Citations

INTRODUCTION: Screening for Barrett's esophagus (BE) is suggested in those with risk factors, but remains underutilized. BE/esophageal adenocarcinoma (EAC) risk prediction tools integrating multiple risk factors have been described. However, accuracy remains modest (area under the receiver-operating curve [AUROC] ≤0.7), and clinical implementation has been challenging. We aimed to develop machine learning (ML) BE/EAC risk prediction models from an electronic health record (EHR) database. METHODS: The Clinical Data Analytics Platform, a deidentified EHR database of 6 million Mayo Clinic patients, was used to predict BE and EAC risk. BE and EAC cases and controls were identified using International Classification of Diseases codes and augmented curation (natural language processing) techniques applied to clinical, endoscopy, laboratory, and pathology notes. Cases were propensity score matched to 5 independent randomly selected control groups. An ensemble transformer-based ML model architecture was used to develop predictive models. RESULTS: We identified 8,476 BE cases, 1,539 EAC cases, and 252,276 controls. The BE ML transformer model had an overall sensitivity, specificity, and AUROC of 76%, 76%, and 0.84, respectively. The EAC ML transformer model had an overall sensitivity, specificity, and AUROC of 84%, 70%, and 0.84, respectively. Predictors of BE and EAC included conventional risk factors and additional novel factors, such as coronary artery disease, serum triglycerides, and electrolytes. DISCUSSION: ML models developed on an EHR database can predict incident BE and EAC risk with improved accuracy compared with conventional risk factor-based risk scores. Such a model may enable effective implementation of a minimally invasive screening technology.

Background Obstructive sleep apnea (OSA) is common in heart failure with preserved ejection fraction (HFpEF). However, current evidence is equivocal regarding the potential benefits of treating OSA with positive airway pressure (PAP) therapy in HFpEF. This study assessed the association between adherence to PAP therapy and health care resource use in patients with OSA and HFpEF. Methods and Results Administrative insurance claims data linked with objective PAP therapy usage data from patients with OSA and HFpEF were used to determine associations between PAP adherence and a composite outcome including hospitalizations and emergency room visits. One‐year PAP adherence was based on an adapted US Medicare definition. Propensity score methods were used to create groups with similar characteristics across PAP adherence levels. The study cohort included 4237 patients (54.0% female, mean age 64.1 years); 40% were considered adherent to PAP therapy (30% intermediate adherent, 30% nonadherent). In the matched cohort, PAP‐adherent patients had fewer health care resource use visits than nonadherent patients, a 57% decrease in hospitalizations, and a 36% decrease in emergency room visits versus the year before PAP initiation. Total health care costs were lower in adherent patients than nonadherent patients ($12 732 versus $15 610, P <0.001). Outcomes for intermediately adherent patients were most similar to those for nonadherent patients. Conclusions Treating OSA with PAP therapy in patients with HFpEF was associated with a reduction in health care resource use. These data highlight the importance of managing concomitant OSA in patients with HFpEF, and the need for strategies to enhance PAP adherence in this population.

Abstract Optical coherence tomography (OCT) is an advanced imaging modality to detect Barrett’s esophagus (BE) dysplasia, providing widefield, cross-sectional imaging and microscopic resolution. BE dysplasia is characterized under OCT by the presence and number of glandular structures with atypical morphology. Accurate detection and interpretation of BE glands under OCT is essential to detect dysplastic lesions. Object Detection using deep learning has the potential to identify glands from OCT images. We developed a YOLO model to identify the presence of glands in BE tissue. The YOLOv4 object detector was trained on a custom BE dataset of 30 patients with confirmed BE who underwent OCT imaging, of which 222 OCT images included at least one gland. Our model identified glands with a high average precision of 88.79% on the test dataset. We showed that the developed model is robust to rotation, brightness, and blur in images. We have implemented an object detection model to identify glands from OCT images with promising results accurately. This model has the potential to improve the diagnosis and surveillance of BE by eliminating human error and missed dysplastic lesions adaptable for capsule endoscopy applications.

Abstract Background: Mutations in RAS is reported in embryonal rhabdomyosarcoma (ERMS). Upregulation of IGF2/IGF1R axis and IGF2 overexpression is a common finding. Targeting RAS downstream is associated with high toxicity. Combination therapies targeting parallel pathways is necessary to overcome treatment resistance. Dual inhibition of RAS downstream and IGF2/IGF1R axis can be a potential mechanism of targeted treatment. The efficacy of this mechanism was studied using a microcancer 3D exvivo tumor cell viability assay in a KRAS mutant ERMS. Design: Tumor was collected for a high-risk pediatric ERMS patient. Comprehensive genomic analysis, for identification of targetable genomic alterations, was performed by integrating WES, MPseq and RNAseq. Tumor spheroids grown in 3D environment were tested against multiple single drugs and combinations. The tumor viability was assed employing the Cell Titer Glow Luminescent Cell Viability Assay. Results were expressed as nMATP and were shown as drug-response curve exhibiting the ATP concentration changes at Cmax. Results: Genomic findings included an oncogenic KRAS (G12D) mutation. BAY-293 has a partial cell killing response as a single agent. Given to the presence of IGF2 over-expression, IGF2/IGF1R pathway was targeted by linsitinib. Addition of BAY-293 to linsitinib led to enhanced effectiveness of this agent (Table 1). Conclusion: Combination of comprehensive genomic profiling and a laboratory-based 3D micro-cancer model using the patient’s own tumor is an effective way to determine drug efficacy. Targeting RAS pathway by single agents, increases the possibility of resistance. Dual targeting of downstream pathways by MEK and PI3K pathway inhibition is highly toxic, and the clinical usage of this combination is limited. Our results have indicated inhibition of IGF-1R/IGF2 axis in combination with other agents targeting RAS pathway, can be a potential method to address these limitations. ABEL 1. Summary of molecular alterations and targeting methods Molecular alteration Drug Target Kill% at Cmax KRAS (G12D) Mutation BAY-293 RAS-SOS1 interaction 55.3 Trametinib MEK1/2 88.3 Selumetinib MEK1/2 76 Paxalisib PI3K/mTOR 89.7 Samotolisib PI3K/mTOR 92.9 Copanlisib PI3K 95.2 Perifosine AKT 89.2 IGF2 Over-expression Linsitinib IGF-1R/InsR 84.1 BMS-754807 IGF-1R/InsR 67.4 NTRK2 mutation Larotrectinib NTRK 41.9 KMT2D mutation Panobinostat HDAC 85.3 Combinations Kill% at Cmax BAY-293+Selumetinib 100 BAY-293+Samotolisib 100 BAY-293+Paxalisib 99.7 BAY-293+Linsitinib 97.5 Selumetinib+Samotolisib 93.1 Citation Format: Dorsay Sadeghian, Wan-Hsin Lin, James Smadbeck, Alan Panheiter, Michael Barrett, Alexa F. Mccune, Giannoula Karagouga, Sarah Johnson, Ryan W. Feathers, Faye R. Harris, Lindsey Kinsella, Taylor Berry, John Cheville, Panos Anastasiadis, Candace Granberg, Patricio Gargollo, Vasmatzis George. Dual inhibition of RAS and the IGF2/IGF1R pathways: a potential targeting strategy for RAS-mutated embryonal rhabdomyosarcoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 3391.

Background Obstructive sleep apnea (OSA) is a common comorbidity in patients with heart failure, although current evidence is equivocal regarding the potential benefits of treating OSA with positive airway pressure (PAP) therapy in patients with heart failure. This study assessed the impact of adherence to PAP therapy on health care resource utilization in patients with OSA and heart failure with reduced ejection fraction. Methods and Results Administrative insurance claims data linked with objective PAP therapy use data from patients with OSA and heart failure with reduced ejection fraction were used to determine associations between PAP adherence and a composite outcome of hospitalizations and emergency room visits. One‐year PAP adherence was based on an adapted US Medicare definition. Propensity score methods were used to create groups with similar characteristics across PAP adherence levels. The study cohort included 3182 patients (69.9% male, mean age 59.7 years); 39% were considered adherent to PAP therapy (29% intermediate adherent, 31% nonadherent). One year after PAP initiation, adherent patients had fewer composite visits than matched nonadherent patients, driven by a 24% reduction in emergency room visits for adherent patients. Composite visit costs were lower in adherent versus nonadherent patients ($3500 versus $5879, P =0.031), although total health care costs were not statistically different ($13 028 versus $14 729, P =0.889). Conclusions PAP therapy adherence in patients with OSA with heart failure with reduced ejection fraction was associated with a reduction in health care resource utilization. This suggests that greater emphasis should be placed on diagnosing and effectively treating OSA with PAP in patients with heart failure with reduced ejection fraction.

AbstractCopanlisib is an intravenously administered phosphatidylinositol 3‐kinase (PI3K) inhibitor which was investigated in pediatric patients with relapsed/refractory solid tumors. A model‐informed approach was undertaken to support and confirm an empirically selected starting dose of 28 mg/m2 for pediatric patients ≥1 year old, corresponding to 80% of the adult recommended dose adjusted for body surface area. An adult physiologically based pharmacokinetic (PBPK) model was initially established using copanlisib physicochemical and disposition properties and clinical pharmacokinetics (PK) data and was shown to adequately capture clinical PK across a range of copanlisib doses in adult cancer patients. The adult PBPK model was then extended to the pediatric population through incorporation of age‐dependent anatomical and physiological changes and used to simulate copanlisib exposures in pediatric cancer patient age groups. The pediatric PBPK model predicted that the copanlisib 28 mg/m2 dose would achieve similar copanlisib exposures across pediatric ages when compared with historical adult exposures following the approved copanlisib 60 mg dose administered on Days 1, 8, and 15 of a 28‐day cycle. Clinical PK were collected from a phase I study in pediatric patients with relapsed/refractory solid tumors (aged ≥4 years). An established adult population PK model was extended to incorporate an allometrically‐scaled effect of body surface area and confirmed that the copanlisib maximum tolerated dose of 28 mg/m2 was appropriate to achieve uniform copanlisib exposures across the investigated pediatric age range and consistent exposures to historical data in adult cancer patients. The model‐informed approach successfully supported and confirmed the copanlisib pediatric dose recommendation.

Esophageal adenocarcinoma (EAC) is rising in incidence and has a poor 5-year survival. Decades of targeted screening for Barrett’s esophagus (BE), the precursor to EAC, have not appreciably reduced EAC incidence. Here, we review the drawbacks of the current screening paradigm, including overreliance on GERD symptoms and sedated upper endoscopy. We debate the merits of a universal BE screening program. We propose an ideal BE screening strategy combining novel risk stratification models and non-invasive screening techniques to expand selective screening to patients without GERD, but at elevated risk for BE.