Citations

e13656 Background: Evidence-based medicine is crucial for clinical decision-making, yet studies suggest that a significant proportion of treatment decisions do not fully incorporate the latest evidence. Large Language Models (LLMs) show promise in bridging this gap, but their reliability for medical recommendations remains uncertain. Methods: We conducted an evaluation study comparing five LLMs’ recommendations across 50 clinical scenarios related to multiple myeloma diagnosis, staging, treatment, and management. The evaluation included three general-purpose LLMs (OpenAI o1-preview, Claude 3.5 Sonnet, Gemini 1.5 Pro), one retrieval-augmented generation (RAG) system (Myelo), and one agentic workflow-based system (HopeAI). General-purpose LLMs generated responses based solely on their internal knowledge, while the RAG system enhanced these capabilities by incorporating external knowledge retrieval. The agentic workflow system extended the RAG approach by implementing multi-step reasoning and coordinating with multiple tools and external systems for complex task execution. Three independent hematologist-oncologists evaluated the LLM-generated responses using standardized scoring criteria developed specifically for this study. Performance assessment encompassed five dimensions: accuracy, relevance, comprehensiveness, hallucination rate, and clinical use readiness. Results: HopeAI demonstrated superior performance across accuracy (82.0%), relevance (85.3%), and comprehensiveness (74.0%), compared to OpenAI o1-preview (64.7%, 57.3%, 36.0%), Claude 3.5 Sonnet (50.0%, 51.3%, 29.3%), Gemini 1.5 Pro (48.0%, 46.0%, 30.0%), and Myelo (58.7%, 56%, 32.7%). Hallucination rates were consistently low across all systems: HopeAI (5.3%), OpenAI o1-preview (3.3%), Claude 3.5 Sonnet (10.0%), Gemini 1.5 Pro (8.0%), and Myelo (5.3%). Clinical use readiness scores were relatively low for all systems: HopeAI (25.3%), OpenAI o1-preview (6.0%), Claude 3.5 Sonnet (2.7%), Gemini 1.5 Pro (4.0%), and Myelo (4.0%). Conclusions: This study demonstrates that while current LLMs show promise in medical decision support, their recommendations require careful clinical supervision to ensure patient safety and optimal care. Further research is needed to improve their clinical use readiness before integration into oncology workflows. These findings provide valuable insights into the capabilities and limitations of LLMs in oncology, guiding future research and development efforts toward integrating AI into clinical workflows.

Rituximab (RTX) is a humanized chimeric anti-CD20 monoclonal antibody that leads to immunosuppression through rapid depletion of B lymphocytes. It has been demonstrated to be useful in treating both thyroid eye disease (TED) and myasthenia gravis (MG), respectively. However, the effectiveness of RTX in concurrent TED and MG cases is unclear and not previously reported in the literature. In this study, we describe a 13-year-old girl who presented with a 10-year history of general MG and a 1-year history of Graves’ disease, bilateral proptosis, eye motility restriction, and lagophthalmos. A comprehensive evaluation confirmed the diagnosis of concurrent TED and MG. Satisfactory effectiveness was observed after RTX treatment without side effects. Based on the described observations, we suggest that RTX should be further explored as a treatment option for patients with concurrent TED and MG.

ABSTRACTBackgroundDeep learning (DL) models for accurate renal tumor characterization may benefit from multi‐center datasets for improved generalizability; however, data‐sharing constraints necessitate privacy‐preserving solutions like federated learning (FL).PurposeTo assess the performance and reliability of FL for renal tumor segmentation and classification in multi‐institutional MRI datasets.Study TypeRetrospective multi‐center study.PopulationA total of 987 patients (403 female) from six hospitals were included for analysis. 73% (723/987) had malignant renal tumors, primarily clear cell carcinoma (n = 509). Patients were split into training (n = 785), validation (n = 104), and test (n = 99) sets, stratified across three simulated institutions.Field Strength/SequenceMRI was performed at 1.5 T and 3 T using T2‐weighted imaging (T2WI) and contrast‐enhanced T1‐weighted imaging (CE‐T1WI) sequences.AssessmentFL and non‐FL approaches used nnU‐Net for tumor segmentation and ResNet for its classification. FL‐trained models across three simulated institutional clients with central weight aggregation, while the non‐FL approach used centralized training on the full dataset.Statistical TestsSegmentation was evaluated using Dice coefficients, and classification between malignant and benign lesions was assessed using accuracy, sensitivity, specificity, and area under the curves (AUCs). FL and non‐FL performance was compared using the Wilcoxon test for segmentation Dice and Delong's test for AUC (p < 0.05).ResultsNo significant difference was observed between FL and non‐FL models in segmentation (Dice: 0.43 vs. 0.45, p = 0.202) or classification (AUC: 0.69 vs. 0.64, p = 0.959) on the test set. For classification, no significant difference was observed between the models in accuracy (p = 0.912), sensitivity (p = 0.862), or specificity (p = 0.847) on the test set.Data ConclusionFL demonstrated comparable performance to non‐FL approaches in renal tumor segmentation and classification, supporting its potential as a privacy‐preserving alternative for multi‐institutional DL models.Evidence Level4.Technical EfficacyStage 2.

ImportanceIntraocular pressure (IOP) reduction with phacogoniotomy (phacoemulsification plus goniosynechialysis plus goniotomy) was not less than that of phacotrabeculectomy for advanced primary angle-closure glaucoma (PACG) with cataract at 1-year follow-up, but longer-term outcomes are needed.ObjectiveTo investigate if phacogoniotomy is noninferior to phacotrabeculectomy for advanced PACG with cataract at 2 years.Design, Setting, and ParticipantsThis multicenter, noninferiority, randomized clinical trial took place in 7 ophthalmology centers in China. The trial started May 31, 2021, and 2-year follow-up ended May 31, 2024. Included in this analysis were patients with advanced PACG and cataract. Study data were analyzed from September 2024 to January 2025.InterventionsRandom assignment (1:1) to phacogoniotomy or phacotrabeculectomy.Main Outcomes and MeasuresThe primary outcome measure was reduction in IOP from baseline to the 2-year visit with a noninferiority margin of 4 mm Hg.ResultsA total of 124 participants (124 eyes) were randomized (mean [SD] age, 66.4 [8.6] years; 67 female [54.0%]), 65 (52.4%) to the phacogoniotomy group and 59 (47.6%) to the phacotrabeculectomy group. A total of 59 patients (90.7%) in the phacogoniotomy group and 52 patients (88.1%) in the phacotrabeculectomy group completed 2-year visits. All participants were Chinese. Mean (SD) IOP reduction was −25.6 (10.2) mm Hg and −24.7 (9.4) mm Hg in the phacogoniotomy and phacotrabeculectomy groups, respectively, and the upper boundary of the CI for difference in change between groups was lower than the 4-mm Hg noninferiority margin (mean difference, −0.5 mm Hg; 97.5% CI, −1.7 mm Hg to 0.8 mm Hg; P = .42). The mean difference for complete success for phacogoniotomy vs phacotrabeculectomy was −6.7% (95% CI, −21.4% to 8.8%; P = .47) and for qualified success was 1.4% (95% CI, −11.0% to 14.3%, P = .30). Median (IQR) number of antiglaucomatous medication was 0 (0) vs 0 (0; Hodges-Lehmann estimate of location shift, 0; 95% CI, 0; P =.12) with phacogoniotomy vs phacotrabeculectomy, respectively (mean difference, 0.13; 95% CI, −0.36 to 0.63; P = .60).Conclusions and RelevanceMean IOP reduction with phacogoniotomy was noninferior to phacotrabeculectomy for advanced PACG and cataract at 2-year follow-up with no differences detected in complete or qualified success or mean number of antiglaucomatous medications. These findings support phacogoniotomy as an alternative to phacotrabeculectomy for patients with advanced PACG and cataract.Trial registrationClinicalTrials.gov Identifier: NCT04878458

Background/Purpose APS ACTION Registry was created to study the natural course of antiphospholipid syndrome (APS) over 10 years in persistently antiphospholipid antibody (aPL) positive patients with or without systemic autoimmune rheumatic diseases (SARDs). Our primary objective was to compare the characteristics of aPL-positive patients with or without thrombocytopenia (TP) and/or autoimmune hemolytic anemia (AIHA). Methods The registry inclusion criteria are positive aPL based on the Revised Sapporo APS Classification Criteria, tested at least twice within 1 year prior to enrollment. For the primary comparison of demographic, clinical, and serologic characteristics in this retrospective study, we divided patients into two groups: TP/AIHA ever and never. Thrombocytopenia was defined as a platelet count of <100,000 x 10 9 /L tested twice at least 12 weeks apart, and AIHA was defined as anemia with hemolysis and a positive direct antiglobulin test (DAT). For the secondary analysis, we compared patients with TP versus AIHA, and the immunosuppressive use stratified by systemic lupus erythematosus (SLE) classification. Results As of April 2022, of 1,039 patients (primary aPL/APS: 618 [59%]; SLE classification: 334 [31%]) included in the registry, 228 (22%) had baseline (historical or current) TP and/or AIHA (TP only: 176 [17%]; AIHA only: 35 [3%], and both: 17 [2%]). Thrombocytopenia and/or AIHA was significantly associated with Asian race, SLE classification, cardiac valve disease, catastrophic/microvascular APS, triple aPL (lupus anticoagulant, anticardiolipin antibody, and anti-β 2 -glycoprotein-I antibody) positivity, and SLE-related serologic and inflammatory markers. When 101/618 (16%) primary aPL/APS patients and 101/334 (34%) SLE patients with TP and/or AIHA were compared, azathioprine and mycophenolate mofetil were more commonly reported in lupus patients, however corticosteroid, intravenous immunoglobulin, and rituximab use were similar between groups. Conclusion In our large multi-center international cohort of persistently aPL-positive patients, approximately one-fifth had active or historical TP and/or AIHA at registry entry; half of these patients had additional SLE. Cardiac valve disease, catastrophic/microvascular APS, and triple aPL-positivity were aPL-related clinical and laboratory manifestations associated with TP and/or AIHA, suggesting a more severe APS clinical phenotype in aPL-patients with TP and/or AIHA.

Hypertension is the leading modifiable risk factor for atrial fibrillation (AF) and is estimated to be present in >70% of AF patients. This Frontiers Review was prepared by 29 expert members of the AF-SCREEN International Collaboration to summarize existing evidence and knowledge gaps on links between hypertension, AF, and their cardiovascular sequelae; simultaneous screening for hypertension and AF; and the prevention of AF through antihypertensive therapy. Hypertension and AF are inextricably connected. Both are easily diagnosed, often silent, and frequently treated inadequately. Together, they additively increase the risk of ischemic stroke, heart failure, and many types of dementia, resulting in greater all-cause mortality, considerable disease burden, and increased health care expenditures. Automated upper arm cuff blood pressure devices with implemented technology can be used to simultaneously detect both hypertension and AF. However, positive screening for AF with an oscillometric blood pressure monitor still requires ECG confirmation. The current evidence suggests that high-risk individuals aged ≥65 years or with treatment-resistant hypertension could benefit from AF screening. Since antihypertensive therapy effectively lowers AF risk, particularly in individuals with left ventricular dysfunction, hypertension should be the key target for AF prediction and prevention rather than merely a comorbidity of AF. Nevertheless, several important gaps in knowledge need to be filled over the next years, including the ideal method and selection of patients for simultaneous screening of hypertension and AF and the optimal antihypertensive drug class and blood pressure targets for AF prevention.

ObjectiveThe objective was to determine the mortality rate as well as the causes and predictors of death in antiphospholipid antibody (aPL)–positive patients with and without antiphospholipid syndrome (APS) classification.MethodsThe inclusion criterion for the multicenter international Antiphospholipid Syndrome Alliance for Clinical Trials and International Networking (APS ACTION) registry is positive aPLs according to the Revised Sapporo Classification Criteria tested within one year before enrollment. Patients are observed every 12 ± 3 months with clinical data and blood collection. For this prospective analysis, we first analyzed the causes of death for patients reported as “deceased.” Secondly, we analyzed risk factors for death using the adjusted Cox proportional hazards model and calculated survival probability using the Kaplan‐Meier model based on different age groups.ResultsOf 967 patients, 43 (5%) were deceased after a median follow‐up of 5.3 years. Based on the univariate analysis, deceased patients, compared to living patients, were more likely to be older and have a history of arterial thrombosis, catastrophic APS, concomitant systemic autoimmune diseases (SAIDs), and baseline cardiovascular disease (CVD) risk factors. Based on the Cox proportional hazards model adjusted for age and for each of the strongest predictors of death, arterial thrombosis (hazard ratio [HR] 2.94, 95% confidence interval [CI] 1.50–5.76), concomitant SAIDs (HR 2.97, 95% 1.56–5.63), and baseline any CVD risk factor (HR 2.43, 95% CI 1.05–5.71) were significantly associated with mortality.ConclusionIn our cohort of persistently aPL‐positive patients, the mortality rate was 5% after a median follow‐up of five years and was highest for patients ≥60 years old at registry entry. History of arterial thrombosis, concomitant SAIDs, and baseline any CVD risk factor independently predicted future death.

ABSTRACTType 2 diabetes mellitus (T2DM), characterized by insulin resistance, is closely associated with Alzheimer's disease (AD). Cerebrovascular dysfunction is manifested in both T2DM and AD, and is often considered as a pathological link between the two diseases. Insulin signaling regulates critical functions of the blood–brain barrier (BBB), and endothelial insulin resistance could lead to BBB dysfunction, aggravating AD pathology. However, insulin signaling is intrinsically dynamic and involves interactions among numerous molecular mediators. Hence, a mechanistic systems biology model is needed to understand how insulin regulates BBB physiology and the consequences of its impairment in T2DM and AD. In this study, we investigated the pharmacodynamic effect of insulin on the expression of vascular cell adhesion molecule 1 (VCAM1), a marker of cerebrovascular inflammation. Intriguingly, normal insulin concentrations selectively activated the PI3K‐AKT pathway, leading to decreased VCAM1 expression. However, exposure to supraphysiological insulin levels, which is present in insulin resistance, activated both PI3K‐AKT and MEK–ERK pathways, and increased VCAM1 expression. We developed a mathematical model that adequately described the dynamics of various insulin signaling nodes and VCAM1 expression. Further, the model was integrated with in vitro proteomics and transcriptomics data from AD patients to simulate VCAM1 expression under pathological conditions. This approach allowed us to establish a quantitative systems pharmacology framework to investigate BBB dysfunction in AD and metabolic syndrome, thereby offering opportunities to identify specific disruptions in molecular networks that will enable us to identify novel therapeutic targets.