RESUMO
Acquisition of a lipid-laden phenotype by immune cells has been defined in infectious diseases and atherosclerosis but remains largely uncharacterized in cancer. Here, in breast cancer models, we found that neutrophils are induced to accumulate neutral lipids upon interaction with resident mesenchymal cells in the premetastatic lung. Lung mesenchymal cells elicit this process through repressing the adipose triglyceride lipase (ATGL) activity in neutrophils in prostaglandin E2-dependent and -independent manners. In vivo, neutrophil-specific deletion of genes encoding ATGL or ATGL inhibitory factors altered neutrophil lipid profiles and breast tumor lung metastasis in mice. Mechanistically, lipids stored in lung neutrophils are transported to metastatic tumor cells through a macropinocytosis-lysosome pathway, endowing tumor cells with augmented survival and proliferative capacities. Pharmacological inhibition of macropinocytosis significantly reduced metastatic colonization by breast tumor cells in vivo. Collectively, our work reveals that neutrophils serve as an energy reservoir to fuel breast cancer lung metastasis.
Assuntos
Neoplasias da Mama/metabolismo , Neoplasias da Mama/patologia , Metabolismo dos Lipídeos , Neoplasias Pulmonares/metabolismo , Neoplasias Pulmonares/secundário , Células-Tronco Mesenquimais/metabolismo , Neutrófilos/metabolismo , Animais , Biomarcadores , Proliferação de Células , Progressão da Doença , Endocitose , Feminino , Imunofluorescência , Humanos , Camundongos , Metástase Neoplásica , Neutrófilos/ultraestruturaRESUMO
The field of computational pathology[1,2] has witnessed remarkable progress in the development of both task-specific predictive models and task-agnostic self-supervised vision encoders[3,4]. However, despite the explosive growth of generative artificial intelligence (AI), there has been limited study on building general purpose, multimodal AI assistants and copilots[5] tailored to pathology. Here we present PathChat, a vision-language generalist AI assistant for human pathology. We build PathChat by adapting a foundational vision encoder for pathology, combining it with a pretrained large language model and finetuning the whole system on over 456,000 diverse visual language instructions consisting of 999,202 question-answer turns. We compare PathChat against several multimodal vision language AI assistants and GPT4V, which powers the commercially available multimodal general purpose AI assistant ChatGPT-4[7]. PathChat achieved state-of-the-art performance on multiple-choice diagnostic questions from cases of diverse tissue origins and disease models. Furthermore, using open-ended questions and human expert evaluation, we found that overall PathChat produced more accurate and pathologist-preferable responses to diverse queries related to pathology. As an interactive and general vision-language AI Copilot that can flexibly handle both visual and natural language inputs, PathChat can potentially find impactful applications in pathology education, research, and human-in-the-loop clinical decision making.
RESUMO
Light and sound are the most ubiquitous forms of waves, associated with a variety of phenomena and physical effects such as rainbows and echoes. Light and sound, both categorized as classical waves, have lately been brought into unexpected connections with exotic topological phases of matter. We are currently witnessing the onset of a second wave of active research into this topic. The past decade has been marked by fundamental advances comprising two-dimensional quantum Hall insulators and quantum spin and valley Hall insulators, whose topological properties are characterized using linear band topology. Here, going beyond these conventional topological systems, we focus on the latest frontiers, including non-Hermitian, nonlinear and non-Abelian topology as well as topological defects, for which the characterization of the topological features goes beyond the standard band-topology language. In addition to an overview of the current state of the art, we also survey future research directions for valuable applications.
RESUMO
Tumor-derived extracellular vesicles are important mediators of cell-to-cell communication during tumorigenesis. Here, we demonstrated that hepatocellular carcinoma (HCC)-derived ectosomes remodel the tumor microenvironment to facilitate HCC progression in an ectosomal PKM2-dependent manner. HCC-derived ectosomal PKM2 induced not only metabolic reprogramming in monocytes but also STAT3 phosphorylation in the nucleus to upregulate differentiation-associated transcription factors, leading to monocyte-to-macrophage differentiation and tumor microenvironment remodeling. In HCC cells, sumoylation of PKM2 induced its plasma membrane targeting and subsequent ectosomal excretion via interactions with ARRDC1. The PKM2-ARRDC1 association in HCC was reinforced by macrophage-secreted cytokines/chemokines in a CCL1-CCR8 axis-dependent manner, further facilitating PKM2 excretion from HCC cells to form a feedforward regulatory loop for tumorigenesis. In the clinic, ectosomal PKM2 was clearly detected in the plasma of HCC patients. This study highlights a mechanism by which ectosomal PKM2 remodels the tumor microenvironment and reveals ectosomal PKM2 as a potential diagnostic marker for HCC.
Assuntos
Proteínas de Transporte/metabolismo , Micropartículas Derivadas de Células/metabolismo , Proteínas de Membrana/metabolismo , Hormônios Tireóideos/metabolismo , Adulto , Idoso , Idoso de 80 Anos ou mais , Animais , Carcinoma Hepatocelular/genética , Carcinoma Hepatocelular/metabolismo , Proteínas de Transporte/genética , Diferenciação Celular/genética , Linhagem Celular Tumoral , Proliferação de Células/genética , Micropartículas Derivadas de Células/genética , Micropartículas Derivadas de Células/patologia , Quimiocina CCL1/metabolismo , Progressão da Doença , Células Hep G2 , Humanos , Neoplasias Hepáticas/genética , Neoplasias Hepáticas/metabolismo , Macrófagos/metabolismo , Masculino , Proteínas de Membrana/genética , Camundongos , Camundongos Endogâmicos C57BL , Pessoa de Meia-Idade , Monócitos/metabolismo , Prognóstico , Fator de Transcrição STAT3/metabolismo , Hormônios Tireóideos/genética , Microambiente Tumoral , Proteínas de Ligação a Hormônio da TireoideRESUMO
Cancer of unknown primary (CUP) origin is an enigmatic group of diagnoses in which the primary anatomical site of tumour origin cannot be determined1,2. This poses a considerable challenge, as modern therapeutics are predominantly specific to the primary tumour3. Recent research has focused on using genomics and transcriptomics to identify the origin of a tumour4-9. However, genomic testing is not always performed and lacks clinical penetration in low-resource settings. Here, to overcome these challenges, we present a deep-learning-based algorithm-Tumour Origin Assessment via Deep Learning (TOAD)-that can provide a differential diagnosis for the origin of the primary tumour using routinely acquired histology slides. We used whole-slide images of tumours with known primary origins to train a model that simultaneously identifies the tumour as primary or metastatic and predicts its site of origin. On our held-out test set of tumours with known primary origins, the model achieved a top-1 accuracy of 0.83 and a top-3 accuracy of 0.96, whereas on our external test set it achieved top-1 and top-3 accuracies of 0.80 and 0.93, respectively. We further curated a dataset of 317 cases of CUP for which a differential diagnosis was assigned. Our model predictions resulted in concordance for 61% of cases and a top-3 agreement of 82%. TOAD can be used as an assistive tool to assign a differential diagnosis to complicated cases of metastatic tumours and CUPs and could be used in conjunction with or in lieu of ancillary tests and extensive diagnostic work-ups to reduce the occurrence of CUP.
Assuntos
Inteligência Artificial , Simulação por Computador , Neoplasias Primárias Desconhecidas/patologia , Estudos de Coortes , Simulação por Computador/normas , Feminino , Humanos , Masculino , Metástase Neoplásica/patologia , Neoplasias Primárias Desconhecidas/diagnóstico , Reprodutibilidade dos Testes , Sensibilidade e Especificidade , Fluxo de TrabalhoRESUMO
CLEC12A, a member of the C-type lectin receptor family involved in immune homeostasis, recognizes MSU crystals released from dying cells. However, the molecular mechanism underlying the CLEC12A-mediated recognition of MSU crystals remains unclear. Herein, we reported the crystal structure of the human CLEC12A-C-type lectin-like domain (CTLD) and identified a unique "basic patch" site on CLEC12A-CTLD that is necessary for the binding of MSU crystals. Meanwhile, we determined the interaction strength between CLEC12A-CTLD and MSU crystals using single-molecule force spectroscopy. Furthermore, we found that CLEC12A clusters at the cell membrane and seems to serve as an internalizing receptor of MSU crystals. Altogether, these findings provide mechanistic insights for understanding the molecular mechanisms underlying the interplay between CLEC12A and MSU crystals.
Assuntos
Lectinas Tipo C , Receptores Mitogênicos , Ácido Úrico , Humanos , Gota/metabolismo , Lectinas Tipo C/química , Lectinas Tipo C/imunologia , Receptores Mitogênicos/química , Receptores Mitogênicos/imunologia , Ácido Úrico/química , Ácido Úrico/imunologia , Domínios Proteicos , Cristalografia por Raios X , Imagem Individual de Molécula , Linhagem CelularRESUMO
DNA methylation affects agronomic traits and the environmental adaptability of crops, but the natural polymorphisms in DNA methylation-related genes and their contributions to phenotypic variation in maize (Zea mays) remain elusive. Here, we show that a polymorphic 10-bp insertion/deletion variant in the 3'UTR of Zea methyltransferase2 (ZMET2) alters its transcript level and accounts for variation in the number of maize husk layers. ZMET2 encodes a chromomethylase and is required for maintaining genome-wide DNA methylation in the CHG sequence context. Disruption of ZMET2 increased the number of husk layers and resulted in thousands of differentially methylated regions, a proportion of which were also distinguishable in natural ZMET2 alleles. Population genetic analyses indicated that ZMET2 was a target of selection and might play a role in the spread of maize from tropical to temperate regions. Our results provide important insights into the natural variation of ZMET2 that confers both global and locus-specific effects on DNA methylation, which contribute to phenotypic diversity in maize.
Assuntos
Metilação de DNA , Proteínas de Plantas , Polimorfismo Genético , Zea mays , Zea mays/genética , Metilação de DNA/genética , Proteínas de Plantas/genética , Proteínas de Plantas/metabolismo , Regulação da Expressão Gênica de Plantas , Alelos , DNA (Citosina-5-)-MetiltransferasesRESUMO
ConspectusThe pursuit of novel and eco-friendly methods in organic synthesis is gaining prominence, with a strong emphasis on green transformations using renewable and sustainable resources. Among these environmentally conscious approaches, water-compatible reactions stand out for their many advantages. Water, as a solvent, offers unmatched abundance, cost-efficiency, and environmental compatibility compared to organic solvents. Its use eliminates the need for complex protection and deprotection steps for reactive functional groups in multistep synthesis and enables the use of water-soluble substrates like proteins and carbohydrates. Water-compatible reactions also provide opportunities to combine with enzymes, resulting in chemoenzymatic transformations that can increase efficiency. Additionally, these reactions facilitate site-specific modification and the bioconjugation of biomolecules, leading to bioconjugate therapeutics.Over nearly three decades, our research group has been dedicated to developing innovative water-compatible methodologies and concepts. This Account provides a comprehensive overview of our contributions since 1994. Our central strategy revolves around integrating green chemistry principles into our methods, focusing on (i) developing reactions that can operate under mild conditions, including room temperature, atmospheric pressure, and physiological pH; (ii) designing atom-economical reactions that minimize waste production; (iii) replacing toxic and flammable organic solvents with eco-friendly alternatives like water and ethanol; and (iv) reducing reliance on metals or halogenated compounds in specific reactions.In this Account, we detail our achievements in developing efficient methodologies in aqueous media, highlighting their scope, limitations, asymmetric control, and applications for synthesizing complex molecules and functionalizing peptides and proteins. Mechanistic investigations underlying these developments are also discussed when applicable. Furthermore, we offer insights into the reasoning behind our work and address future opportunities and challenges in this area of research. We hope that this Account will inspire continued interest and foster new breakthroughs. By exploring innovative and broadly applicable strategies that expand the water-compatible synthetic toolbox, we aim to pave the way for the truly green and sustainable synthesis of complex molecules and pharmaceuticals.
RESUMO
The NLR family pyrin domain containing 3 (NLRP3) inflammasome plays an important role in the pathogenesis of a wide variety of human diseases. So far, drugs directly and specifically targeting the NLRP3 inflammasome are not available for clinical use since the safety and efficacy of new compounds are often unclear. A promising approach is thus to identify NLRP3 inhibitors from existing drugs that are already in clinical use. Here, we show that mefloquine, a well-known antimalarial drug, is a highly selective and potent NLRP3 inhibitor by screening a FDA-approved drug library. Mechanistically, mefloquine directly binds to the NLRP3 NACHT and LRR domains to prevent NLRP3 inflammasome activation. More importantly, mefloquine treatment attenuates the symptoms of lipopolysaccharide-induced systemic inflammation and Parkinson's disease-like neural damage in mice. Our findings identify mefloquine as a potential therapeutic agent for NLRP3-driven diseases and migth expand its clinical use considerably.
RESUMO
AIMS: Although cefiderocol (FDC) is not prescribed in China, FDC-resistant pandrug-resistant hypervirulent Klebsiella pneumoniae (PDR-hvKp) is emerging. In this study, we performed FDC susceptibility testing of clinical Kp isolates to explore the prevalence of FDC-resistant isolates and the mechanism of FDC-resistance. METHODS: We retrospectively selected 151 carbapenem-resistant Kp isolates to assess FDC susceptibility. Seven isolates harboring blaSHV-12 from two patients were enrolled for whole-genome sequencing. The antimicrobial resistance, virulence, blaSHV-12 expression, and fitness costs in different media were examined. The amplification of blaSHV-12 was further investigated by qPCR and long-read sequencing. RESULTS: The 151 isolates showed a low MIC50/MIC90 (1/4 mg/L) of FDC. The seven isolates were ST11 PDR-hvKp, and two represented FDC-resistance (MIC=32 mg/L). The IncR/IncFII plasmids of two FDC-resistant isolates harbored 6 and 15 copies of blaSHV-12, whereas four FDC-susceptible isolates carried one copy and one harbored three copies. These blaSHV-12 genes concatenated together and were located within the same 7.3 kb fragment flanked by IS26, which contributed to the increased expression and FDC resistance without fitness costs. The amplification of blaSHV-12 and FDC resistance could be induced by FDC in vitro and reversed during continuous passage. CONCLUSIONS: The amplification of blaSHV-12 and the consequent dynamic within-host heteroresistance are important concerns for the rational application of antibiotics. Long-read sequencing might be a superior way to detect resistance gene amplification rapidly and accurately.
Assuntos
Infecções por Klebsiella , Klebsiella pneumoniae , Humanos , Klebsiella pneumoniae/genética , Cefiderocol , Estudos Retrospectivos , Infecções por Klebsiella/tratamento farmacológico , Infecções por Klebsiella/epidemiologia , Testes de Sensibilidade Microbiana , Antibacterianos/farmacologia , Antibacterianos/uso terapêuticoRESUMO
Upconversion nanoparticles are popular as imaging probes due to their advantages in photostability and controllable emission dimensions. However, upconversion polarization remains largely uncharted with previous reports limited to microstructures. In this work, we report the observation of polarized upconversion emissions from ß-NaYF4 single nanostructures below 100 nm. At the sub-100 nm scale, nanorods, nanodiscs, and nanoplates exhibit distinctive polarization degrees despite the same doping concentrations of lanthanides. We find this varied polarization degree results from the crystallographic orientation of nanostructure in relation to the light field and can be linked to the distinctive emission spectrum profile with varied Stark splitting transition ratios from Er3+. Our findings provide a comprehensive understanding of the polarization properties of upconversion nanoparticles, revealing a previously unexplored aspect of light emission. This discovery expands our knowledge of upconversion nanoparticles and also opens new possibilities for their use in future imaging and sensing applications, where polarization sensitivity is crucial.
RESUMO
Stationary energy storage infrastructure based on zinc-ion transport and storage chemistry is attracting more attention due to favorable metrics, including cost, safety, and recycling feasibility. However, splitting water and liquid electrolyte fluidity lead to cathode dissolution and Zn corrosion, resulting in rapid attenuation of the capacity and service life. Herein, a new architecture of solid-state electrolytes with high zinc ionic conductivity at room temperature was prepared via solidification of deep eutectic solvents utilizing MXene as nucleation additives. The ionic conductivity of MXene/ZCEs reached 6.69 × 10-4 S cm-1 at room temperature. Dendrite-free Zn plating/stripping with high reversibility can remain for over 2500 h. Subsequently, the fabricated solid-state zinc-ion battery with eliminated HER and suppressed Zn dendrites exhibited excellent cycling performance and could work normally in a range from -10 to 60 °C. This design inspired by eutectic solidification affords new insights into the multivalent solid electrochemistry suffering from slow ion migration.
RESUMO
AIMS/HYPOTHESIS: Mutations in Isl1, encoding the insulin enhancer-binding protein islet-1 (ISL1), may contribute to attenuated insulin secretion in type 2 diabetes mellitus. We made an Isl1E283D mouse model to investigate the disease-causing mechanism of diabetes mellitus. METHODS: The ISL1E283D mutation (c. 849A>T) was identified by whole exome sequencing on an early-onset type 2 diabetes family and then the Isl1E283D knockin (KI) mouse model was created and an IPGTT and IPITT were conducted. Glucose-stimulated insulin secretion (GSIS), expression of Ins2 and other ISL1 target genes and interacting proteins were evaluated in isolated pancreas islets. Transcriptional activity of Isl1E283D was evaluated by cell-based luciferase reporter assay and electrophoretic mobility shift assay, and the expression levels of Ins2 driven by Isl1 wild-type (Isl1WT) and Isl1E283D mutation in rat INS-1 cells were determined by RT-PCR and western blotting. RESULTS: Impaired GSIS and elevated glucose level were observed in Isl1E283D KI mice while expression of Ins2 and other ISL1 target genes Mafa, Pdx1, Slc2a2 and the interacting protein NeuroD1 were downregulated in isolated islets. Transcriptional activity of the Isl1E283D mutation for Ins2 was reduced by 59.3%, and resulted in a marked downregulation of Ins2 expression when it was overexpressed in INS-1 cells, while overexpression of Isl1WT led to an upregulation of Ins2 expression. CONCLUSIONS/INTERPRETATION: Isl1E283D mutation reduces insulin expression and secretion by regulating insulin and other target genes, as well as its interacting proteins such as NeuroD1, leading to the development of glucose intolerance in the KI mice, which recapitulated the human diabetic phenotype. This study identified and highlighted the Isl1E283D mutation as a novel causative factor for type 2 diabetes, and suggested that targeting transcription factor ISL1 could offer an innovative avenue for the precise treatment of human type 2 diabetes.
Assuntos
Diabetes Mellitus Tipo 2 , Proteínas com Homeodomínio LIM , Mutação de Sentido Incorreto , Fatores de Transcrição , Animais , Diabetes Mellitus Tipo 2/genética , Diabetes Mellitus Tipo 2/metabolismo , Proteínas com Homeodomínio LIM/genética , Proteínas com Homeodomínio LIM/metabolismo , Camundongos , Fatores de Transcrição/genética , Fatores de Transcrição/metabolismo , Humanos , Masculino , Insulina/metabolismo , Feminino , Ratos , Secreção de Insulina/genética , Ilhotas Pancreáticas/metabolismoRESUMO
Retinoic acid (RA), a vitamin A derivative, is an effective cell differentiating factor which plays critical roles in neuronal differentiation induction and the production of neurotransmitters in neurons. However, the specific changes in phosphorylation levels and downstream signalling pathways associated with RA remain unclear. This study employed qualitative and quantitative phosphoproteomics approaches based on mass spectrometry to investigate the phosphorylation changes induced by RA in C17.2 neural stem cells (NSCs). Dimethyl labelling, in conjunction with TiO2 phosphopeptide enrichment, was utilized to profile the phosphoproteome of self-renewing and RA-induced differentiated cells in C17.2 NSCs. The results of our study revealed that, qualitatively, 230 and 14 phosphoproteins were exclusively identified in the self-renewal and RA-induced groups respectively. Quantitatively, we successfully identified and quantified 177 unique phosphoproteins, among which 70 exhibited differential phosphorylation levels. Analysis of conserved phosphorylation motifs demonstrated enrichment of motifs corresponding to cyclin-dependent kinase and MAPK in the RA-induced group. Additionally, through a comprehensive literature and database survey, we found that the differentially expressed proteins were associated with the Wnt/ß-catenin and Hippo signalling pathways. This work sheds light on the changes in phosphorylation levels induced by RA in C17.2 NSCs, thereby expanding our understanding of the molecular mechanisms underlying RA-induced neuronal differentiation.
Assuntos
Células-Tronco Neurais , Tretinoína , Tretinoína/farmacologia , Tretinoína/metabolismo , Células-Tronco Neurais/metabolismo , Neurônios/metabolismo , Diferenciação Celular , Fosfoproteínas/genética , Fosfoproteínas/metabolismoRESUMO
The rapid and controlled synthesis of high-molecular-weight (HMW) polysarcosine (pSar), a potential polyethylene glycol (PEG) alternative, via the ring-opening polymerization (ROP) of N-carboxyanhydride (NCA) is rare and challenging. Here, we report the well-controlled ROP of sarcosine NCA (Sar-NCA) that is catalyzed by various carboxylic acids, which accelerate the polymerization rate up to 50 times, and enables the robust synthesis of pSar with an unprecedented ultrahigh molecular weight (UHMW) up to 586 kDa (DP ⼠8200) and exceptionally narrow dispersity (D̵) below 1.05. Mechanistic experiments and density functional theory calculations together elucidate the role of carboxylic acid as a bifunctional catalyst that significantly facilitates proton transfer processes and avoids charge separation and suggest the ring opening of NCA, rather than decarboxylation, as the rate-determining step. UHMW pSar demonstrates improved thermal and mechanical properties over the low-molecular-weight counterparts. This work provides a simple yet highly efficient approach to UHMW pSar and generates a new fundamental understanding useful not only for the ROP of Sar-NCA but also for other NCAs.
RESUMO
Tailoring the surface ligands of metal nanoclusters is important for engineering unique configurations of metal nanoclusters. Thiacalix[4]arene has found extensive applications in the construction of metal nanoclusters. In this investigation, we present the synthesis and characterization of the first all-calixarene-protected silver nanoclusters, [Ag(CH3CN)4]2[Ag44(BTCA)6] (Ag44, H4BTCA = p-tert-butylthiacalix[4]arene). Single-crystal X-ray structural analysis reveals that all silver atoms are in a face-centered cubic (fcc) arrangement. The formation of such an fcc structure is attributed to the selectively passivation on {100} facets by BTCA4-. Thiacalixarene substantially facilitates the stability of Ag44 due to its multiple coordination sites and bulkiness. Mass spectrometry and theoretical calculations reveal that Ag44 is a superatomic silver nanocluster with 22 free electrons in the following configuration: 1S21P61D61F22S21D4. This work not only elucidates the impact of macrocyclic ligands on the stabilization of silver clusters but also furnishes an approach for assembling atomically precise fcc nanoclusters.
RESUMO
Mesenchymal stem cells (MSCs) are ubiquitous multipotent cells exhibiting significant therapeutic potential for various diseases. It is generally accepted that clinical application requires massive expansion of MSCs, which is often accompanied by the occurrence of replicative senescence. Additionally, senescent MSCs exhibit significantly reduced proliferation, differentiation, and therapeutic potential. The scale-up of MSCs production and cellular senescence are major challenges for translational applications. This study first collected extracellular vesicles (EVs) from gingival MSCs (GMSCs) under hypoxia preconditioning combined with 3D dynamic culture (obtained EVs designed as H-3D-EVs). Subsequently, we further explored the effects and mechanisms of H-3D-EVs on aging-GMSCs. The results showed that H-3D-EVs improved the proliferation ability and cell activity of aging-GMSCs, and ameliorated their senescence. mRNA sequencing reveals transcriptomic changes in aging-GMSCs. It was found that H-3D-EVs up-regulated genes related to mitochondrial dynamics, cell cycle, and DNA repair, while down-regulated aging-related genes. Furthermore, we verified that H-3D-EVs corrected the mitochondrial dysfunction of aging-GMSCs by improving mitochondrial dynamics. In summary, this study provides a promising strategy for improving the culture methods of GMSCs and avoiding its senescence in large-scale production.
Assuntos
Senescência Celular , Vesículas Extracelulares , Células-Tronco Mesenquimais , Mitocôndrias , Vesículas Extracelulares/metabolismo , Células-Tronco Mesenquimais/metabolismo , Células-Tronco Mesenquimais/citologia , Mitocôndrias/metabolismo , Humanos , Hipóxia Celular , Células Cultivadas , Proliferação de Células , Envelhecimento/metabolismo , Envelhecimento/genética , Dinâmica MitocondrialRESUMO
BACKGROUND: The programmed death 1 inhibitor toripalimab plus the angio-immuno kinase inhibitor surufatinib showed a tolerable safety profile and preliminary efficacy in patients with advanced solid tumors in a phase I study. METHODS: This open-label, multi-cohort study in China enrolled patients with advanced solid tumors who had failed or were intolerable to standard treatment into tumor-specific cohorts. Patients received surufatinib (250 mg orally, once daily) plus toripalimab (240 mg intravenously, once every three weeks). Results for three cohorts (gastric/gastroesophageal junction [GC/GEJ] adenocarcinoma, esophageal squamous cell carcinoma [ESCC], and biliary tract carcinoma [BTC]) are reported here. The primary endpoint was investigator-assessed objective response rate (ORR) per Response Evaluation criteria in Solid Tumors version 1.1. RESULTS: Between December 17, 2019, and January 29, 2021, 60 patients were enrolled (GC/GEJ, n = 20; ESCC, n = 20; BTC, n = 20). At data cutoff (February 28, 2023), ORRs were 31.6%, 30.0%, and 11.1%, respectively. Median progression-free survival was 4.1, 2.7, and 2.9 months, respectively. Median overall survival was 13.7, 10.4, and 7.0 months, respectively. Overall, grade ≥ 3 treatment-related adverse events occurred in 28 (46.7%) patients. CONCLUSIONS: Surufatinib plus toripalimab showed promising antitumor activity and a tolerable safety profile in immunotherapy-naïve patients with GC/GEJ adenocarcinoma, ESCC, or BTC. These findings warrant further study in larger randomized trials comparing surufatinib plus toripalimab with standard therapies in these tumors. CLINICALTRIALS: gov NCT04169672.
Assuntos
Adenocarcinoma , Anticorpos Monoclonais Humanizados , Protocolos de Quimioterapia Combinada Antineoplásica , Neoplasias do Sistema Biliar , Neoplasias Esofágicas , Carcinoma de Células Escamosas do Esôfago , Humanos , Masculino , Feminino , Pessoa de Meia-Idade , Idoso , Anticorpos Monoclonais Humanizados/administração & dosagem , Anticorpos Monoclonais Humanizados/efeitos adversos , Anticorpos Monoclonais Humanizados/uso terapêutico , Protocolos de Quimioterapia Combinada Antineoplásica/uso terapêutico , Protocolos de Quimioterapia Combinada Antineoplásica/efeitos adversos , Neoplasias Esofágicas/tratamento farmacológico , Neoplasias Esofágicas/patologia , Neoplasias do Sistema Biliar/tratamento farmacológico , Neoplasias do Sistema Biliar/patologia , Neoplasias do Sistema Biliar/mortalidade , Adulto , Carcinoma de Células Escamosas do Esôfago/tratamento farmacológico , Carcinoma de Células Escamosas do Esôfago/patologia , Adenocarcinoma/tratamento farmacológico , Adenocarcinoma/patologia , Adenocarcinoma/mortalidade , Neoplasias Gástricas/tratamento farmacológico , Neoplasias Gástricas/mortalidade , Neoplasias Gástricas/patologia , Junção Esofagogástrica/patologia , Imidazóis/administração & dosagem , Imidazóis/uso terapêutico , Imidazóis/efeitos adversos , Idoso de 80 Anos ou mais , Estudos de CoortesRESUMO
Background Large language models (LLMs) hold substantial promise for medical imaging interpretation. However, there is a lack of studies on their feasibility in handling reasoning questions associated with medical diagnosis. Purpose To investigate the viability of leveraging three publicly available LLMs to enhance consistency and diagnostic accuracy in medical imaging based on standardized reporting, with pathology as the reference standard. Materials and Methods US images of thyroid nodules with pathologic results were retrospectively collected from a tertiary referral hospital between July 2022 and December 2022 and used to evaluate malignancy diagnoses generated by three LLMs-OpenAI's ChatGPT 3.5, ChatGPT 4.0, and Google's Bard. Inter- and intra-LLM agreement of diagnosis were evaluated. Then, diagnostic performance, including accuracy, sensitivity, specificity, and area under the receiver operating characteristic curve (AUC), was evaluated and compared for the LLMs and three interactive approaches: human reader combined with LLMs, image-to-text model combined with LLMs, and an end-to-end convolutional neural network model. Results A total of 1161 US images of thyroid nodules (498 benign, 663 malignant) from 725 patients (mean age, 42.2 years ± 14.1 [SD]; 516 women) were evaluated. ChatGPT 4.0 and Bard displayed substantial to almost perfect intra-LLM agreement (κ range, 0.65-0.86 [95% CI: 0.64, 0.86]), while ChatGPT 3.5 showed fair to substantial agreement (κ range, 0.36-0.68 [95% CI: 0.36, 0.68]). ChatGPT 4.0 had an accuracy of 78%-86% (95% CI: 76%, 88%) and sensitivity of 86%-95% (95% CI: 83%, 96%), compared with 74%-86% (95% CI: 71%, 88%) and 74%-91% (95% CI: 71%, 93%), respectively, for Bard. Moreover, with ChatGPT 4.0, the image-to-text-LLM strategy exhibited an AUC (0.83 [95% CI: 0.80, 0.85]) and accuracy (84% [95% CI: 82%, 86%]) comparable to those of the human-LLM interaction strategy with two senior readers and one junior reader and exceeding those of the human-LLM interaction strategy with one junior reader. Conclusion LLMs, particularly integrated with image-to-text approaches, show potential in enhancing diagnostic medical imaging. ChatGPT 4.0 was optimal for consistency and diagnostic accuracy when compared with Bard and ChatGPT 3.5. © RSNA, 2024 Supplemental material is available for this article.
Assuntos
Nódulo da Glândula Tireoide , Humanos , Feminino , Adulto , Nódulo da Glândula Tireoide/diagnóstico por imagem , Estudos Retrospectivos , Idioma , Redes Neurais de Computação , Curva ROCRESUMO
Background Noninvasive tests can be used to screen patients with chronic liver disease for advanced liver fibrosis; however, the use of single tests may not be adequate. Purpose To construct sequential clinical algorithms that include a US deep learning (DL) model and compare their ability to predict advanced liver fibrosis with that of other noninvasive tests. Materials and Methods This retrospective study included adult patients with a history of chronic liver disease or unexplained abnormal liver function test results who underwent B-mode US of the liver between January 2014 and September 2022 at three health care facilities. A US-based DL network (FIB-Net) was trained on US images to predict whether the shear-wave elastography (SWE) value was 8.7 kPa or higher, indicative of advanced fibrosis. In the internal and external test sets, a two-step algorithm (Two-step#1) using the Fibrosis-4 Index (FIB-4) followed by FIB-Net and a three-step algorithm (Three-step#1) using FIB-4 followed by FIB-Net and SWE were used to simulate screening scenarios where liver stiffness measurements were not or were available, respectively. Measures of diagnostic accuracy were calculated using liver biopsy as the reference standard and compared between FIB-4, SWE, FIB-Net, and European Association for the Study of the Liver guidelines (ie, FIB-4 followed by SWE), along with sequential algorithms. Results The training, validation, and test data sets included 3067 (median age, 42 years [IQR, 33-53 years]; 2083 male), 1599 (median age, 41 years [IQR, 33-51 years]; 1124 male), and 1228 (median age, 44 years [IQR, 33-55 years]; 741 male) patients, respectively. FIB-Net obtained a noninferior specificity with a margin of 5% (P < .001) compared with SWE (80% vs 82%). The Two-step#1 algorithm showed higher specificity and positive predictive value (PPV) than FIB-4 (specificity, 79% vs 57%; PPV, 44% vs 32%) while reducing unnecessary referrals by 42%. The Three-step#1 algorithm had higher specificity and PPV compared with European Association for the Study of the Liver guidelines (specificity, 94% vs 88%; PPV, 73% vs 64%) while reducing unnecessary referrals by 35%. Conclusion A sequential algorithm combining FIB-4 and a US DL model showed higher diagnostic accuracy and improved referral management for all-cause advanced liver fibrosis compared with FIB-4 or the DL model alone. © RSNA, 2024 Supplemental material is available for this article. See also the editorial by Ghosh in this issue.