (TN) antibody (mAb9) could hold unforeseen therapeutic potential.

Methods: By mapping mAb9's epitope to P2 (residues 55-70), a region crucial for TN's protective functions, we developed P2-1, a water-soluble derivative as a targeted therapy. We then employed animal models of sepsis (cecal ligation and puncture) and arthritis (collagen antibody-induced arthritis) to evaluate the therapeutic effects of P2, P2-1, and a procathepsin L (pCTS-L)-neutralizing antibody by assessing septic survival, arthritis severity, pain sensitivity, and joint tissue histology. In parallel, we utilized a surface plasmon resonance (SPR) assay and computational modeling to examine the P2-1/high mobility group box 1 (HMGB1) interaction. Finally, we elucidate the effect of P2-1 on the HMGB1-induced release of pCTS-L and other cytokines and chemokines using primary human peripheral blood mononuclear cells (PBMCs).

Results: P2-1 significantly improved survival and reduced systemic inflammation in a sepsis model, and attenuated arthritis severity and pain sensitivity in an RA model, even with therapeutic administration after disease onset. Mechanistically, P2-1 exhibited high-affinity binding to HMGB1 and selectively suppressed HMGB1-induced cathepsin L (Ctsl) mRNA upregulation and pCTS-L secretion from human immune cells, crucially without perturbing other HMGB1-induced cytokines and chemokines. We further validated pCTS-L as a therapeutic target by demonstrating that a neutralizing antibody conferred potent antiarthritic effects, reducing joint inflammation, pain, and structural damage.

Conclusions: Our findings introduce a paradigm-shifting drug discovery strategy that transforms insights from harmful antibody action into targeted therapeutics for the HMGB1-pCTS-L axis. This approach not only delivers P2-1 as a potent therapy but also establishes pCTS-L as a crucial mediator in inflammatory diseases such as sepsis and RA.

Supplementary information: The online version contains supplementary material available at 10.1186/s40779-026-00686-8.

Artificial intelligence (AI) and big data are reshaping the healthcare landscape. However, clinical value depends on how well systems augment clinicians and fit into routine workflows. To this end, we introduce the TRIAD framework: trustworthy governance, real-world clinical value, and integrated adaptive deployment, to guide the development, validation, and deployment of clinical AI. TRIAD requires explicit data provenance and intended use, fairness auditing, and calibrated uncertainty. This framework evaluates the human-AI team in real workflows using team-level metrics, including accuracy, safety, workload, and patterns of acceptance, editing, and overriding. Deployment proceeds via staged rollouts with preregistered guardrails and continuous monitoring of performance and subgroup impact. TRIAD views intelligence as a property of the human-AI team rather than the AI model alone. Aligning governance, evaluation, and deployment around clinicians and patients enables durable gains in safety, equity, efficiency, and experience, thereby elevating clinical value.

The cytokine storm, a life-threatening systemic inflammatory syndrome, is the primary driver of multiorgan failure in different clinical situations, including severe infections, autoimmune diseases, chimeric antigen receptor (CAR) T cell immunotherapy for cancer, and genetic syndromes. This review focuses primarily on cytokine storms triggered by severe infections such as viral pneumonia and bacterial sepsis, and explores the underlying mechanisms of cytokine storms and potential therapeutic interventions. Cytokine storms are characterized primarily by the excessive release of proinflammatory cytokines, which are triggered by pathogen-associated molecular patterns (PAMPs), damage-associated molecular patterns (DAMPs), and PANoptosis, all of which activate immune signaling cascades. Amplification mechanisms involve positive feedback loops and the failure of negative feedback mechanisms, leading to uncontrolled inflammation. Like a pyrrhic victory, the excessive activation of the immune system eliminated invading pathogens but caused catastrophic damage due to multiple organ dysfunction syndrome (MODS), turning the life-saving response into a life-threatening war. Therapeutic strategies, including cytokine antagonists, Janus kinase (JAK) inhibitors, caspase inhibitors, glucocorticoids, and blood purification therapies, aim to interrupt the self-amplifying cycle of inflammation that propagates organ injury, thereby reducing MODS and mortality. Challenges include optimizing the treatment timing and patient stratification. Future research should focus on combination therapies and personalized medicine based on the heterogeneity of infections and sepsis. Advances in multiomics and targeted therapies provide new hope for managing infections and sepsis.

Background: Despite the predictive impact of circulating tumor DNA (ctDNA) minimal residual disease (MRD), accurate prediction of failure risk after curative-intent treatments for early-stage or localized non-small cell lung cancer (NSCLC) patients to guide personalized therapy remains challenging. This study aimed to develop and validate an interpretable artificial intelligence-assisted model using global data resources.

Methods: Liquid biopsy data, blood-based genomic alterations, clinicopathological features, and survival outcomes of stage I-III NSCLC patients who underwent surgery or definitive chemoradiotherapy were collected from 6 cohorts. PRIME (Progression Risk prediction by Interpretable Machine learning on ctDNA-MRD, Mutations, and clinical-therapeutic features) was trained by 6 machine learning algorithms across 4 cohorts and validated in 2 independent cohorts. Model performance was evaluated by the area under the curve (AUC) and interpreted by SHapley Additive exPlanations (SHAP). Whole-exome sequencing (WES) or whole-genome sequencing (WGS) of tumor tissue from 430 stage II-III NSCLC patients and RNA-sequencing (RNA-seq) data from 1149 subjects, sourced from The Cancer Genome Atlas, were used to validate the prognostic effect of mutations identified in peripheral blood and investigate the underlying mechanisms.

Results: A global dataset encompassing 781 blood samples from 493 patients was analyzed. Clinical stage, pre-treatment ctDNA, post-treatment MRD, blood-based Kelch-like ECH-associated protein 1 (KEAP1), serine/threonine kinase 11 (STK11), and cyclin-dependent kinase inhibitor 2A (CDKN2A) mutations, and treatment modality were significantly associated with the risk of disease progression and were thereby included in the model training. WES/WGS and RNA-seq confirmed the poor prognostic effect of KEAP1, STK11, and CDKN2A mutations, which were characterized by the suppressive tumor microenvironment and attenuated humoral immunity. The neural network (NN) model exhibited optimal prediction of treatment failure risk in the training (AUC = 0.85, 95% CI 0.81-0.89) and validation sets (AUC = 0.82, 95% CI 0.74-0.89). SHAP analysis indicated that MRD (+0.306), treatment modality (+0.128), and pre-treatment ctDNA (+0.043) ranked in the top 3 contributions. NN-PRIME outperformed single liquid biopsy biomarkers and clinical-therapeutic signatures, and demonstrated consistent robustness across different clinical scenarios. High-risk patients identified by NN-PRIME had poorer prognoses but derived significant benefits from adjuvant therapy after surgery.

Conclusions: As an interpretable model integrating readily-accessible and crucial clinical-genomic predictors, PRIME achieves enhanced performance, allowing for early outcome prediction, refined risk stratification, and personalized clinical decision-making.

Artificial intelligence (AI) offers transformative potential in pathology, where histopathological images remain the diagnostic gold standard due to their rich morphological and molecular information. While the rapid development of AI-driven computational pathology tools is revolutionizing disease interpretation, these technologies have not yet been systematically evaluated. Therefore, this review systematically evaluates AI applications across the diagnostic continuum, from image preprocessing and tumor classification to prognostic stratification and the discovery of predictive biomarkers. It presents a technical taxonomy of the algorithms and foundation models powering these applications, benchmarking their performance across diverse diagnostic tasks through rigorous comparative analyses. It also identifies critical challenges in clinical translation, including computational scaling, noisy annotations, interpretability gaps, and domain shifts. Finally, it proposes a roadmap for advancing AI applications in precision oncology and pathological research. By bridging technological innovation with clinical needs, this review aims to accelerate the integration of robust, unified, scalable AI solutions into diagnostic workflows.

Background: In 2022, China experienced an unprecedented heatwave event, raising concerns about the health impacts of heatwaves. This study aims to understand the devastating health risk of the exceptional heatwave in 2022 by comparing heatwave-related mortality burden in 2022 with that during 2000-2021.

Methods: We collected daily mortality and daily maximum temperature (DMT) during 2006-2017 in 364 locations (counties/districts) of China. Heatwave was defined as an event with 2 or more consecutive days of DMT exceeding the 92.5th percentile. We employed a distributed lag nonlinear model (DLNM) and a meta-analysis to examine the heatwave-mortality association based on the data from 364 counties/districts, and then this association was used to assess the mortality burden attributable to heatwaves during 2000-2022 in 368 cities in China. A percentage change (%) indicator, comparing the 2022 mortality burden to the average value from 2000 to 2021, was further calculated to highlight the severity of heatwaves in 2022.

Results: In the past 2 decades, the frequency and intensity of heatwaves in China significantly increased, with the cumulative excessive degree-day increasing to 31,626 in 2022 compared with the annual average value of 13,772 during 2000-2021 across China. In 2022, we observed 62,961 [95% confidence internal (CI) 54,945-70,413] heatwave-related deaths in China, which was much higher than the annual average [35,987 (95% CI 31,252-40,471)] attributable to heatwaves during 2000-2021. The vulnerability groups of heatwave-related mortality in 2022 primarily included patients with cardiovascular diseases [40,567 (95% CI 35,313-45,404)], females [35,876 (95% CI 31,035-41,005)], and people aged over 65 years [56,208 (95% CI 49,023-62,864)]; and greater heatwave-related mortality was found in eastern-central China. The attributable fraction (AF) of heatwave-related deaths increased from an annual average of 11.01‰ (95% CI 9.56-12.38) during 2000-2021 to 18.11‰ (95% CI 15.80-20.25) in 2022 with 64.43% increment (95% CI 38.10-93.78), and the increase rates were greater in Xizang Autonomous Region (159.77%, 95% CI 12.84-477.87) and Sichuan Province (133.64%, 95% CI 3.84-416.61).

Conclusions: This study indicated that the frequency and intensity of heatwaves significantly increased in the past 2 decades in China, and the 2022 heatwaves were linked to a substantial mortality burden in China, with significant population and regional heterogeneity. Our findings underscore the need for developing comprehensive heat adaptation plans in the context of rapid aging and ongoing global warming.

Background: Anemia is a major global health problem. There were 89% of all anemia-related disabilities in developing countries. We aim to analyze the burden of anemia and its underlying causes in China from 1990 to 2023.

Methods: Utilizing the data of the 2023 Global Burden of Disease (GBD 2023) study, this study analyzed the burden of anemia in China between 1990 and 2023. Then we analyzed the number and rate of anemia attributed to 16 underlying causes for all genders and ages. Drivers of change in prevalence and years lived with disability (YLD) numbers due to anemia were explored by decomposition analysis. And locally weighted regression was used to estimate the relationship between socio-demographic index (SDI) and age-standardized prevalence rate (ASPR) and age-standardized YLD rate due to anemia.

Results: From 1990 to 2023, the ASPR and age-standardized YLD rate showed a downward trend among all anemia types (P < 0.05), and the ASPR and age-standardized YLD rate of anemia in females were higher than those in males. The highest number and rate of prevalence were found in mild anemia, and the highest number and rate of YLD were found in moderate anemia. As age increased, the prevalence and YLD rate of anemia increased, with a significant increase in females aged 20-54, in particular of moderate anemia. In 2023, the highest ASPR and age-standardized YLD rate among all anemia types were in the Northwestern regions. Compared to 1990, 31 provinces, Hong Kong, and Macao exhibited declines in both the ASPR and the age-standardized YLD rate for anemia. In China, most of the prevalent cases and YLD were attributable to dietary iron deficiency in 2023. The total prevalence of anemia decreased by 46.14% [95% uncertainty interval (UI) 27.54-61.02], of which age-specific rate, population growth, and population aging accounted for -77.32%, 21.33%, and 9.84%, respectively. A negative association between SDI and the ASPR and age-standardized YLD rate of anemia was shown in China.

Conclusions: From 1990 to 2023, the burden of anemia in China has decreased but remained heavy among women of childbearing age, the elderly, and in the Northwestern region. Tailored prevention and control strategies should be strengthened to reduce the burden of anemia in high-risk areas.

Background: Physiological, pharmacological, and pathological alterations of consciousness provide critical windows into its neural substrates. Given the inherent complexity and multidimensionality of consciousness, defining quantitative, dynamic signatures of neural activity, and translating them into clinically applicable tools remains challenge. This study aimed to build an electroencephalography (EEG)-based methodological guideline for clinical consciousness assessment.

Methods: EEG signals were systematically categorized across periodic and aperiodic activity, connectivity and network topology, spatiotemporal dynamics, self-organized criticality, and transcranial magnetic stimulation (TMS)-evoked responses. These biomarkers were mapped onto a conceptual framework of consciousness, comprising wakefulness and internal/external awareness, based on their validation across clinical conditions. The discriminative efficacy of various biomarkers was then evaluated across 4 independent datasets.

Results: Integrated EEG features each captured distinct yet complementary dimensions of consciousness, supporting a unified neurophysiological architecture underlying diverse alterations of consciousness. Spectral power and peak frequency tracked the loss of consciousness during propofol anesthesia and sleep. Steeper aperiodic slopes, loss of frontoparietal connectivity, disrupted small-world organization, and reduced effective dimensionality were particularly effective in distinguishing minimally conscious state (MCS) from unresponsive wakefulness syndrome (UWS). Additionally, spatiotemporal patterns exhibited consciousness-specific alterations, with both pharmacological and pathological alterations influencing specific microstate dynamics.

Conclusions: Synthesizing integrated neural dynamics and multidimensional consciousness, this guideline establishes both methodological and theoretical foundations for translating neurophysiological biomarkers into clinical applications. While this work advances both conceptual clarity and practical methodology, large-scale validation across expanded clinical cohorts, experimental models, and multimodal platforms is essential to fully establish causal linkages and translational utility.