Explore the vast range of titles available.

Mesothelioma is a highly lethal and poorly biologically understood disease which presents diagnostic challenges due to its morphological complexity. This study uses self-supervised AI (Artificial Intelligence) to map the histomorphological landscape of the disease. The resulting atlas consists of recurrent patterns identified from 3446 Hematoxylin and Eosin (H&E) stained images scanned from resected tumour slides. These patterns generate highly interpretable predictions, achieving state-of-the-art performance with 0.65 concordance index (c-index) for outcomes and 88% AUC in subtyping. Their clinical relevance is endorsed by comprehensive human pathological assessment. Furthermore, we characterise the molecular underpinnings of these diverse, meaningful, predictive patterns. Our approach both improves diagnosis and deepens our understanding of mesothelioma biology, highlighting the power of this self-learning method in clinical applications and scientific discovery. Mesothelioma is a highly lethal cancer that remains challenging to diagnose. Here, the authors curate a histomorphological atlas of resected mesothelioma and map it using self-supervised AI endorsed by human pathological assessment, revealing patterns that generate highly interpretable predictions.

Malignant mesothelioma (MpM) is an aggressive, invariably fatal tumour that is causally linked with asbestos exposure. The disease primarily results from loss of tumour suppressor gene function and there are no ‘druggable’ driver oncogenes associated with MpM. To identify opportunities for management of this disease we have carried out polysome profiling to define the MpM translatome. We show that in MpM there is a selective increase in the translation of mRNAs encoding proteins required for ribosome assembly and mitochondrial biogenesis. This results in an enhanced rate of mRNA translation, abnormal mitochondrial morphology and oxygen consumption, and a reprogramming of metabolic outputs. These alterations delimit the cellular capacity for protein biosynthesis, accelerate growth and drive disease progression. Importantly, we show that inhibition of mRNA translation, particularly through combined pharmacological targeting of mTORC1 and 2, reverses these changes and inhibits malignant cell growth in vitro and in ex-vivo tumour tissue from patients with end-stage disease. Critically, we show that these pharmacological interventions prolong survival in animal models of asbestos-induced mesothelioma, providing the basis for a targeted, viable therapeutic option for patients with this incurable disease. Treating malignant pleural mesothelioma (MpM) is challenging due to a lack of druggable genes, but other molecular features could be clinically useful. Here the authors profile mRNA translation dysregulation in MpM cell lines using polysome profiling, and identify mTORC1 and 2 as potential pharmacological targets.

Malignant pleural mesothelioma (MpM) is a lethal tumour closely linked to asbestos exposure and is a cancer of unmet clinical need with no known oncogenic drivers. Recent advancements in technologies to identify RNA-binding proteins (RBPs) has uncovered an emerging role for RBP-RNA interactions in cancer progression and we therefore assessed changes in the RBPome of patient-derived MpM cell lines. We identify over 350 RBPs showing altered RNA binding, with functions consistent with key cancer hallmarks, and we identified YBX3 as a potential oncoprotein driving cell proliferation in MpM. Mechanistically we show the impact of YBX3 on cell growth is achieved through its control of the expression of the amino acid transporter SLC7A5/LAT1, with increased amino acid uptake increasing protein synthesis rates. Notably, we show the inhibition of cell growth by YBX3 deletion is recapitulated by the clinically relevant SLC7A5/LAT1 inhibitor JPH203. Finally, we demonstrate that JPH203 sensitizes MpM cells to radiotherapy, which could provide a promising therapeutic strategy for MpM.Competing Interest StatementThe authors have declared no competing interest.

Applications of artificial intelligence (AI) to histopathology are now common, but most require supervision which inherently limits their scope. By using self-supervised learning (SSL), we discover and quantify the full range of histopathological appearances in a disease, and associate them with clinicopathological ground truths such as prognosis. We used this approach to discover under-appreciated morphologies of lung adenocarcinoma (LUAD), using a highly characterised resected tumour cohort of over 4000 slides from over 1000 patients. By constructing an authoritative lexicon of recurrent LUAD appearances, we ab initio discovered several stromal morphologies strongly predictive of outcome. With multimodal data integration and external dataset validation, we propose that epithelial discohesion is lethal, but only in the context of immunologically cold stroma. Both these morphological features are independent of current prognostic schema. Crucially, we describe these features in the context of real-world diagnostic histopathology, giving them immediate clinical translatability. Histopathology relies heavily on epithelial morphology, often neglecting the stroma. We use self-supervised AI to identify under-appreciated morphologies in LUAD linked to poor outcome and validate these observations in an external cohort, demonstrating the utility of self-supervised AI as a powerful biological discovery tool.

Mesothelioma is a highly lethal and poorly biologically understood disease which presents diagnostic challenges due to its morphological complexity. This study uses self-supervised AI (Artificial Intelligence) to map the histomorphological landscape of the disease. The resulting atlas consists of recurrent patterns identified from 3446 Hematoxylin and Eosin (H&E) stained images scanned from resected tumour slides. These patterns generate highly interpretable predictions, achieving state-of-the-art performance with 0.65 concordance index (c-index) for outcomes and 85% AUC in subtyping. Their clinical relevance is endorsed by comprehensive human pathological assessment. Furthermore, we characterise the molecular underpinnings of these diverse, meaningful, predictive patterns. Our approach both improves diagnosis and deepens our understanding of mesothelioma biology, highlighting the power of this self-learning method in clinical applications and scientific discovery.Competing Interest StatementThe authors have declared no competing interest.Footnotes* The text and figures are slightly updated.

Mrs. Maertz' class made seaweed pancakes when they were learning about whales because it was a favorite breakfast food for crews of whaling ships in the 1800s. This recipe is adapted from \"A Guide to Some of the Common Seaweeds\" by G. McCarthy and K. Ehlingis. Although these pancakes don't taste anyting like the kind we are used to eating today, seaweed is a good source of vitamins: