Citations

Introduction Translating findings from cell and animal models to the clinic is difficult because these systems insufficiently capture human disease biology. Computational approaches, notably pathways modeling, have overcome interspecies differences by mapping shared signaling pathways between species. We considered whether this approach could be generalized to bridge animal and virtual drug screening data to effects measured in patients. We emphasized translational psychiatry datasets because of the unmet need for effective therapies. Methods We conducted three parallel analyses using published drug screening data in zebrafish, structure-function computational screening data, and clinically-reported efficacy data for experimental and approved antipsychotic drugs and predicted drug pathways using the PathFX algorithm Results Despite choosing screens developed for unique aspects of psychiatry and without careful curation of shared drugs, we found pathways associated with drugs that performed “well” across these distinct platforms—specifically, among drugs measured as favorable (expected to improve disease) or unfavorable (expected to aggravate disease) across screens, these drugs connected to distinct pathway proteins; however, these drugs are connected to similar protein families and shared Gene Ontology functional terms. By comparing screens, we discovered that favorable drugs may influence G-proteins, solute carrier family proteins, adrenoceptors, and steroid hydroxylase activity and that unfavorable drugs affect serotonin receptors and phosphodiesterase activity. Discussion This suggests that predictive platforms could emphasize functional information as features that could overcome differences in distinct screening platforms to eventually improve translational approaches.

357 Background: Prostate cancer (PCa) is the most incident cancer in adult males in the United States, and the second highest in the world. CTCAE is the standardized method to grade the severity of treatment-related adverse events (AEs), but are tedious to collect and subject to inter-observer variability. This study aimed to evaluate the performance of LLMs in automatically extracting and grading CTCAE toxicities from clinical notes and patient-reported outcomes of PCa patients from a clinical trial. Methods: A total of 55 patients from NCT02874014 trial, undergoing proton therapy (PT) for localized PCa, were included. The ground truth of CTCAE graded toxicities was obtained from the trial’s results. Clinical notes within 3 days, and the most recent patient questionnaires within 90 days of toxicity assessment were retrieved from the electronic health record. A comprehensive prompt was developed to extract and CTCAE grade toxicities from the retrieved records. LLMs used for this analysis were OpenAI GPT-4o and Gemini 2.0 Flash. Phase 1 evaluated both LLMs' performance against the ground truth. To ensure the metrics reflected the models’ performance based only on the information available in the text provided, cases where both LLMs contradicted the ground truth were manually reviewed. Phase 2 recalculated LLMs' performance metrics using this adapted ground truth. A conservative stepwise ensemble model was developed to leverage the complementary strengths of both LLMs. Results: In the conservative stepwise ensemble model, Gemini 2.0 Flash served as the initial screening platform to identify AEs in step 1. All cases with identified toxicity events were then reviewed and confirmed using OpenAI GPT-4o in step 2. This effectively reduced false positive rates while preserving high sensitivity. The ensemble model achieved strong performance metrics: binary accuracy (0.954), sensitivity (0.961), specificity (0.953), F1 score (0.836), and grade accuracy (0.932). Conclusions: This study demonstrates the feasibility of the use of LLMs for extraction and CTCAE toxicity grading in PCa patients treated with RT. The ensemble model achieved strong performance metrics. Further validation is needed for other cancer diagnosis and treatment modalities. Phase Model Binary Accuracy Precision Sensitivity Specificity F1 Score Grade Accuracy 1 Gemini 0.946 0.672 1.000 0.939 0.804 0.919 1 OpenAI 0.940 0.785 0.637 0.978 0.703 0.895 2 Gemini 0.956 0.735 1.000 0.950 0.848 0.927 2 OpenAI 0.954 0.917 0.679 0.991 0.780 0.903

Introduction Burnout among healthcare professionals remains a critical public health issue linked to impaired cognition, emotional exhaustion and diminished clinical performance. Structured breathing practices have demonstrated promise in improving autonomic regulation and cerebral oxygenation, yet their feasibility, acceptability and implementation in real-world healthcare settings remain underexplored. Objectives This single-arm pilot feasibility trial aims to evaluate the feasibility, acceptability and implementation appropriateness of a structured breathing-based intervention for healthcare professionals across community-based Mayo Clinic Health System (MCHS) sites. Secondary objectives include assessing usability and engagement with the mobile breathing platform, while exploratory analyses will describe magnitude of variability and feasibility of collecting psychological and cerebral haemodynamic measures. This study will commence in November 2025 and is expected to be completed by June 2026. Methods and analysis A total of 40 clinicians (MD/DO/MBBS/PA) and nurses reporting moderate or greater burnout will be enrolled across four MCHS sites. Participants will complete a 4-month structured breathing programme delivered primarily online, supported by a mobile application for practice tracking. Assessments will occur at baseline, 2 months and 4 months, with functional near-infrared spectroscopy (fNIRS) measures of cerebral oxygenation collected at baseline and 4 months in a population subset. Primary outcomes include (1) recruitment yield, retention and adherence rates; (2) acceptability and participant satisfaction (survey and qualitative feedback); and (3) implementation appropriateness measured by the Acceptability of Intervention Measure (AIM), Intervention Appropriateness Measure (IAM) and Feasibility of Intervention Measure (FIM). Secondary outcomes include digital engagement and usability through mobile analytics and the System Usability Scale. Exploratory outcomes are psychological indicators (burnout, depression, anxiety, perceived stress, sleep, fatigue, professional fulfilment and resilience) and physiological endpoints (fNIRS). Analyses will be descriptive, summarising feasibility metrics with 95% CIs. Progression criteria (recruitment≥75%, retention≥80%, adherence≥70%, AIM/IAM/FIM≥4.0) will determine readiness for a definitive hybrid effectiveness–implementation trial. Ethics and dissemination The study is approved by the Mayo Clinic Institutional Review Board (IRB # 25-009320). All participants will provide informed consent. Study procedures ensure confidentiality, cultural sensitivity and participant safety. Data will be securely stored in REDCap and disseminated through peer-reviewed publications and scientific conference Trial registration number NCT07218458 .

Abstract Introduction: Accurate detection and prioritization of tumor-specific somatic variants is foundational to the development of personalized immunotherapies. Traditional pipelines for neoepitope identification rely heavily on short-read next-generation sequencing (NGS) and DNA-based variant calling, followed by RNA expression filtering and HLA-binding prediction. However, this approach is limited by time-intensive workflows, potential false positives from DNA-only calls, and poor capture of complex isoforms or RNA-level editing. Moreover, the accuracy of neoepitope prediction is constrained by conventional sequence-based HLA binding algorithms that do not account for structural and biophysical features of peptide-MHC interactions, particularly across underrepresented HLA alleles. Methods: To overcome these limitations, we developed a streamlined, sequencing-to-vaccine framework that leverages long-read RNA sequencing for direct detection of expressed somatic variants, including fusions, indels, and single nucleotide variants. Using Oxford Nanopore Technologies’ cDNA-PCR approach, we enable comprehensive, full-length transcript characterization from both fresh and frozen breast tumor samples. Our bioinformatic pipeline, NeoTarget, integrates alignment, variant calling, and neoepitope inference from tumor RNA in a single, modular workflow. A core innovation of this pipeline is HLA-Inception, a deep convolutional neural network trained to predict peptide-MHC class I binding using features derived from molecular dynamics-based modeling of electrostatic potential at the peptide-HLA interface, enhancing generalizability across diverse HLA subtypes. Results: Expanding on previous results, the current NeoTarget pipeline identifies, evaluates, and prioritizes single nucleotide variants, indels, and fusions. To validate this system, we used a set of 9 human breast cancer tumors benchmarked with short read sequencing. Identified neoantigens included most variants shown by clinical short read sequencing and also novel structural variants (i.e., fusions) only revealed with long read sequencing with high confidence. Conclusion: Here, we describe the design and performance of the NeoTarget pipeline, and show proof-of-concept feasibility of applying this fast and cost-effective workflow for HLA-restricted epitope discovery in breast tumors using long-read RNA sequencing. Overall, these methods can be applied to rapidly identify and prioritize neoantigens for personalized neoantigen-based immune therapies. Citation Format: F. Batalini, S. Henry, Y. Zhang, R. Foster, J. M. Gayton, A. Hostal, E. A. Wilson, A. Singharoy, M. Gustafson, B. Pockaj, J. Park, K. S. Anderson. Somatic variant detection using long read RNA sequencing reveals HLA-restricted targets for immunotherapy [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS3-06-08.

Abstract Progression from monoclonal gammopathy of undetermined significance (MGUS) to multiple myeloma (MM) is driven by coordinated metabolic reprogramming within clonal plasma cells and the bone marrow microenvironment. We applied high-resolution MALDI–FT-ICR mass spectrometry imaging (MSI) to archived FFPE bone-marrow biopsies, integrated with matched bone marrow plasma metabolomics, to map spatial and systemic metabolic alterations. Spatial clustering delineated plasma-cell–rich niches, while Hill-based diversity and β-diversity metrics quantified intra- and inter-compartment heterogeneity. MM niches exhibited elevated 3-hydroxykynurenine, rewired tryptophan–kynurenine flux, and increased nucleotide and bioactive lipid metabolism associated with proliferation. Notably, some MGUS-like samples displayed MM-like metabolic niches undetectable in bone marrow plasma alone, underscoring spatial heterogeneity. Cross-compartment integration revealed conserved metabolic signatures and systemic redistribution of key metabolites, consistent with ecological reorganization and niche divergence during progression. These findings establish spatial metabolomics of biopsies as a framework to dissect intramedullary metabolic heterogeneity and enable metabolite-based risk stratification in plasma-cell disorders.

Abstract Objective To estimate the incidence of de novo chronic rhinosinusitis and identify associated risk factors in kidney and liver transplant recipients without pre‐existing sinonasal complaints. Study design Retrospective cohort study. Setting Multisite study across Mayo Clinic Enterprise locations in Arizona, Florida, Minnesota, and Wisconsin. Methods Records of all patients who underwent kidney or liver transplantation between November 1, 2021, and November 1, 2022 were reviewed. Patients with documented sinonasal complaints prior to transplantation were excluded. Diagnoses were based on ICD‐10 codes assigned by board‐certified otolaryngologists, with follow‐up extending through November 2024. Demographic and clinical variables were collected, and multivariable logistic regression was used to identify predictors of de novo chronic rhinosinusitis. Results Among 1459 transplant recipients (mean age 55.1 years, 59.8% male), the cumulative 2‐year incidence of de novo chronic rhinosinusitis was 2.4% (24/986) in kidney and 2.3% (11/473) in liver transplant recipients ( P = .89). The overall incidence rate was 12.0 cases per 1000 person‐years, exceeding general population estimates. Transplant rejection (OR 3.2, 95% CI 2.5‐3.9, P < .001) and additional organ transplantation (OR 5.0, 95% CI 3.4‐6.5, P = .041) were independently associated with chronic rhinosinusitis development. Conclusion Kidney and liver transplant recipients experience a higher incidence of de novo chronic rhinosinusitis compared to the general population. Transplant rejection and additional organ transplantation significantly increase this risk. Early recognition and management of chronic rhinosinusitis in this population may support improved postoperative outcomes.

Background/Objectives: Precocious puberty (PP), defined as the onset of secondary sexual characteristics before 8 years in girls and 9 years in boys, is associated with psychosocial distress, compromised adult height, and long-term metabolic risk. Early identification remains challenging, as current diagnostic approaches are largely reactive and rely on invasive or resource-intensive testing. This narrative review examines how artificial intelligence (AI) can support earlier risk prediction and detection of PP through integration of clinical, hormonal, imaging, lifestyle, and environmental data. Methods: A narrative literature review was conducted using PubMed, Scopus, Embase, Web of Science, and Google Scholar to identify relevant studies published between 2005 and 2025. Eligible studies included original research and high-quality reviews that examined AI-based approaches, such as machine learning and deep learning, in pediatric endocrinology, particularly for the prediction or diagnosis of central or peripheral precocious puberty. Studies incorporating clinical, hormonal, radiological, lifestyle, environmental, or multi-omics data relevant to AI modeling were included. Results: AI models, including XGBoost, random forest, convolutional neural networks, and regression-based approaches, have demonstrated potential utility in predicting central precocious puberty using hormonal, imaging, and growth data. Reported applications include automated bone age assessment, lifestyle and dietary risk stratification, and exploratory use of wearable-derived behavioral data. However, progress is limited by small pediatric datasets, population bias, limited interpretability, and unresolved ethical challenges related to privacy, consent, and equity. Conclusions: Artificial intelligence represents a promising decision-support approach for earlier, non-invasive, and individualized risk assessment in precocious puberty. Future progress will depend on the integration of longitudinal, multimodal data, the development of ethical models, and interdisciplinary collaboration among pediatric endocrinologists, data scientists, and public health stakeholders.

BACKGROUND: With pulmonary hypertension (PH), a pulmonary artery wedge pressure (PAWP)>15 mm Hg is used to diagnose left heart dysfunction, but some patients with adjudicated group 1 PH demonstrate PAWP>15 mm Hg. The primary objective of the study was to evaluate group 1 PH with high PAWP>15 mm Hg. METHODS: Patients with adjudicated group 1 PH from PVDOMICS between 2016 and 2019 were separated into high PAWP(>15 mm Hg) or normal PAWP and compared with adjudicated combined pre- and postcapillary (Cpc) PH related to heart failure with preserved ejection fraction (HFpEF). Participants underwent dynamic right heart catheterization and metabolomics. Findings were validated in 3 independent cohorts with adjudicated group 1 PH (validation cohorts 1 and 2 with exercise right heart catheterization and validation cohort 3 with resting right heart catheterization). RESULTS: Of 325 patients with group 1 PH (73% women, mean age 53.0±14.3 years), 15% (n=48) had high PAWP. Group 1 PH+high PAWP demonstrated greater obesity, left ventricular hypertrophy ( P <0.0002), left ventricular strain impairment ( P <0.0001), and decreased left ventricular compliance ( P <0.0001) compared with group 1 PH+normal PAWP, with changes comparable to Cpc-PH HFpEF (n=75). Compared with Cpc-PH HFpEF, left atrial function was better in group 1 PH+high PAWP with lower PAWP V wave, higher left atrial compliance, and left atrial ejection fraction ( P <0.0001 for all). Metabolomics demonstrated little difference between group 1 PH with high versus normal PAWP but large differences between group 1 PH+high PAWP versus Cpc-PH HFpEF (100 metabolites altered at false discovery rate P <0.05). Elevated PAWP was also observed in 18% of group 1 PH in the validation cohort 1 (n=402), with exercise PAWP response intermediately abnormal in group 1 PH+high PAWP relative to Cpc-PH HFpEF and group 1 PH+normal PAWP (interaction P <0.0001). Elevated PAWP was similarly observed in 22% and 19% of group 1 PH in validation cohorts 2 (n=55) and 3 (n=787), respectively. CONCLUSIONS: Approximately 1 in 5 adjudicated patients with group 1 PH has elevation in resting PAWP despite severe pulmonary vascular dysfunction and metabolomics consistent with traditionally defined group 1 PH. Despite resting PAWP elevation, these patients with group 1 PH were metabolomically and biologically distinct from Cpc PH HFpEF, with better left atrial function and diastolic reserve during exercise. These data emphasize the limitations of using resting PAWP alone to separate group 1 PH from HFpEF and call for development of more integrated clinical diagnostic criteria.