RESUMEN
Pancreatic ductal adenocarcinoma (PDA) is one of the most lethal human malignancies, owing in part to its propensity for metastasis. Here, we used an organoid culture system to investigate how transcription and the enhancer landscape become altered during discrete stages of disease progression in a PDA mouse model. This approach revealed that the metastatic transition is accompanied by massive and recurrent alterations in enhancer activity. We implicate the pioneer factor FOXA1 as a driver of enhancer activation in this system, a mechanism that renders PDA cells more invasive and less anchorage-dependent for growth in vitro, as well as more metastatic in vivo. In this context, FOXA1-dependent enhancer reprogramming activates a transcriptional program of embryonic foregut endoderm. Collectively, our study implicates enhancer reprogramming, FOXA1 upregulation, and a retrograde developmental transition in PDA metastasis.
Asunto(s)
Adenocarcinoma/genética , Carcinoma Ductal Pancreático/genética , Elementos de Facilitación Genéticos , Regulación Neoplásica de la Expresión Génica , Factor Nuclear 3-alfa del Hepatocito/genética , Neoplasias Pancreáticas/genética , Adenocarcinoma/metabolismo , Adenocarcinoma/patología , Animales , Carcinoma Ductal Pancreático/metabolismo , Carcinoma Ductal Pancreático/patología , Línea Celular Tumoral , Modelos Animales de Enfermedad , Epigenómica , Femenino , Perfilación de la Expresión Génica , Humanos , Masculino , Ratones , Ratones Endogámicos C57BL , Metástasis de la Neoplasia , Organoides/metabolismo , Páncreas/metabolismo , Neoplasias Pancreáticas/metabolismo , Neoplasias Pancreáticas/patologíaRESUMEN
Pancreatic cancer is a deadly malignancy that lacks effective therapeutics. We previously reported that oncogenic Kras induced the redox master regulator Nfe2l2/Nrf2 to stimulate pancreatic and lung cancer initiation. Here, we show that NRF2 is necessary to maintain pancreatic cancer proliferation by regulating mRNA translation. Specifically, loss of NRF2 led to defects in autocrine epidermal growth factor receptor (EGFR) signaling and oxidation of specific translational regulatory proteins, resulting in impaired cap-dependent and cap-independent mRNA translation in pancreatic cancer cells. Combined targeting of the EGFR effector AKT and the glutathione antioxidant pathway mimicked Nrf2 ablation to potently inhibit pancreatic cancer ex vivo and in vivo, representing a promising synthetic lethal strategy for treating the disease.
Asunto(s)
Factor 2 Relacionado con NF-E2/metabolismo , Neoplasias Pancreáticas/metabolismo , Biosíntesis de Proteínas , Animales , Comunicación Autocrina , Cisteína/metabolismo , Glutatión/metabolismo , Humanos , Ratones , Organoides/metabolismo , Neoplasias Pancreáticas/patología , Proteínas Proto-Oncogénicas p21(ras)/metabolismo , Transducción de SeñalRESUMEN
Lipid peroxidation of polyunsaturated fatty acids (PUFAs) is an endogenous source of α,ß-unsaturated aldehydes that react with DNA producing a variety of cyclic adducts. The mutagenic cyclic adducts, specifically those derived from oxidation of ω-6 PUFAs, may contribute to the cancer promoting activities associated with ω-6 PUFAs. ( E)-4-Hydroxy-2-nonenal (HNE) is a unique product of ω-6 PUFAs oxidation. HNE reacts with deoxyguanosine (dG) yielding mutagenic 1, N2-propanodeoxyguanosine adducts (HNE-dG). Earlier studies showed HNE can also be oxidized to its epoxide (EH), and EH can react with deoxyadenosine (dA) forming the well-studied εdA and the substituted etheno adducts. Using a liquid chromatography-based tandem mass spectroscopic (LC-MS/MS) method, we previously reported the detection of EH-derived 7-(1',2'-dihydroxyheptyl)-1, N6-ethenodeoxyadenosine (DHHεdA) as a novel endogenous background adduct in DNA from rodent and human tissues. The formation, repair, and mutagenicity of DHHεdA and its biological consequences in cells have not been investigated. To understand the roles of DHHεdA in carcinogenesis, it is important to develop an immuno-based assay to detect DHHεdA in cells and tissues. In this study we describe the development of monoclonal antibodies specifically against DHHεdA and its application to detect DHHεdA in human cells.
Asunto(s)
Anticuerpos Monoclonales/inmunología , Aductos de ADN/química , Aductos de ADN/inmunología , Ácidos Grasos Omega-6/química , Adenosina/análogos & derivados , Adenosina/química , Adenosina/inmunología , Aldehídos/química , Animales , Carcinógenos , Separación Celular , Cromatografía Liquida/métodos , ADN/química , Ensayo de Inmunoadsorción Enzimática/métodos , Compuestos Epoxi/farmacología , Citometría de Flujo , Células Hep G2 , Hepatocitos/efectos de los fármacos , Humanos , Ratones , Espectrometría de Masas en Tándem/métodosRESUMEN
With Artificial Intelligence (AI) increasingly permeating various aspects of society, including healthcare, the adoption of the Transformers neural network architecture is rapidly changing many applications. Transformer is a type of deep learning architecture initially developed to solve general-purpose Natural Language Processing (NLP) tasks and has subsequently been adapted in many fields, including healthcare. In this survey paper, we provide an overview of how this architecture has been adopted to analyze various forms of healthcare data, including clinical NLP, medical imaging, structured Electronic Health Records (EHR), social media, bio-physiological signals, biomolecular sequences. Furthermore, which have also include the articles that used the transformer architecture for generating surgical instructions and predicting adverse outcomes after surgeries under the umbrella of critical care. Under diverse settings, these models have been used for clinical diagnosis, report generation, data reconstruction, and drug/protein synthesis. Finally, we also discuss the benefits and limitations of using transformers in healthcare and examine issues such as computational cost, model interpretability, fairness, alignment with human values, ethical implications, and environmental impact.
Asunto(s)
Aprendizaje Profundo , Procesamiento de Lenguaje Natural , Humanos , Inteligencia Artificial , Atención a la Salud/organización & administración , Redes Neurales de la Computación , Registros Electrónicos de SaludRESUMEN
On average, more than 5 million patients are admitted to intensive care units (ICUs) in the US, with mortality rates ranging from 10 to 29%. The acuity state of patients in the ICU can quickly change from stable to unstable, sometimes leading to life-threatening conditions. Early detection of deteriorating conditions can assist in more timely interventions and improved survival rates. While Artificial Intelligence (AI)-based models show potential for assessing acuity in a more granular and automated manner, they typically use mortality as a proxy of acuity in the ICU. Furthermore, these methods do not determine the acuity state of a patient (i.e., stable or unstable), the transition between acuity states, or the need for life-sustaining therapies. In this study, we propose APRICOT-M (Acuity Prediction in Intensive Care Unit-Mamba), a 1M-parameter state space-based neural network to predict acuity state, transitions, and the need for life-sustaining therapies in real-time among ICU patients. The model integrates ICU data in the preceding four hours (including vital signs, laboratory results, assessment scores, and medications) and patient characteristics (age, sex, race, and comorbidities) to predict the acuity outcomes in the next four hours. Our state space-based model can process sparse and irregularly sampled data without manual imputation, thus reducing the noise in input data and increasing inference speed. The model was trained on data from 107,473 patients (142,062 ICU admissions) from 55 hospitals between 2014-2017 and validated externally on data from 74,901 patients (101,356 ICU admissions) from 143 hospitals. Additionally, it was validated temporally on data from 12,927 patients (15,940 ICU admissions) from one hospital in 2018-2019 and prospectively on data from 215 patients (369 ICU admissions) from one hospital in 2021-2023. Three datasets were used for training and evaluation: the University of Florida Health (UFH) dataset, the electronic ICU Collaborative Research Database (eICU), and the Medical Information Mart for Intensive Care (MIMIC)-IV dataset. APRICOT-M significantly outperforms the baseline acuity assessment, Sequential Organ Failure Assessment (SOFA), for mortality prediction in both external (AUROC 0.95 CI: 0.94-0.95 compared to 0.78 CI: 0.78-0.79) and prospective (AUROC 0.99 CI: 0.97-1.00 compared to 0.80 CI: 0.65-0.92) cohorts, as well as for instability prediction (external AUROC 0.75 CI: 0.74-0.75 compared to 0.51 CI: 0.51-0.51, and prospective AUROC 0.69 CI: 0.64-0.74 compared to 0.53 CI: 0.50-0.57). This tool has the potential to help clinicians make timely interventions by predicting the transition between acuity states and decision-making on life-sustaining within the next four hours in the ICU.
RESUMEN
Delirium is a syndrome of acute brain failure which is prevalent amongst older adults in the Intensive Care Unit (ICU). Incidence of delirium can significantly worsen prognosis and increase mortality, therefore necessitating its rapid and continual assessment in the ICU. Currently, the common approach for delirium assessment is manual and sporadic. Hence, there exists a critical need for a robust and automated system for predicting delirium in the ICU. In this work, we develop a machine learning (ML) system for real-time prediction of delirium using Electronic Health Record (EHR) data. Unlike prior approaches which provide one delirium prediction label per entire ICU stay, our approach provides predictions every 12 hours. We use the latest 12 hours of ICU data, along with patient demographic and medical history data, to predict delirium risk in the next 12-hour window. This enables delirium risk prediction as soon as 12 hours after ICU admission. We train and test four ML classification algorithms on longitudinal EHR data pertaining to 16,327 ICU stays of 13,395 patients covering a total of 56,297 12-hour windows in the ICU to predict the dynamic incidence of delirium. The best performing algorithm was Categorical Boosting which achieved an area under receiver operating characteristic curve (AUROC) of 0.87 (95% Confidence Interval; C.I, 0.86-0.87). The deployment of this ML system in ICUs can enable early identification of delirium, thereby reducing its deleterious impact on long-term adverse outcomes, such as ICU cost, length of stay and mortality.
RESUMEN
INTRODUCTION: The purpose of this study is to examine the effect of admission glucose in patients hospitalized with COVID-19 with and without diabetes mellitus in a largely African American cohort. DESIGN AND METHODS: This study included 708 adults (89% non-Hispanic Black) admitted with COVID-19 to an urban hospital between 1 March and 15 May 2020. Patients with diabetes were compared with those without and were stratified based on admission glucose of 140 and 180 mg/dL. Adjusted ORs were calculated for outcomes of mortality, intubation, intensive care unit (ICU) admission, acute kidney injury (AKI), and length of stay based on admission glucose levels. RESULTS: Patients with diabetes with admission glucose >140 mg/dL (vs <140 g/dL) had 2.4-fold increased odds of intubation (95% CI 1.2 to 4.6) and 2.1-fold increased odds of ICU admission (95% CI 1.0 to 4.3). Patients with diabetes with admission glucose >180 mg/dL (vs <180 g/dL) had a 1.9-fold increased mortality (95% CI 1.2 to 3.1). Patients without diabetes with admission glucose >140 mg/dL had a 2.3-fold increased mortality (95% CI 1.3 to 4.3), 2.7-fold increased odds of ICU admission (95% CI 1.3 to 5.4), 1.9-fold increased odds of intubation (95% CI 1.0 to 3.7) and 2.2-fold odds of AKI (95% CI 1.1 to 3.8). Patients without diabetes with glucose >180 mg/dL had 4.4-fold increased odds of mortality (95% CI 1.9 to 10.4), 2.7-fold increased odds of intubation (95% CI 1.2 to 5.8) and 3-fold increased odds of ICU admission (95% CI 1.3 to 6.6). CONCLUSION: Our results show hyperglycemia portends worse outcomes in patients with COVID-19 with and without diabetes. While our study was limited by its retrospective design, our findings suggest that patients presenting with hyperglycemia require closer observation and more aggressive therapies.
Asunto(s)
Lesión Renal Aguda , COVID-19 , Diabetes Mellitus , Hiperglucemia , Lesión Renal Aguda/epidemiología , Lesión Renal Aguda/terapia , Adulto , Negro o Afroamericano , COVID-19/complicaciones , COVID-19/epidemiología , Diabetes Mellitus/epidemiología , Glucosa , Humanos , Hiperglucemia/epidemiología , Estudios Retrospectivos , AzúcaresRESUMEN
Transformer model architectures have revolutionized the natural language processing (NLP) domain and continue to produce state-of-the-art results in text-based applications. Prior to the emergence of transformers, traditional NLP models such as recurrent and convolutional neural networks demonstrated promising utility for patient-level predictions and health forecasting from longitudinal datasets. However, to our knowledge only few studies have explored transformers for predicting clinical outcomes from electronic health record (EHR) data, and in our estimation, none have adequately derived a health-specific tokenization scheme to fully capture the heterogeneity of EHR systems. In this study, we propose a dynamic method for tokenizing both discrete and continuous patient data, and present a transformer-based classifier utilizing a joint embedding space for integrating disparate temporal patient measurements. We demonstrate the feasibility of our clinical AI framework through multi-task ICU patient acuity estimation, where we simultaneously predict six mortality and readmission outcomes. Our longitudinal EHR tokenization and transformer modeling approaches resulted in more accurate predictions compared with baseline machine learning models, which suggest opportunities for future multimodal data integrations and algorithmic support tools using clinical transformer networks.
RESUMEN
We describe a case of a 44-year-old woman with locally advanced aggressive angiomyxoma with a novel translocation high-mobility group AT-hook 2-yes-associated protein 1 (HMGA2-YAP1) fusion, implying a t(11;12)(q22.1;q14.3) translocation. She was started on gonadotropin-releasing hormone agonist injection and an aromatase inhibitor for persistent disease, which responded to treatment; she was subsequently treated with radiation before a more definitive operation was conducted. This case report indicates that HGMA2-YAP1-translocated aggressive angiomyxoma is responsive to oestrogen antagonism and hopefully will allow for the development of diagnostics useful for this rare but often morbid neoplasm. This case also highlights the importance of appropriate workup of a soft tissue mass.
Asunto(s)
Proteínas Adaptadoras Transductoras de Señales/genética , Proteína HMGA2/genética , Mixoma/genética , Factores de Transcripción/genética , Neoplasias de la Vulva/genética , Adulto , Anastrozol/uso terapéutico , Antineoplásicos/uso terapéutico , Inhibidores de la Aromatasa/uso terapéutico , Diagnóstico Diferencial , Antagonistas de Estrógenos/uso terapéutico , Femenino , Humanos , Leuprolida/uso terapéutico , Imagen por Resonancia Magnética , Mixoma/tratamiento farmacológico , Mixoma/cirugía , Enfermedades Raras , Translocación Genética/genética , Resultado del Tratamiento , Neoplasias de la Vulva/tratamiento farmacológico , Neoplasias de la Vulva/cirugía , Proteínas Señalizadoras YAPRESUMEN
Pancreatic ductal adenocarcinoma (PDA) is a lethal malignancy with limited treatment options. Although metabolic reprogramming is a hallmark of many cancers, including PDA, previous attempts to target metabolic changes therapeutically have been stymied by drug toxicity and tumour cell plasticity. Here, we show that PDA cells engage an eIF4F-dependent translation program that supports redox and central carbon metabolism. Inhibition of the eIF4F subunit, eIF4A, using the synthetic rocaglate CR-1-31-B (CR-31) reduced the viability of PDA organoids relative to their normal counterparts. In vivo, CR-31 suppresses tumour growth and extends survival of genetically-engineered murine models of PDA. Surprisingly, inhibition of eIF4A also induces glutamine reductive carboxylation. As a consequence, combined targeting of eIF4A and glutaminase activity more effectively inhibits PDA cell growth both in vitro and in vivo. Overall, our work demonstrates the importance of eIF4A in translational control of pancreatic tumour metabolism and as a therapeutic target against PDA.
Asunto(s)
Biosíntesis de Proteínas , Animales , Carcinogénesis , Carcinoma Ductal Pancreático/metabolismo , Línea Celular Tumoral , Factor 4A Eucariótico de Iniciación/antagonistas & inhibidores , Factor 4A Eucariótico de Iniciación/metabolismo , Glutatión/metabolismo , Humanos , Ratones Endogámicos C57BL , Terapia Molecular Dirigida , Oxidación-Reducción , Neoplasias Pancreáticas/metabolismo , Neoplasias PancreáticasRESUMEN
Glycosylation alterations are indicative of tissue inflammation and neoplasia, but whether these alterations contribute to disease pathogenesis is largely unknown. To study the role of glycan changes in pancreatic disease, we inducibly expressed human fucosyltransferase 3 and ß1,3-galactosyltransferase 5 in mice, reconstituting the glycan sialyl-Lewisa, also known as carbohydrate antigen 19-9 (CA19-9). Notably, CA19-9 expression in mice resulted in rapid and severe pancreatitis with hyperactivation of epidermal growth factor receptor (EGFR) signaling. Mechanistically, CA19-9 modification of the matricellular protein fibulin-3 increased its interaction with EGFR, and blockade of fibulin-3, EGFR ligands, or CA19-9 prevented EGFR hyperactivation in organoids. CA19-9-mediated pancreatitis was reversible and could be suppressed with CA19-9 antibodies. CA19-9 also cooperated with the KrasG12D oncogene to produce aggressive pancreatic cancer. These findings implicate CA19-9 in the etiology of pancreatitis and pancreatic cancer and nominate CA19-9 as a therapeutic target.