Citations

Abstract Introduction Large Language Models (LLMs) have proven effective at analyzing unstructured and structured data from electronic health records (EHR). Their reliance on static EHR data means they miss the dynamic, high-resolution physiological signals that are the first indicators of patient decline, creating a fundamental data lag as clinical documentation of an event may occur hours after its onset. This represents a critical gap in the ability to develop truly effective triage tools. Furthermore, while LLMs can synthesize multi-source data for acuity assessment, they exhibit known limitations, including hallucinations that constrain autonomous use in safety-critical settings. In contrast, a Vision-Language Model (VLM) with strong image-grounded clinical reasoning can interpret vital sign trend images, surpassing the limitations of current EHR-based models. This approach provides immediate triage evaluations that lead to earlier detection of patient deterioration. Methods This mixed-methods study deployed a VLM (Gemini 2.5) integrated within the eICU dashboard. We used near-real-time two-hour trends in heart rate, blood pressure, respiratory rate, and oxygen saturation from consecutive patient cases during alerts in hospital floor units monitored by eICU that admit acute medical and surgical patients. To increase the diversity of the sample, we sought routine cases from patients without alerts and emergent cases from patients admitted to the medical, surgical, trauma, and mixed ICUs. A structured context prompt was used to generate a standardized summary including clinical interpretation, data quality and artifact assessment, triage recommendation, and the model’s chain-of-thought reasoning. Results We included patients from Mayo Clinic Rochester, MN, using an 800-patient cohort for prompt development and a 100-patient cohort for validation against clinically adjudicated results. In the validation cohort the VLM achieved 80% accuracy (80/100). An analysis of the 20 incorrect predictions revealed a clear pattern of conservative over-triage: 80% (16/20) of errors were clinically acceptable adjacent-class discrepancies (routine↔urgent or urgent↔emergent). Only 4/20 discrepancies represented a substantive misclassification. Most errors therefore reflected conservative prioritization rather than incorrect assessment. The model demonstrated a perfect safety profile, underestimating patient acuity in 0% of cases, and thus avoiding the most clinically dangerous error. These results confirm the feasibility of VLM-based approach and provide a strong foundation for the proposed work. Conclusion This VLM performs cross-modal reasoning by fusing deep visual perception with broad clinical context. The result is a fully contextualized and explainable triage prioritization, which provides direct decision support and integrates into the clinical workflow, allowing for faster and more confident acute care assessments. This abstract is funded by: none

PURPOSE External beam radiation therapy (EBRT) has gained delayed acceptance as a recommended first-line treatment modality for patients with hepatocellular carcinoma (HCC), given limited evidence that it improves overall survival (OS). We analyzed individual patient data (IPD) from an international cohort to assess OS among patients with HCC treated with EBRT. METHODS We performed a systematic review of publications that assessed EBRT, met prespecified technical standards for HCC, and reported OS (search date December 15, 2022). Corresponding authors were invited to submit IPD for the study. We performed Kaplan-Meier survival analyses to determine OS and restricted mean survival time (RMST) stratified by Barcelona Clinic Liver Cancer (BCLC) stage and treatment status (ie, treatment-naïve and experienced). We performed random effects Cox proportional hazards modeling to assess clinical characteristics associated with OS. RESULTS Data were provided on 4,913 patients treated with EBRT with a median follow-up time of 5.0 years. The median OS was 6.8 years (95% CI, 5.7 to 8.7) for BCLC-0 and 4.6 years (95% CI, 4.1 to 5.1) for BCLC-A. Among treatment-naïve patients, the median OS was not reached (95% CI, 8.6 to not reached) for BCLC-0 and was 5.4 years (95% CI, 4.5 to 6.7) for BCLC-A. In multivariable models, more advanced BCLC stage, higher tumor burden, worse performance status, and Child-Pugh class B or C were associated with a higher risk of mortality. Ablative radiation dose and more recent year of treatment were associated with a reduced risk of death. CONCLUSION To our knowledge, this study represents the largest multinational cohort of patients with HCC treated with EBRT. OS outcomes with EBRT for very early- and early-stage HCC appear to be comparable with resection, thermal ablation, and other ablative locoregional therapies. These data support the inclusion of EBRT in the BCLC HCC clinical decision-making process.

ABSTRACT Background Atrial fibrillation (AF) is common in patients with hypertrophic cardiomyopathy (HCM) and is associated with increased morbidity, heart failure (HF) exacerbations, and adverse long‐term outcomes. Although catheter ablation is an established rhythm‐control strategy, data comparing its long‐term clinical outcomes with anti‐arrhythmic drug (AAD) therapy in patients with HCM remain limited. Objective To compare long‐term clinical outcomes of catheter ablation versus medical therapy with AADs in patients with HCM and AF. Methods We conducted a retrospective cohort study using the TriNetX Research Network, a global federated database of electronic health records. Adult patients with HCM and AF between January 2010 and December 2020 were identified and stratified according to treatment with catheter ablation or AAD therapy. Outcomes included all‐cause mortality, HF exacerbations, AAD‐related adverse events, all‐cause hospitalization, ischemic stroke, repeat ablation, need for cardioversion, need for long‐term AAD use, and atrioventricular nodal ablation. A 1:1 propensity score matching (PSM) analysis was performed to balance baseline characteristics. Kaplan–Meier survival analysis and Cox proportional hazards models were used to estimate hazard ratios (HRs). Results After PSM, 3144 patients were included (1572 in each cohort) with a mean follow‐up of approximately 3.4 years. Catheter ablation was associated with lower all‐cause mortality compared with AAD therapy (9.5% vs. 14.8%; HR: 0.60; 95% CI: 0.49–0.74; p < 0.001). Ablation was also associated with reduced risk of HF exacerbations (38.4% vs. 43.3%; HR: 0.79; 95% CI: 0.71–0.88; p < 0.001) and fewer AAD‐related adverse events (8.4% vs. 11.2%; HR: 0.71; 95% CI: 0.55–0.93; p = 0.011). Rates of all‐cause hospitalization (HR: 0.93; p = 0.11) and ischemic stroke (HR: 0.78; p = 0.10) were similar between groups. Conclusion In this large real‐world cohort of patients with HCM and AF, catheter ablation was associated with lower mortality, reduced HF hospitalizations, and decreased AAD therapy‐related adverse events, as compared with medical therapy. These findings suggest that catheter ablation may provide meaningful clinical benefits in selected patients with HCM and AF, although prospective studies are needed to confirm these observations.

Abstract Honoring a patient’s wishes for end-of-life care can be challenging if the patient loses decision-making capacity and the identified alternate decision-maker will not respect the patient’s wishes. This article discusses how to proceed ethically and legally when the alternate decision-maker and care team disagree about respecting a patient’s end-of-life preferences.

Abstract Background Meningiomas exhibit clinical heterogeneity. Radiotherapy (RT) remains the only adjuvant therapy, but tumor-control is variable, and biomarkers are limited. NRG/RTOG-0539 is the first prospective phase 2 trial to stratify meningioma patients for adjuvant RT based on clinical risk. Here, we apply modern molecular tools to this cohort and identify correlates of RT response. Methods Tumor tissue from 100 RTOG-0539 patients was profiled using DNA methylation arrays, RNA sequencing, and whole-exome sequencing. Recurrence scores, Molecular Groups, gene expression, and copy number alterations were compared across clinical groups and between RT responders and non-responders; non-response was defined as progression or death within 3 years. Results Modern grading criteria, including brain invasion, TERT mutation, CDKN2A/B deletion, and 1p/1q status, would reclassify 10% of tumors and alter treatment group assignment in 7%. Non-responders to RT exhibited more frequent 1p and 14q loss, and more copy number alterations. Transcriptomic and epigenetic profiling revealed immune-related signatures in responders and cell cycle–related pathways in non-responders, several of which overlapped with targets of vorinostat, a histone deacetylase inhibitor previously validated in aggressive meningioma models. The Proliferative Molecular Group was an independent predictor of post-RT recurrence in multivariable analysis, outperforming WHO grade. Conclusion Multi-omic analysis of the NRG/RTOG-0539 cohort shows that updated WHO grading criteria, incorporating molecular and cytogenetic features, improve risk stratification. However, molecular classification, particularly the Proliferative group, remains an independent and stronger predictor of RT response. These findings support integrating molecular biomarkers alongside modern grading frameworks to guide treatment and trial design in meningioma. Clinical Trial Information NCT00895622.

Introduction Female distance runners are at elevated risk for impaired bone remodeling due to high mechanical loading, potential low energy availability, and sustained inflammatory stress. Collagen peptide (CP) supplementation has been proposed as a nutritional strategy to support type I collagen synthesis and modulate osteoimmune signaling; however, evidence in premenopausal endurance athletes is limited. This pilot randomized, double-blind, placebo-controlled trial examined the effects of four weeks of high-dose CP supplementation on markers of bone metabolism and inflammatory activity in endurance-trained premenopausal women. Methods Twenty-two participants (18–35 years; ≥35 miles/week [&gt;56km/week]) were randomized to CP (INT; 20 g/day) or isocaloric maltodextrin (CON). Pre- and post-intervention assessments, conducted in the early follicular phase, included serum procollagen type I N-terminal propeptide (P1NP), plasma C-terminal telopeptide of type I collagen (CTX-1), serum soluble receptor activator of nuclear factor-κB ligand (sRANKL), osteoprotegerin (OPG), the sRANKL/OPG ratio, and interleukin-6 (IL-6). Repeated-measures ANCOVAs were performed, adjusting for accumulated running distance and vitamin D status. Results A significant group × time interaction was observed for P1NP ( p = 0.04), with increases in INT and no change in CON. No significant interaction was observed for CTX-1 ( p = 0.13). Significant interactions were also observed for sRANKL ( p = 0.046) and IL-6 ( p = 0.03). No significant interaction effects were detected for sRANKL, OPG, or the sRANKL/OPG ratio. Discussion Short-term CP supplementation increased a marker of bone formation, altered osteoclast-related signaling, and reduced IL-6 in endurance-trained premenopausal women. These findings support the potential for CP-mediated modulation of bone turnover and inflammatory activity and warrant further investigation in larger, adequately powered trials incorporating structural bone outcomes.

PURPOSE The Lynch syndrome (LS) genes, MLH1 , MSH2 , MSH6 , and PMS2 , are considered to be some of the highest-priority genes to be tested for population genomic screening, with the aim to prevent and detect cancers early in those with a higher risk for such diagnoses. The objective of this study was to determine the risk of cancer associated with pathogenic variants in MLH1 , MSH2 , MSH6 , or PMS2 in the general population to build evidence-based guidelines for population genomic screening METHODS This retrospective case-control study uses Helix Research Network data from 230,159 participants across nine US health systems. Variant interpretation was performed following the American College of Medical Genetics and Association for Molecular Pathology guidelines. Clinical diagnoses were identified from electronic health records and claims data for 11 cancer types associated with LS, including colorectal and endometrial cancers. RESULTS Individuals with pathogenic variants in MLH1 , MSH2 , and MSH6 showed substantial risk for LS-related cancers compared with those without pathogenic variants. Conversely, individuals with pathogenic variants in PMS2 had no increase in risk for all 11 LS-associated cancers with hazard ratio (HR) of 1.0 (95% CI, 0.6 to 1.8) compared with those without a pathogenic variant. There was a small increase in risk when restricting the analysis to colorectal cancer with an HR of 3.5 (1.7-7.4). This increase in colorectal cancer risk was predominantly observed after age 60 years, 10 years after US Preventive Services Task Force guidelines recommend starting colonoscopies for the average population. Up to age 60 years, 2.6% of individuals with PMS2 pathogenic variants were diagnosed with colorectal cancer, closer to the general population (0.7%) than those with MLH1 (37.3%), MSH2 (28.9%), or MSH6 (6.7%) pathogenic variants. For 100,000 individuals screened for PMS2 , 854 additional colonoscopies would be performed, which would prevent or detect early 3.2 colorectal cancers. CONCLUSION This analysis across nine health systems highlights the clinical phenotypes of PMS2 pathogenic germline variants in the entire population.

Background Pancreatic ductal adenocarcinoma (PDAC), the most common malignant type of pancreatic cancer, is characterized by a dense desmoplastic stroma, low neoantigen burden, and a highly immunosuppressive tumor microenvironment (TME). Methods We present a novel strategy harnessing acute transplant rejection mechanisms by employing a recombinant oncolytic rVMG vector engineered to express murine alloantigens H-2Kb (rVMG-H-2Kb) or H-2Kk (rVMG-H-2Kk), thereby inducing tumor-specific antigenic mismatch responses. Results In vitro, rVMG-H-2Kb and rVMG-H-2Kk exhibited strong replication and cytolytic activity while inducing cell surface expression of both H-2Kk and endogenous H-2Kb, in addition to upregulation of antigen presentation genes (β2-microglobulin, Tap1, and Tapbp). In two immunocompetent PDAC models, intratumoral and systemic delivery of rVMG-H-2Kb and rVMG-H-2Kk delayed tumor progression, with rVMG-H-2Kk conferring a survival advantage. Multiplex immunohistochemistry and immunophenotyping revealed substantial TME remodeling, marked by increased effector T-cell infiltration, regulatory T-cell depletion, and reduced fibrosis. Spatial transcriptomics further showed compartment-specific immune activation and epithelial metabolic reprogramming, corroborating with enhanced tumor immunogenicity. Despite these effects, rVMG-H-2Kk also induced compensatory immunosuppressive pathways, including upregulation of antiviral response genes and immune checkpoint receptors such as programmed death-ligand-2. Importantly, combination therapy with rVMG-H-2Kk and murine checkpoint blockade (anti-programmed cell death protein-1 and anti-cytotoxic T-lymphocyte-associated protein 4) drastically improved survival compared to checkpoint blockade alone. Strikingly, surviving mice resisted tumor rechallenge, indicating the establishment of durable antitumor memory. Conclusion These findings establish rVMG-mediated alloantigen delivery as a novel immunotherapeutic platform capable of converting immune-cold tumors into immune-hot, sensitizing tumors to immune checkpoint inhibitors, and establishing durable antitumor immunity in PDAC.

Abstract Inflammatory bowel disease (IBD) is a chronic immune-mediated disorder of the gastrointestinal tract whose genetic basis is only partly resolved because most risk variants identified by genome-wide association studies (GWAS) lie in non-coding regions, limiting direct gene assignment and biological interpretation 1,2 . Here we analysed whole-exome and whole-genome sequencing data from 86,213 IBD cases and 478,363 controls of European ancestry. We identified 68 IBD genes directly implicated by conditionally independent protein-coding associations across the allele frequency spectrum. Many newly implicated IBD genes are supported by orthogonal genomic or pleiotropic evidence, pointing to disease-related pathways and nominating targets with therapeutic relevance. We further identified allelic series and non-additive effects at key loci such as NOD2 and TYK2. These results show that large-scale sequencing can resolve disease genes and pathways that remain ambiguous from non-coding association alone, providing a more direct route from human genetics to biological insight and therapeutic hypotheses.