Citations

Abstract Rationale In fungal endemic regions, differentiating benign from malignant pulmonary nodules on CT scans remains a major diagnostic challenge. Coccidioides nodules frequently resemble malignant lesions on CT and PET imaging, presenting significant challenges for differentiation, even among radiologists. Consequently, the benign biopsy and resection rate, especially in the Southwestern US, remains high. We previously developed [DOI:10.1371/journal.pone.0196910] and externally validated [DOI:10.1183/13993003.02485-2020] the Benign versus aggRessive nODule Evaluation using Radiomic Stratification (BRODERS) machine learning (ML) classifier to non-invasively distinguish between benign and malignant pulmonary nodules with an AUC of 0.87. However, its performance in certain benign nodules with malignant surface and texture characteristics like in coccidioidomycosis had a suboptimal AUC of 0.65. In this study, we redesigned our radiomics model by including numerous clinical features. Methods In this retrospective study, 113 pulmonary nodules (66 Coccidioides and 47 adenocarcinoma spectrum) detected on CTs with ground truth pathology diagnosis were identified by screening for key terms like “nodule”, “cocci”, and “adenocarcinoma” from both CT and pathology reports using a large language model trained on clinical textual data (Illuminate®). Clinical features such as demographics, smoking history, and laboratory studies were extracted by trained clinicians from the electronic medical record. All nodules were segmented and radiomic features extracted using a semi-automated seed growing algorithm previously described [DOI: 10.1097/JTO.0b013e3182843721]. The database was randomly split into an 80%-20% development and hold-out set prior to ML modeling. Feature selection for variables was conducted utilizing Elastic Net Regression, which provides a balanced approach between Lasso and Ridge regression. A logistic regression model was developed using the selected features and ground truth information of the training cohort. The true performance of the model was tested on the hold-out set. Results 160 radiomics and 22 clinical features were extracted. Elastic Net regression identified 44 features that contributed the most. Some of these clinical features include presence of satellite lesions, smoking history, and presence of positive IgG coccidiomycosis serology. The AUC of the ROC curve for the hold-out test set was 0.86. Conclusions The clinical features of significance that emerged during multimodal modeling are explainable. Designing explainable ML models is important, especially in the Southwestern US where Coccidioides nodules can appear to be malignant to both human experts and ML models. Although coccidioidomycosis serologies alone lack sensitivity and specificity to diagnose Coccidioides nodules, integrating them with a multimodal ML model can improve differentiation between benign Coccidioides and malignant adenocarcinoma nodules. This abstract is funded by: None

Background: Pleuroparenchymal fibroelastosis (PPFE) is a rare interstitial lung disease (ILD) associated with systemic autoimmune rheumatic diseases (SARDs). Its clinical characteristics and outcomes in SARDs remain poorly defined. Objectives: To describe demographics, imaging patterns, and outcomes in SARD-associated PPFE. Methods: Patients diagnosed with PPFE and SARDs at Mayo Clinic (2000-2024) were included. PPFE diagnosis was confirmed by high-resolution computed tomography, and SARD diagnoses were established by rheumatologists. Results: Fifteen of 62 patients with PPFE had SARDs with majority having systemic sclerosis (SSc)(n= 7, 46%). Median age at PPFE diagnosis was 57.2 years; 80% were female; 40% were ever-smokers; 40% required supplemental oxygen; 5 had pulmonary hypertension. Coexisting ILD patterns included UIP (26.7%) and NSIP (20%). Coexisting ILD patterns included usual interstitial pneumonia (26.7%) and nonspecific interstitial pneumonia (20%). Three patients died, including two with SSc. Conclusion: Systemic sclerosis was the most frequently observed SARD among patients with PPFE, and may coexist with other ILD patterns; recognition may inform risk stratification, multidisciplinary evaluation, and longitudinal follow-up.

Abstract Pancreatic cancer disproportionately affects Black individuals in the United States, but they have limited representation in genetic studies of pancreatic ductal adenocarcinoma (PDAC). To address this gap, we performed admixture mapping and genome-wide association analysis (GWAS) in genetically inferred African ancestry individuals (1,030 cases and 889 controls). Admixture mapping identified three regions with a significantly higher proportion of African ancestry in cases compared to controls (5q33.3, 10p1, 22q12.3). GWAS identified a genome-wide significant association at 5p15.33 ( CLPTM1L , rs383009:T>C, T Allele Frequency=0.51, OR:1.45, P value=1.24×10 -8 ), a locus previously associated with PDAC. Known loci at 5p15.33, 7q32.3, 8q24.21 and 7q25.1 also replicated (P value <0.01). Multi-ancestral fine-mapping identified two potential causal SNPs (rs3830069 and rs2735940) at 5p15.33. Collectively these findings identified novel PDAC risk loci and expanded our understanding of this deadly cancer in underrepresented populations, emphasizing the multifactorial nature of PDAC risk including inherited genetic and non-genetic factors. Statement of Significance To understand how genetic variation contributes to PDAC risk in Black people in North American, we studied individuals of genetically-inferred African ancestry. We identified novel risk loci and differences in the contribution of known loci. This demonstrates that ancestry-informed genetic analyses improve our understanding of PDAC risk and enhances discovery.

Background and Clinical Significance: Nonbacterial thrombotic endocarditis (NBTE) is a sterile fibrin-platelet valvular condition associated with malignancy and hypercoagulable states. It produces friable vegetations prone to systemic embolization, often presenting as multifocal ischemic stroke. While modestly linked to advanced adenocarcinomas, its association with melanoma is exceedingly rare; Case Presentation: We present a 43-year-old man with recently diagnosed metastatic melanoma who presented with fever, confusion and abdominal pain. Brain magnetic resonance imaging (MRI) revealed multifocal bilateral acute infarcts. Additional imaging demonstrated splenic and bilateral renal infarcts. Transesophageal echocardiography (TEE) revealed an 8 mm × 7 mm multilobar lesion on the posterior mitral valve leaflet. Blood cultures remained persistently negative; autoimmune and infectious workup were unrevealing, and positron emission tomography-computed tomography (PET-CT) showed no cardiac hypermetabolism. Despite empiric antibiotics for suspected infective endocarditis (IE), progressive embolic infarcts occurred. After exclusion of infection, NBTE was considered, and therapeutic enoxaparin was initiated, resulting in clinical stabilization without hemorrhagic conversion; Conclusions: Distinguishing NBTE from IE remains challenging due to overlapping and nonspecific imaging findings. TEE is the preferred diagnostic modality because of its high sensitivity for detecting small valvular vegetations. Adjunctive imaging modalities such as brain MRI and PET-CT may support the diagnosis by demonstrating embolic patterns or excluding metabolically active infectious vegetations. Management primarily relies on systemic anticoagulation, while percutaneous vegetation aspiration may represent a potential diagnostic and therapeutic strategy. Clinicians should maintain high suspicion of this condition in patients with advanced melanoma and other malignancies presenting with multifocal embolic phenomena and negative cultures to enable timely anticoagulation.

Abstract Aims Routine aspirin use for primary prevention yields modest cardiovascular benefit but increases bleeding risk in average-risk adults. Whether individuals with elevated lipoprotein(a) [Lp(a)] levels or high-risk LPA variants derive greater benefit remains uncertain. Methods and results Following Cochrane and PRISMA guidelines, we systematically searched PubMed/MEDLINE, Embase, and Cochrane Central for studies of adults without ASCVD, elevated Lp(a) (≥50 mg/dL), high-risk LPA genotypes, or elevated genetic risk scores, comparing aspirin vs. no aspirin use. Random-effects meta-analyses using restricted maximum-likelihood (REML) estimation with Hartung–Knapp–Sidik–Jonkman (HKSJ) adjustment were used to pool hazard ratios (HR) and 95% confidence intervals (CI). In six studies, 6628 participants were included. Pooled random effects showed no significant reduction in major adverse cardiovascular events (MACE) with aspirin (HR = 0.60; 95% CI 0.31–1.17; P = 0.1) or clinical bleeding events (HR = 1.24; 95% CI 0.83–1.87; P = 0.18). Subgroup analyses showed a nonsignificant reduction MACE among elevated Lp(a) (HR = 0.63; 95% CI 0.16 to 2.49; P = 0.284) and statistically significant reduction among LPA rs3798220 carriers (HR = 0.37; 95% CI 0.19 to 0.71; P = 0.003), though this finding was not consistent in the sensitivity analyses. Conclusion Among adults without ASCVD but with elevated Lp(a) or high-risk LPA genotypes, aspirin use did not confer a statistically significant cardiovascular benefit and was associated with numerically higher bleeding risk. The apparent reduction seen in rs3798220 carriers was inconsistent and likely reflects small-sample bias.

Introduction: Apolipoprotein B (ApoB) and coronary artery calcium score (CAC) are well-established risk factors for coronary heart disease (CHD). However, the interaction between these risk factors on CHD risk has not been well-studied. We investigated the association between ApoB levels and CAC with CHD risk. Methods: We analyzed 4,649 participants from the Multiethnic Study of Atherosclerosis cohort (median follow-up 13.9 years) without baseline CHD or statin use. High ApoB was defined as ≥120 mg/dL. Participants were divided into 6 groups; group 1: CAC = 0 with normal ApoB, group 2: CAC = 0 with high ApoB, group 3: CAC 1-99 with normal ApoB, group 4: CAC 1-99 with high ApoB, group 5: CAC ≥100 with normal ApoB, group 6: CAC ≥100 with high ApoB. CHD was defined as myocardial infarction, resuscitated cardiac arrest, or CHD death. Survival analysis using Kaplan-Meier and multivariable Cox proportional hazard models were performed to assess the relationship between ApoB, CAC, and time to first CHD event. Results: In an adjusted model, high ApoB, CAC (1-99), and CAC (≥100) were independently associated with CHD (HR: 1.48; 95% CI: 1.04, 2.11; HR: 2.54; 95% CI: 1.78, 3.64; and HR: 4.41; 95% CI: 3.08, 6.31, respectively). Compared to group 1 (CAC 0 with normal ApoB), CAC 0 and high ApoB (group 2) was not associated with increased risk of CHD (HR: 1.15; 95% CI: 0.59, 2.23). However, a statistically significant greater risk of CHD was observed in those with CAC 1-99 and normal ApoB (group 3) (HR: 2.12; 95% CI: 1.37, 3.28), CAC 1-99 and high ApoB (group 4) (HR: 4.12; 95% CI: 2.42, 7.00), CAC ≥100 and normal ApoB (group 5) (HR: 4.20; 95% CI: 2.78, 6.34), CAC ≥100 and high ApoB (group 6) (HR: 5.72; 95% CI: 3.40, 9.63). Conclusions: In this analysis, ApoB and CAC are independently associated with CHD risk. Individuals with high ApoB and CAC ≥100 had the highest risk of CHD. In contrast, individuals with CAC = 0 and high ApoB had no significant increase in CHD risk compared to individuals with normal ApoB levels. These findings suggest that ApoB and CAC may be complementary tools for CHD risk stratification in the context of primary prevention.

Backround: Atrial fibrillation (AF) is a common complication following veno-venous extracorporeal membrane oxygenation (VV-ECMO), leading to further complications from anticoagulation or stroke. The Artificial Intelligence AF Prediction Dashboard developed at Mayo Clinic generates two predictive outputs: (1) a continuous probability score and (2) a binary threshold-based signal to predict the likelihood of AF at an individual level after 12-lead electrocardiogram (ECG) is obtained. Using both continuous risk (%) and binary (yes/no) likelihood, we present a novel method to understand AF risk after VV-ECMO cannulation based on pre-ECMO 12-lead ECG. Methods: We reviewed a total of 147 patients on ECMO between 2018 and 2024, from which 85 VV-ECMO patients with complete data were included. AI Dashboard outputs were quantified as binary (1/0) or percent likelihood for AF. R was utilized for statistical analysis. Results: 85 patients met our inclusion criteria from which AF after cannulation occurred in 31 (36%). 17 patients had a history of AF prior to VV-ECMO initiation. The mean age of our cohort was 49 years with a mean BMI of 29.3. Baseline AI Dashboard continuous assessment identified 39 patients (46%) compared to the binary assessment identifying 31 patients (36%). Logistic regression in the binary assessment demonstrated an odds ratio of 1.08 (95% CI 1.04 – 1.12) with a p value of <0.001. The binary AF Prediction yielded a 58% sensitivity, 76% specificity, 58% PPV, and 76% NPV, with an accuracy of 67%. The continuous AF Probability score achieved 72% balanced accuracy with sensitivity 74%, specificity 70%, PPV 59% and NPV 83%, indicating fair overall performance and suggesting good utility for ruling out AF in low-risk patients. Although both outputs provided moderate predictive accuracy, the continuous score reported before ECMO cannulation demonstrated stronger overall performance with fewer false negatives and higher rule-out value. Conclusions: We demonstrate the novel use of an AI predictive model in the VV-ECMO population for the prediction of AF risk with both continuous and binary risk cutoffs from baseline ECG. This data highlights the need for further development in ECMO-specific AI models to help mitigate risk after the onset of atrial arrhythmias. As AI-driven tools become increasingly integrated into care, their performance must be continuously validated and recalibrated for specific populations to ensure their safe and effective use.