Deep Learning Predicts Lymphovascular Invasion Status in Muscle Invasive Bladder Cancer Histopathology.

BACKGROUND: Lymphovascular invasion (LVI) is linked to poor prognosis in patients with muscle-invasive bladder cancer (MIBC). Accurately identifying the LVI status in MIBC patients is crucial for effective risk stratification and precision treatment. We aim to develop a deep learning model to identify the LVI status in whole-slide images (WSIs) of MIBC patients. PATIENTS AND METHODS: A cohort from The Cancer Genome Atlas (TCGA) database was used to train a deep learning model, slide-based lymphovascular invasion predictor (SBLVIP), based on multiple-instance learning. This model was externally validated using the Renmin Hospital of Wuhan University (RHWU) and People's Hospital of Hanchuan City (PHHC) cohorts. Kaplan-Meier curves, along with univariate and multivariate Cox models, were employed to evaluate the association between the LVI status predicted by SBLVIP and the survival outcomes of MIBC patients. RESULTS: In the TCGA cohort, the SBLVIP model achieved an average accuracy of 0.804 [95% confidence interval (CI) 0.712-0.895] and an area under the receiver operating characteristic curve (AUC) of 0.77 (95% CI 0.63-0.84) in the training set. In the internal validation set, the model's average accuracy and AUC were 0.774 (95% CI, 0.701-0.846) and 0.76 (95% CI, 0.60-0.83), respectively. In the RHWU cohort, the SBLVIP model achieved an average accuracy of 0.807 (95% CI 0.734-0.880) and an AUC of 0.74 (95% CI 0.55-0.83). In the PHHC cohort, SBLVIP demonstrated an average accuracy of 0.821 (95% CI 0.737-0.909) and an AUC of 0.74 (95% CI 0.58-0.89). Moreover, the LVI status predicted by SBLVIP showed significant independent prognostic value (P = 1 × 10-6). CONCLUSIONS: We developed a deep learning model named SBLVIP to predict the LVI status in routine WSIs of MIBC patients.

UroAngel: a single-kidney function prediction system based on computed tomography urography using deep learning.

BACKGROUND: Accurate estimation of the glomerular filtration rate (GFR) is clinically crucial for determining the status of obstruction, developing treatment strategies, and predicting prognosis in obstructive nephropathy (ON). We aimed to develop a deep learning-based system, named UroAngel, for non-invasive and convenient prediction of single-kidney function level. METHODS: We retrospectively collected computed tomography urography (CTU) images and emission computed tomography diagnostic reports of 520 ON patients. A 3D U-Net model was used to segment the renal parenchyma, and a logistic regression multi-classification model was used to predict renal function level. We compared the predictive performance of UroAngel with the Modification of Diet in Renal Disease (MDRD), Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equations, and two expert radiologists in an additional 40 ON patients to validate clinical effectiveness. RESULTS: UroAngel based on 3D U-Net convolutional neural network could segment the renal cortex accurately, with a Dice similarity coefficient of 0.861. Using the segmented renal cortex to predict renal function stage had high performance with an accuracy of 0.918, outperforming MDRD and CKD-EPI and two radiologists. CONCLUSIONS: We proposed an automated 3D U-Net-based analysis system for direct prediction of single-kidney function stage from CTU images. UroAngel could accurately predict single-kidney function in ON patients, providing a novel, reliable, convenient, and non-invasive method.

Machine Learning Quantified Tumor-Stroma Ratio Is an Independent Prognosticator in Muscle-Invasive Bladder Cancer.

Although the tumor-stroma ratio (TSR) has prognostic value in many cancers, the traditional semi-quantitative visual assessment method has inter-observer variability, making it impossible for clinical practice. We aimed to develop a machine learning (ML) algorithm for accurately quantifying TSR in hematoxylin-and-eosin (H&E)-stained whole slide images (WSI) and further investigate its prognostic effect in patients with muscle-invasive bladder cancer (MIBC). We used an optimal cell classifier previously built based on QuPath open-source software and ML algorithm for quantitative calculation of TSR. We retrospectively analyzed data from two independent cohorts to verify the prognostic significance of ML-based TSR in MIBC patients. WSIs from 133 MIBC patients were used as the discovery set to identify the optimal association of TSR with patient survival outcomes. Furthermore, we performed validation in an independent external cohort consisting of 261 MIBC patients. We demonstrated a significant prognostic association of ML-based TSR with survival outcomes in MIBC patients (p < 0.001 for all comparisons), with higher TSR associated with better prognosis. Uni- and multivariate Cox regression analyses showed that TSR was independently associated with overall survival (p < 0.001 for all analyses) after adjusting for clinicopathological factors including age, gender, and pathologic stage. TSR was found to be a strong prognostic factor that was not redundant with the existing staging system in different subgroup analyses (p < 0.05 for all analyses). Finally, the expression of six genes (DACH1, DEEND2A, NOTCH4, DTWD1, TAF6L, and MARCHF5) were significantly associated with TSR, revealing possible potential biological relevance. In conclusion, we developed an ML algorithm based on WSIs of MIBC patients to accurately quantify TSR and demonstrated its prognostic validity for MIBC patients in two independent cohorts. This objective quantitative method allows application in clinical practice while reducing the workload of pathologists. Thus, it might be of significant aid in promoting precise pathology services in MIBC.

Targeting Enteropeptidase with Reversible Covalent Inhibitors To Achieve Metabolic Benefits.

Inhibition of the serine protease enteropeptidase (EP) opens a new avenue to the discovery of chemotherapeutics for the treatment of metabolic diseases. Camostat has been used clinically for treating chronic pancreatitis in Japan; however, the mechanistic basis of the observed clinical efficacy has not been fully elucidated. We demonstrate that camostat is a potent reversible covalent inhibitor of EP, with an inhibition potency (k inact/KI) of 1.5 × 104 M-1s-1 High-resolution liquid chromatography-mass spectrometry (LC-MS) showed addition of 161.6 Da to EP after the reaction with camostat, consistent with insertion of the carboxyphenylguanidine moiety of camostat. Covalent inhibition of EP by camostat is reversible, with an enzyme reactivation half-life of 14.3 hours. Formation of a covalent adduct was further supported by a crystal structure resolved to 2.19 Å, showing modification of the catalytic serine of EP by a close analog of camostat, leading to formation of the carboxyphenylguanidine acyl enzyme identical to that expected for the reaction with camostat. Of particular note, minor structural modifications of camostat led to changes in the mechanism of inhibition. We observed from other studies that sustained inhibition of EP is required to effect a reduction in cumulative food intake and body weight, with concomitant improved blood glucose levels in obese and diabetic leptin-deficient mice. Thus, the structure-activity relationship needs to be driven by not only the inhibition potency but also the mechanistic and kinetic characterization. Our findings support EP as a target for the treatment of metabolic diseases and demonstrate that reversible covalent EP inhibitors show clinically relevant efficacy. SIGNIFICANCE STATEMENT: Interest in targeted covalent drugs has expanded in recent years, particularly so for kinase targets, but also more broadly. This study demonstrates that reversible covalent inhibition of the serine protease enteropeptidase is a therapeutically viable approach to the treatment of metabolic diseases and that mechanistic details of inhibition are relevant to clinical efficacy. Our mechanistic and kinetic studies outline a framework for detailed inhibitor characterization that is proving essential in guiding discovery efforts in this area.

Oxysterols direct B-cell migration through EBI2.

EBI2 (also called GPR183) is an orphan G-protein-coupled receptor that is highly expressed in spleen and upregulated upon Epstein-Barr-virus infection. Recent studies indicated that this receptor controls follicular B-cell migration and T-cell-dependent antibody production. Oxysterols elicit profound effects on immune and inflammatory responses as well as on cholesterol metabolism. The biological effects of oxysterols have largely been credited to the activation of nuclear hormone receptors. Here we isolate oxysterols from porcine spleen extracts and show that they are endogenous ligands for EBI2. The most potent ligand and activator is 7α,25-dihydroxycholesterol (OHC), with a dissociation constant of 450 pM for EBI2. In vitro, 7α,25-OHC stimulated the migration of EBI2-expressing mouse B and T cells with half-maximum effective concentration values around 500 pM, but had no effect on EBI2-deficient cells. In vivo, EBI2-deficient B cells or normal B cells desensitized by 7α,25-OHC pre-treatment showed reduced homing to follicular areas of the spleen. Blocking the synthesis of 7α,25-OHC in vivo with clotrimazole, a CYP7B1 inhibitor, reduced the content of 7α,25-OHC in the mouse spleen and promoted the migration of adoptively transferred pre-activated B cells to the T/B boundary (the boundary between the T-zone and B-zone in the spleen follicle), mimicking the phenotype of pre-activated B cells from EBI2-deficient mice. Our results show an unexpected causal link between EBI2, an orphan G-protein-coupled receptor controlling B-cell migration, and the known immunological effects of certain oxysterols, thus uncovering a previously unknown role for this class of molecules.

Oxysterols are agonist ligands of RORγt and drive Th17 cell differentiation.

The RAR-related orphan receptor gamma t (RORγt) is a nuclear receptor required for generating IL-17-producing CD4(+) Th17 T cells, which are essential in host defense and may play key pathogenic roles in autoimmune diseases. Oxysterols elicit profound effects on immune and inflammatory responses as well as on cholesterol and lipid metabolism. Here, we describe the identification of several naturally occurring oxysterols as RORγt agonists. The most potent and selective activator for RORγt is 7ß, 27-dihydroxycholesterol (7ß, 27-OHC). We show that these oxysterols reverse the inhibitory effect of an RORγt antagonist, ursolic acid, in RORγ- or RORγt-dependent cell-based reporter assays. These ligands bind directly to recombinant RORγ ligand binding domain (LBD), promote recruitment of a coactivator peptide, and reduce binding of a corepressor peptide to RORγ LBD. In primary cells, 7ß, 27-OHC and 7α, 27-OHC enhance the differentiation of murine and human IL-17-producing Th17 cells in an RORγt-dependent manner. Importantly, we showed that Th17, but not Th1 cells, preferentially produce these two oxysterols. In vivo, administration of 7ß, 27-OHC in mice enhanced IL-17 production. Mice deficient in CYP27A1, a key enzyme in generating these oxysterols, showed significant reduction of IL-17-producing cells, including CD4(+) and γδ(+) T cells, similar to the deficiency observed in RORγt knockout mice. Our results reveal a previously unknown mechanism for selected oxysterols as immune modulators and a direct role for CYP27A1 in generating these RORγt agonist ligands, which we propose as RORγt endogenous ligands, driving both innate and adaptive IL-17-dependent immune responses.

Fluoro analogs of bioactive oxy-sterols: Synthesis of an EBI2 agonist with enhanced metabolic stability.

Synthesis of several 7-hydroxy oxysterols and their potential roles as signaling molecules in the innate and adaptive immune responses is discussed. Discovery of a new, fluorinated, synthetic analog of the 7α,25-dihydroxycholesterol-the endogenous ligand of GPR 183 (EBI2), a G-protein coupled receptor highly expressed upon Epstein-Barr virus infection is described. Fluoro oxysterol 12 showed good metabolic stability while maintaining excellent EBI2 agonist activity.

Phenotyping drug polypharmacology via eicosanoid profiling of blood.

It is widely accepted that small-molecule drugs, despite their selectivity at primary targets, exert pharmacological effects (and safety liabilities) through a multiplicity of pathways. As such, it has proved extremely difficult to experimentally assess polypharmacology in an agnostic fashion. Profiling of metabolites produced as part of physiological responses to pharmacological stimuli provides a unique opportunity to explore drug pharmacology. A total of 122 eicosanoid lipids in human whole blood were monitored from 10 different donors upon stimulation with several inducers of immunological responses and treatment with modulators of prostaglandin (PG) and leukotriene biosynthesis, including clinical and investigational molecules. Such analysis revealed differentiation between drugs nominally targeting different eicosanoid biosynthetic enzymes, or even those designed to target the same enzyme. Profiled agents, some of them marketed products, affect eicosanoid biosynthesis in ways that cannot be predicted from information on their intended targets. As an example, we used this platform to discriminate drugs based on their ability to silence PG biosynthesis in response to bacterial lipopolysaccharide, resulting in differential pharmacological activity in an in vivo model of endotoxemia. Some of the observed effects are subject to variability among individuals, indicating a potential application of this methodology to the patient stratification, based on their responses to benchmark drugs and experimental compounds read on the eicosanome via a simple blood test.

GPR139, an Orphan Receptor Highly Enriched in the Habenula and Septum, Is Activated by the Essential Amino Acids L-Tryptophan and L-Phenylalanine.

GPR139 is an orphan G-protein-coupled receptor expressed in the central nervous system. To identify its physiologic ligand, we measured GPR139 receptor activity from recombinant cells after treatment with amino acids, orphan ligands, serum, and tissue extracts. GPR139 activity was measured using guanosine 5'-O-(3-[(35)S]thio)-triphosphate binding, calcium mobilization, and extracellular signal-regulated kinases phosphorylation assays. Amino acids L-tryptophan (L-Trp) and L-phenylalanine (L-Phe) activated GPR139, with EC50 values in the 30- to 300-µM range, consistent with the physiologic concentrations of L-Trp and L-Phe in tissues. Chromatography of rat brain, rat serum, and human serum extracts revealed two peaks of GPR139 activity, which corresponded to the elution peaks of L-Trp and L-Phe. With the purpose of identifying novel tools to study GPR139 function, a high-throughput screening campaign led to the identification of a selective small-molecule agonist [JNJ-63533054, (S)-3-chloro-N-(2-oxo-2-((1-phenylethyl)amino)ethyl) benzamide]. The tritium-labeled JNJ-63533054 bound to cell membranes expressing GPR139 and could be specifically displaced by L-Trp and L-Phe. Sequence alignment revealed that GPR139 is highly conserved across species, and RNA sequencing studies of rat and human tissues indicated its exclusive expression in the brain and pituitary gland. Immunohistochemical analysis showed specific expression of the receptor in circumventricular regions of the habenula and septum in mice. Together, these findings suggest that L-Trp and L-Phe are candidate physiologic ligands for GPR139, and we hypothesize that this receptor may act as a sensor to detect dynamic changes of L-Trp and L-Phe in the brain.

Prediction of HER2 Status Based on Deep Learning in H&E-Stained Histopathology Images of Bladder Cancer.

Epidermal growth factor receptor 2 (HER2) has been widely recognized as one of the targets for bladder cancer immunotherapy. The key to implementing personalized treatment for bladder cancer patients lies in achieving rapid and accurate diagnosis. To tackle this challenge, we have pioneered the application of deep learning techniques to predict HER2 expression status from H&E-stained pathological images of bladder cancer, bypassing the need for intricate IHC staining or high-throughput sequencing methods. Our model, when subjected to rigorous testing within the cohort from the People's Hospital of Wuhan University, which encompasses 106 cases, has exhibited commendable performance on both the validation and test datasets. Specifically, the validation set yielded an AUC of 0.92, an accuracy of 0.86, a sensitivity of 0.87, a specificity of 0.83, and an F1 score of 86.7%. The corresponding metrics for the test set were 0.88 for AUC, 0.67 for accuracy, 0.56 for sensitivity, 0.75 for specificity, and 77.8% for F1 score. Additionally, in a direct comparison with pathologists, our model demonstrated statistically superior performance, with a p-value less than 0.05, highlighting its potential as a powerful diagnostic tool.

A highly efficient, high-throughput lipidomics platform for the quantitative detection of eicosanoids in human whole blood.

We have developed an ultra-performance liquid chromatography-multiple reaction monitoring/mass spectrometry (UPLC-MRM/MS)-based, high-content, high-throughput platform that enables simultaneous profiling of multiple lipids produced ex vivo in human whole blood (HWB) on treatment with calcium ionophore and its modulation with pharmacological agents. HWB samples were processed in a 96-well plate format compatible with high-throughput sample processing instrumentation. We employed a scheduled MRM (sMRM) method, with a triple-quadrupole mass spectrometer coupled to a UPLC system, to measure absolute amounts of 122 distinct eicosanoids using deuterated internal standards. In a 6.5-min run, we resolved and detected with high sensitivity (lower limit of quantification in the range of 0.4-460 pg) all targeted analytes from a very small HWB sample (2.5 µl). Approximately 90% of the analytes exhibited a dynamic range exceeding 1000. We also developed a tailored software package that dramatically sped up the overall data quantification and analysis process with superior consistency and accuracy. Matrix effects from HWB and precision of the calibration curve were evaluated using this newly developed automation tool. This platform was successfully applied to the global quantification of changes on all 122 eicosanoids in HWB samples from healthy donors in response to calcium ionophore stimulation.

Genome-wide mapping of RNA Pol-II promoter usage in mouse tissues by ChIP-seq.

Alternative promoters that are differentially used in various cellular contexts and tissue types add to the transcriptional complexity in mammalian genome. Identification of alternative promoters and the annotation of their activity in different tissues is one of the major challenges in understanding the transcriptional regulation of the mammalian genes and their isoforms. To determine the use of alternative promoters in different tissues, we performed ChIP-seq experiments using antibody against RNA Pol-II, in five adult mouse tissues (brain, liver, lung, spleen and kidney). Our analysis identified 38 639 Pol-II promoters, including 12 270 novel promoters, for both protein coding and non-coding mouse genes. Of these, 6384 promoters are tissue specific which are CpG poor and we find that only 34% of the novel promoters are located in CpG-rich regions, suggesting that novel promoters are mostly tissue specific. By identifying the Pol-II bound promoter(s) of each annotated gene in a given tissue, we found that 37% of the protein coding genes use alternative promoters in the five mouse tissues. The promoter annotations and ChIP-seq data presented here will aid ongoing efforts of characterizing gene regulatory regions in mammalian genomes.

Cyanidin-3-O-glucoside plays a protective role against renal ischemia/ reperfusion injury via the JAK/STAT pathway.

PURPOSE: To investigate the role of cyanidin-3-O-glucoside (C3G) in renal ischemia/reperfusion (I/R) injury and the potential mechanisms. METHODS: Mouse models were established by clamping the left renal vessels, and in vitro cellular models were established by hypoxic reoxygenation. RESULTS: Renal dysfunction and tissue structural damage were significantly higher in the I/R group. After treatment with different concentrations of C3G, the levels of renal dysfunction and tissue structural damage decreased at different levels. And its protective effect was most pronounced at 200 mg/kg. The use of C3G reduced apoptosis as well as the expression of endoplasmic reticulum stress (ERS)-related proteins. Hypoxia/reoxygenation (H/R)-induced apoptosis and ERS are dependent on oxidative stress in vitro. In addition, both AG490 and C3G inhibited the activation of JAK/STAT pathway and attenuated oxidative stress, ischemia-induced apoptosis and ERS. CONCLUSIONS: The results demonstrated that C3G blocked renal apoptosis and ERS protein expression by preventing reactive oxygen species (ROS) production after I/R via the JAK/STAT pathway, suggesting that C3G may be a potential therapeutic agent for renal I/R injury.

Predicting Lymph Node Metastasis Status from Primary Muscle-Invasive Bladder Cancer Histology Slides Using Deep Learning: A Retrospective Multicenter Study.

BACKGROUND: Accurate prediction of lymph node metastasis (LNM) status in patients with muscle-invasive bladder cancer (MIBC) before radical cystectomy can guide the use of neoadjuvant chemotherapy and the extent of pelvic lymph node dissection. We aimed to develop and validate a weakly-supervised deep learning model to predict LNM status from digitized histopathological slides in MIBC. METHODS: We trained a multiple instance learning model with an attention mechanism (namely SBLNP) from a cohort of 323 patients in the TCGA cohort. In parallel, we collected corresponding clinical information to construct a logistic regression model. Subsequently, the score predicted by the SBLNP was incorporated into the logistic regression model. In total, 417 WSIs from 139 patients in the RHWU cohort and 230 WSIs from 78 patients in the PHHC cohort were used as independent external validation sets. RESULTS: In the TCGA cohort, the SBLNP achieved an AUROC of 0.811 (95% confidence interval [CI], 0.771-0.855), the clinical classifier achieved an AUROC of 0.697 (95% CI, 0.661-0.728) and the combined classifier yielded an improvement to 0.864 (95% CI, 0.827-0.906). Encouragingly, the SBLNP still maintained high performance in the RHWU cohort and PHHC cohort, with an AUROC of 0.762 (95% CI, 0.725-0.801) and 0.746 (95% CI, 0.687-0.799), respectively. Moreover, the interpretability of SBLNP identified stroma with lymphocytic inflammation as a key feature of predicting LNM presence. CONCLUSIONS: Our proposed weakly-supervised deep learning model can predict the LNM status of MIBC patients from routine WSIs, demonstrating decent generalization performance and holding promise for clinical implementation.

Identification of structural motifs critical for epstein-barr virus-induced molecule 2 function and homology modeling of the ligand docking site.

Epstein-Barr virus-induced molecule 2 (EBI2) (also known as G-protein-coupled receptor 183) is a G-protein-coupled receptor (GPCR) that is best known for its role in B cell migration and localization. Our recent deorphanization effort led to the discovery of 7α,25-dihydroxycholesterol (7α,25-OHC) as the endogenous ligand for EBI2, which provides a tool for mechanistic studies of EBI2 function. Because EBI2 is the first GPCR known to bind and to be activated by an oxysterol, the goal of this study was to understand the molecular and structural bases for its ligand-dependent activation; this was achieved by identifying structural moieties in EBI2 or in 7α,25-OHC that might affect receptor-ligand interactions. By using a series of chemically related OHC analogs, we demonstrated that all three hydroxyl groups in 7α,25-OHC contributed to ligand-induced activation of the receptor. To determine the location and composition of the ligand binding domain in EBI2, we used a site-directed mutagenesis approach and generated mutant receptors with single amino acid substitutions at selected positions of interest. Biochemical and pharmacological profiling of these mutant receptors allowed for structure-function analyses and revealed critical motifs that likely interact with 7α,25-OHC. By using a hybrid ß(2)-adrenergic receptor-C-X-C chemokine receptor type 4 structure as a template, we created a homology model for EBI2 and optimized the docking of 7α,25-OHC into the putative ligand binding site, so that the hydroxyl groups interact with residues Arg87, Asn114, and Glu183. This model of ligand docking yields important structural insight into the molecular mechanisms mediating EBI2 function and may facilitate future efforts to design novel therapeutic agents that target EBI2.

Automatic analysis of quantitative NMR data of pharmaceutical compound libraries.

In drug discovery, chemical library compounds are usually dissolved in DMSO at a certain concentration and then distributed to biologists for target screening. Quantitative (1)H NMR (qNMR) is the preferred method for the determination of the actual concentrations of compounds because the relative single proton peak areas of two chemical species represent the relative molar concentrations of the two compounds, that is, the compound of interest and a calibrant. Thus, an analyte concentration can be determined using a calibration compound at a known concentration. One particularly time-consuming step in the qNMR analysis of compound libraries is the manual integration of peaks. In this report is presented an automated method for performing this task without prior knowledge of compound structures and by using an external calibration spectrum. The script for automated integration is fast and adaptable to large-scale data sets, eliminating the need for manual integration in ~80% of the cases.

3,5-Dihydroxybenzoic acid, a specific agonist for hydroxycarboxylic acid 1, inhibits lipolysis in adipocytes.

Niacin raises high-density lipoprotein and lowers low-density lipoprotein through the activation of the ß-hydroxybutyrate receptor hydroxycarboxylic acid 2 (HCA2) (aka GPR109a) but with an unwanted side effect of cutaneous flushing caused by vascular dilation because of the stimulation of HCA2 receptors in Langerhans cells in skin. HCA1 (aka GPR81), predominantly expressed in adipocytes, was recently identified as a receptor for lactate. Activation of HCA1 in adipocytes by lactate results in the inhibition of lipolysis, suggesting that agonists for HCA1 may be useful for the treatment of dyslipidemia. Lactate is a metabolite of glucose, suggesting that HCA1 may also be involved in the regulation of glucose metabolism. The low potency of lactate to activate HCA1, coupled with its fast turnover rate in vivo, render it an inadequate tool for studying the biological role of lactate/HCA1 in vivo. In this article, we demonstrate the identification of 3-hydroxybenzoic acid (3-HBA) as an agonist for both HCA2 and HCA1, whereas 3,5-dihydroxybenzoic acid (3,5-DHBA) is a specific agonist for only HCA1 (EC(50) ∼150 µM). 3,5-DHBA inhibits lipolysis in wild-type mouse adipocytes but not in HCA1-deficient adipocytes. Therefore, 3,5-DHBA is a useful tool for the in vivo study of HCA1 function and offers a base for further HCA1 agonist design. Because 3-HBA and 3,5-DHBA are polyphenolic acids found in many natural products, such as fruits, berries, and coffee, it is intriguing to speculate that other heretofore undiscovered natural substances may have therapeutic benefits.

Quantitative Assessment of Tumor-Infiltrating Lymphocytes Using Machine Learning Predicts Survival in Muscle-Invasive Bladder Cancer.

(1) Purpose: Although assessment of tumor-infiltrating lymphocytes (TILs) has been acknowledged to have important predictive prognostic value in muscle-invasive bladder cancer (MIBC), it is limited by inter- and intra-observer variability, hampering widespread clinical application. We aimed to evaluate the prognostic value of quantitative TILs score based on a machine learning (ML) algorithm to identify MIBC patients who might benefit from immunotherapy or the de-escalation of therapy. (2) Methods: We constructed an artificial neural network classifier for tumor cells, lymphocytes, stromal cells, and "ignore" cells from hematoxylin-and-eosin-stained slide images using the QuPath open source software. We defined four unique TILs variables based on ML to analyze TILs measurements. Pathological slide images from 133 MIBC patients were retrospectively collected as the discovery set to determine the optimal association of ML-read TILs variables with patient survival outcomes. For validation, we evaluated an independent external validation set consisting of 247 MIBC patients. (3) Results: We found that all four TILs variables had significant prognostic associations with survival outcomes in MIBC patients (p < 0.001 for all comparisons), with higher TILs score being associated with better prognosis. Univariate and multivariate Cox regression analyses demonstrated that electronic TILs (eTILs) variables were independently associated with overall survival after adjustment for clinicopathological factors including age, sex, and pathological stage (p < 0.001 for all analyses). Results analyzed in different subgroups showed that the eTILs variable was a strong prognostic factor that was not redundant with pre-existing clinicopathological features (p < 0.05 for all analyses). (4) Conclusion: ML-driven cell classifier-defined TILs variables were robust and independent prognostic factors in two independent cohorts of MIBC patients. eTILs have the potential to identify a subset of high-risk stage II or stage III-IV MIBC patients who might benefit from adjuvant immunotherapy.

Accurate Diagnosis and Survival Prediction of Bladder Cancer Using Deep Learning on Histological Slides.

(1) Background: Early diagnosis and treatment are essential to reduce the mortality rate of bladder cancer (BLCA). We aimed to develop deep learning (DL)-based weakly supervised models for the diagnosis of BLCA and prediction of overall survival (OS) in muscle-invasive bladder cancer (MIBC) patients using whole slide digitized histological images (WSIs). (2) Methods: Diagnostic and prognostic models were developed using 926 WSIs of 412 BLCA patients from The Cancer Genome Atlas cohort. We collected 250 WSIs of 150 BLCA patients from the Renmin Hospital of Wuhan University cohort for external validation of the models. Two DL models were developed: a BLCA diagnostic model (named BlcaMIL) and an MIBC prognostic model (named MibcMLP). (3) Results: The BlcaMIL model identified BLCA with accuracy 0.987 in the external validation set, comparable to that of expert uropathologists and outperforming a junior pathologist. The C-index values for the MibcMLP model on the internal and external validation sets were 0.631 and 0.622, respectively. The risk score predicted by MibcMLP was a strong predictor independent of existing clinical or histopathologic indicators, as demonstrated by univariate Cox (HR = 2.390, p < 0.0001) and multivariate Cox (HR = 2.414, p < 0.0001) analyses. The interpretability of DL models can help in the analysis of critical regions associated with tumors to enrich the information obtained from WSIs. Furthermore, the expression of six genes (ANAPC7, MAPKAPK5, COX19, LINC01106, AL161431.1 and MYO16-AS1) was significantly associated with MibcMLP-predicted risk scores, revealing possible potential biological correlations. (4) Conclusions: Our study developed DL models for accurately diagnosing BLCA and predicting OS in MIBC patients, which will help promote the precise pathological diagnosis of BLCA and risk stratification of MIBC to improve clinical treatment decisions.

Discovery of JNJ-64264681: A Potent and Selective Covalent Inhibitor of Bruton's Tyrosine Kinase.

Bruton's tyrosine kinase (BTK) is a Tec family kinase that plays an essential role in B-cell receptor (BCR) signaling as well as Fcγ receptor signaling in leukocytes. Pharmacological inhibition of BTK has been shown to be effective in treating hematological malignancies and is hypothesized to provide an effective strategy for the treatment of autoimmune diseases such as rheumatoid arthritis and systemic lupus erythematosus. We report the discovery and preclinical properties of JNJ-64264681 (13), a covalent, irreversible BTK inhibitor with potent whole blood activity and exceptional kinome selectivity. JNJ-64264681 demonstrated excellent oral efficacy in both cancer and autoimmune models with sustained in vivo target coverage amenable to once daily dosing and has advanced into human clinical studies to investigate safety and pharmacokinetics.