Lactate Levels as a Predictor of Emergency Department Revisits in Infants With Acute Bronchiolitis.

OBJECTIVE: This study aimed to identify predictive biomarkers for unscheduled emergency department (ED) revisits within 24 hours of discharge in infants diagnosed with acute bronchiolitis (AB). METHODS: A retrospective observational study was conducted on infants diagnosed with AB who visited 3 emergency medical centers between January 2020 and December 2022. The study excluded infants with comorbidities, congenital diseases, and prematurity and infants who revisited the ED after 24 hours of discharge. Demographic data, vital signs, and laboratory results were collected from the medical records. Univariable and multivariable logistic regression analyses were performed on factors with P of less than 0.1 in univariable analysis. Receiver operator curve analysis was used to assess the accuracy of lactate measurements in predicting ED revisits within 24 hours of discharge. RESULTS: Out of 172 participants, 100 were in the revisit group and 72 in the discharge group. The revisit group was significantly younger and exhibited higher lactate levels, lower pH values, and higher pCO 2 levels compared to the discharge group. Univariable logistic regression identified several factors associated with revisits. Multivariable analysis found that only lactate was a variable correlated with predicting ED revisits (odds ratio, 18.020; 95% confidence interval [CI], 5.764-56.334). The receiver operator curve analysis showed an area under the curve of 0.856, with an optimal lactate cutoff value of 2.15. CONCLUSION: Lactate value in infants diagnosed with AB were identified as a potential indicator of predicting unscheduled ED revisits within 24 hours of discharge. The predictive potential of lactate levels holds promise for enhancing prognosis prediction, reducing health care costs, and alleviating ED overcrowding. However, given the study's limitations, a more comprehensive prospective investigation is recommended to validate these findings.

Design of Wideband Flextensional Hydrophone.

Flextensional transducers have been widely used as low-frequency projectors, and these characteristics can be used to develop hydrophones with wider receiver bandwidth and higher sensitivity than conventional products in low-frequency ranges. In this work, we designed flextensional hydrophones of all classes, and compared their acoustic receiver performance to select the most suitable class for a low-frequency broadband hydrophone. For this purpose, basic models of the hydrophones were constructed for all classes and the effects of various structural parameters on the acoustic receiver characteristics of the hydrophones were analyzed. Based on the results, the structure of the flextensional hydrophone of each class was designed to have the maximum receiver bandwidth by an optimization technique while maintaining the receiver voltage sensitivity over a certain level. A comparison of the designed performance led to the selection of the class IV flextensional hydrophone as the most promising one with the widest receiver fractional bandwidth and highest sensitivity.

Integrated metagenomics and metabolomics analysis illustrates the systemic impact of the gut microbiota on host metabolism after bariatric surgery.

AIM: To explore how bariatric surgery (BS) modified the obesity-associated gut microbiome, the host metabolome, and their interactions in obese Korean patients. MATERIALS AND METHODS: Stool and fasting blood samples were obtained before and 1, 3, 6, and 12 months after BS from 52 patients enrolled in the Korean Obesity Surgical Treatment Study. We analysed the gut microbiome by 16S rRNA gene sequencing and the serum metabolome, including bile acids, by nuclear magnetic resonance spectroscopy and ultrahigh-performance liquid chromatography/triple quadrupole mass spectrometry. RESULTS: Stool metagenomics showed that 27 microbiota were enriched and 14 microbiota were reduced after BS, whereas the abundances and diversity of observed features were increased. The levels of branched-chain amino acids and metabolites of energy metabolism in serum were decreased after surgery, whereas the levels of metabolites related to microbial metabolism, including dimethyl sulphone, glycine, and secondary bile acids, were increased in the serum samples. In addition, we found notable mutual associations among metabolites and gut microbiome changes attributed to BS. CONCLUSIONS: Changes in the gut microbiome community and systemic levels of amino acids and sugars were directly derived from anatomical changes in the gastrointestinal tract after BS. We hypothesized that the observed increases in microbiome-related serum metabolites were a result of complex and indirect changes derived from BS. Ethnic-specific environmental or genetic factors could affect Korean-specific postmetabolic modification in obese patients who undergo BS.

Deep Learning with Multimodal Integration for Predicting Recurrence in Patients with Non-Small Cell Lung Cancer.

Due to high recurrence rates in patients with non-small cell lung cancer (NSCLC), medical professionals need extremely accurate diagnostic methods to prevent bleak prognoses. However, even the most commonly used diagnostic method, the TNM staging system, which describes the tumor-size, nodal-involvement, and presence of metastasis, is often inaccurate in predicting NSCLC recurrence. These limitations make it difficult for clinicians to tailor treatments to individual patients. Here, we propose a novel approach, which applies deep learning to an ensemble-based method that exploits patient-derived, multi-modal data. This will aid clinicians in successfully identifying patients at high risk of recurrence and improve treatment planning.

Copper-Catalyzed Aza-Michael Addition of Aromatic Amines or Aromatic Aza-Heterocycles to α,ß-Unsaturated Olefins.

A highly efficient and mild Cu-catalyzed conjugate addition reaction of aromatic amines and aromatic aza-heterocycles to α,ß-unsaturated olefins is described. The transformation is promoted by 3-7 mol % of a Cu complex generated in situ from a mixture of inexpensive CuCl, a readily available phosphine or imidazolium salt, and KOt-Bu at ambient temperature. A wide range of ß-amino sulfone, ß-amino nitrile, and ß-amino carbonyl compounds is efficiently and selectively synthesized in high yields (62-99%).

4-Isopropyl-2,6-bis(1-phenylethyl)aniline 1, an Analogue of KTH-13 Isolated from Cordyceps bassiana, Inhibits the NF-κB-Mediated Inflammatory Response.

The Cordyceps species has been a good source of compounds with anticancer and anti-inflammatory activities. Recently, we reported a novel compound (4-isopropyl-2,6-bis(1-phenylethyl)phenol, KTH-13) with anticancer activity isolated from Cordyceps bassiana and created several derivatives to increase its pharmacological activity. In this study, we tested one of the KTH-013 derivatives, 4-isopropyl-2,6-bis(1-phenylethyl)aniline 1 (KTH-13-AD1), with regard to anti-inflammatory activity under macrophage-mediated inflammatory conditions. KTH-13-AD1 clearly suppressed the production of nitric oxide (NO) and reactive oxygen species (ROS) in lipopolysaccharide (LPS) and sodium nitroprusside- (SNP-) treated macrophage-like cells (RAW264.7 cells). Similarly, this compound also reduced mRNA expression of inducible NO synthase (iNOS) and tumor necrosis factor-α (TNF-α), as analyzed by RT-PCR and real-time PCR. Interestingly, KTH-13-AD1 strongly diminished NF-κB-mediated luciferase activities and nuclear translocation of NF-κB family proteins. In accordance, KTH-13-AD1 suppressed the upstream signaling pathway of NF-κB activation, including IκBα, IKKα/ß, AKT, p85/PI3K, and Src in a time- and dose-dependent manner. The autophosphorylation of Src and NF-κB observed during the overexpression of Src was also suppressed by KTH-13-AD1. These results strongly suggest that KTH-13-AD1 has strong anti-inflammatory features mediated by suppression of the Src/NF-κB regulatory loop.

The neuropharmacological properties of α-pyrrolidinobutiothiophenone, a new synthetic cathinone, in rodents; role of the dopaminergic system.

BACKGROUND AND PURPOSE: α-Pyrrolidinobutiothiophenone (α-PBT) is a chemical derivative of cathinone, a structural analogue of amphetamine. Until now, there have been a few previous neurochemical or neurobehavioural studies on the abuse potential of α-PBT. EXPERIMENTAL APPROACH: We examined the abuse potential of α-PBT by measuring psychomotor, rewarding, and reinforcing properties and methamphetamine-like discriminative stimulus effects in rodents using locomotor activity, conditioned place preference, self-administration, and drug discrimination studies. To clarify the underlying neuropharmacological mechanisms, we measured dopamine levels and neuronal activation in the dorsal striatum. In addition, we investigated the role of the dopamine D1 receptor or D2 receptors in α-PBT-induced hyperlocomotor activity, conditioned place preference, and the methamphetamine-like discriminative stimulus effect of α-PBT in rodents. KEY RESULTS: α-PBT promoted hyperlocomotor activity in mice. α-PBT induced drug-paired place preference in mice and supported self-administration in rats. In a drug discrimination experiment, α-PBT fully substituted for the discriminative stimulus effects of methamphetamine in rats. Furthermore, α-PBT increased dopamine levels and c-Fos expression in the dorsal striatum of mice, which was associated with these behaviours. Finally, pretreatment with the D1 receptor antagonist SCH23390 or the D2 receptors antagonist eticlopride significantly attenuated acute or repeated α-PBT-induced hyperlocomotor activity, place preference, and the methamphetamine-like discriminative stimulus effects in rodents. CONCLUSIONS AND IMPLICATIONS: These findings suggest that α-PBT has abuse potential at the highest dose tested via enhanced dopaminergic transmission in the dorsal striatum of rodents. The results provide scientific evidence for the legal restrictions of the recreational use of α-PBT.

Prediction of lung cancer using novel biomarkers based on microbiome profiling of bronchoalveolar lavage fluid.

There is an unmet need for biomarkers for the diagnosis of lung cancer and decision criteria for lung biopsy. We comparatively investigated the lung microbiomes of patients with lung cancer and benign lung diseases. Patients who underwent bronchoscopy at Chungnam National University Hospital between June 2021 and June 2022 were enrolled. Bronchoalveolar lavage fluid (BALF) was collected from 24 patients each with lung cancer and benign lung diseases. The samples were analyzed using 16S rRNA-based metagenomic sequencing. We found that alpha diversity and the beta diversity distribution (P = 0.001) differed significantly between patients with benign lung diseases and those with lung cancer. Firmicutes was the most abundant phylum in patients with lung cancer (33.39% ± 17.439), whereas Bacteroidota was the most abundant phylum in patients with benign lung disease (31.132% ± 22.505), respectively. In differential abundance analysis, the most differentially abundant microbiota taxon was unclassified_SAR202_clade, belonging to the phylum Chloroflexi. The established prediction model distinguished patients with benign lung disease from those with lung cancer with a high accuracy (micro area under the curve [AUC] = 0.98 and macro AUC = 0.99). The BALF microbiome may be a novel biomarker for the detection of lung cancer.

Fecal microbiota transplantation improves anti-PD-1 inhibitor efficacy in unresectable or metastatic solid cancers refractory to anti-PD-1 inhibitor.

The gut microbiome significantly influences immune responses and the efficacy of immune checkpoint inhibitors. We conducted a clinical trial (NCT04264975) combining an anti-programmed death-1 (PD-1) inhibitor with fecal microbiota transplantation (FMT) from anti-PD-1 responder in 13 patients with anti-PD-1-refractory advanced solid cancers. FMT induced sustained microbiota changes and clinical benefits in 6 of 13 patients, with 1 partial response and 5 stable diseases, achieving an objective response rate of 7.7% and a disease control rate of 46.2%. The clinical response correlates with increased cytotoxic T cells and immune cytokines in blood and tumors. We isolated Prevotella merdae Immunoactis from a responder to FMT, which stimulates T cell activity and suppresses tumor growth in mice by enhancing cytotoxic T cell infiltration. Additionally, we found Lactobacillus salivarius and Bacteroides plebeius may inhibit anti-tumor immunity. Our findings suggest that FMT with beneficial microbiota can overcome resistance to anti-PD-1 inhibitors in advanced solid cancers, especially gastrointestinal cancers.

Advancing diabetes prediction with a progressive self-transfer learning framework for discrete time series data.

Although diabetes mellitus is a complex and pervasive disease, most studies to date have focused on individual features, rather than considering the complexities of multivariate, multi-instance, and time-series data. In this study, we developed a novel diabetes prediction model that incorporates these complex data types. We applied advanced techniques of data imputation (bidirectional recurrent imputation for time series; BRITS) and feature selection (the least absolute shrinkage and selection operator; LASSO). Additionally, we utilized self-supervised algorithms and transfer learning to address the common issues with medical datasets, such as irregular data collection and sparsity. We also proposed a novel approach for discrete time-series data preprocessing, utilizing both shifting and rolling time windows and modifying time resolution. Our study evaluated the performance of a progressive self-transfer network for predicting diabetes, which demonstrated a significant improvement in metrics compared to non-progressive and single self-transfer prediction tasks, particularly in AUC, recall, and F1 score. These findings suggest that the proposed approach can mitigate accumulated errors and reflect temporal information, making it an effective tool for accurate diagnosis and disease management. In summary, our study highlights the importance of considering the complexities of multivariate, multi-instance, and time-series data in diabetes prediction.

Intra-person multi-task learning method for chronic-disease prediction.

In the medical field, various clinical information has been accumulated to help clinicians provide personalized medicine and make better diagnoses. As chronic diseases share similar characteristics, it is possible to predict multiple chronic diseases using the accumulated data of each patient. Thus, we propose an intra-person multi-task learning framework that jointly predicts the status of correlated chronic diseases and improves the model performance. Because chronic diseases occur over a long period and are affected by various factors, we considered features related to each chronic disease and the temporal relationship of the time-series data for accurate prediction. The study was carried out in three stages: (1) data preprocessing and feature selection using bidirectional recurrent imputation for time series (BRITS) and the least absolute shrinkage and selection operator (LASSO); (2) a convolutional neural network and long short-term memory (CNN-LSTM) for single-task models; and (3) a novel intra-person multi-task learning CNN-LSTM framework developed to predict multiple chronic diseases simultaneously. Our multi-task learning method between correlated chronic diseases produced a more stable and accurate system than single-task models and other baseline recurrent networks. Furthermore, the proposed model was tested using different time steps to illustrate its flexibility and generalization across multiple time steps.

A multi-kernel and multi-scale learning based deep ensemble model for predicting recurrence of non-small cell lung cancer.

Predicting recurrence in patients with non-small cell lung cancer (NSCLC) before treatment is vital for guiding personalized medicine. Deep learning techniques have revolutionized the application of cancer informatics, including lung cancer time-to-event prediction. Most existing convolutional neural network (CNN) models are based on a single two-dimensional (2D) computational tomography (CT) image or three-dimensional (3D) CT volume. However, studies have shown that using multi-scale input and fusing multiple networks provide promising performance. This study proposes a deep learning-based ensemble network for recurrence prediction using a dataset of 530 patients with NSCLC. This network assembles 2D CNN models of various input slices, scales, and convolutional kernels, using a deep learning-based feature fusion model as an ensemble strategy. The proposed framework is uniquely designed to benefit from (i) multiple 2D in-plane slices to provide more information than a single central slice, (ii) multi-scale networks and multi-kernel networks to capture the local and peritumoral features, (iii) ensemble design to integrate features from various inputs and model architectures for final prediction. The ensemble of five 2D-CNN models, three slices, and two multi-kernel networks, using 5 × 5 and 6 × 6 convolutional kernels, achieved the best performance with an accuracy of 69.62%, area under the curve (AUC) of 72.5%, F1 score of 70.12%, and recall of 70.81%. Furthermore, the proposed method achieved competitive results compared with the 2D and 3D-CNN models for cancer outcome prediction in the benchmark studies. Our model is also a potential adjuvant treatment tool for identifying NSCLC patients with a high risk of recurrence.

Intracellular Adhesion Molecule-1 Improves Responsiveness to Immune Checkpoint Inhibitor by Activating CD8+ T Cells.

Immune checkpoint inhibitor (ICI) clinically benefits cancer treatment. However, the ICI responses are only achieved in a subset of patients, and the underlying mechanisms of the limited response remain unclear. 160 patients with non-small cell lung cancer treated with anti-programmed cell death protein-1 (anti-PD-1) or anti-programmed death ligand-1 (anti-PD-L1) are analyzed to understand the early determinants of response to ICI. It is observed that high levels of intracellular adhesion molecule-1 (ICAM-1) in tumors and plasma of patients are associated with prolonged survival. Further reverse translational studies using murine syngeneic tumor models reveal that soluble ICAM-1 (sICAM-1) is a key molecule that increases the efficacy of anti-PD-1 via activation of cytotoxic T cells. Moreover, chemokine (CXC motif) ligand 13 (CXCL13) in tumors and plasma is correlated with the level of ICAM-1 and ICI efficacy, suggesting that CXCL13 might be involved in the ICAM-1-mediated anti-tumor pathway. Using sICAM-1 alone and in combination with anti-PD-1 enhances anti-tumor efficacy in anti-PD-1-responsive tumors in murine models. Notably, combinatorial therapy with sICAM-1 and anti-PD-1 converts anti-PD-1-resistant tumors to responsive ones in a preclinical study. These findings provide a new immunotherapeutic strategy for treating cancers using ICAM-1.

Accuracy and reliability of length measurements on three-dimensional computed tomography using open-source OsiriX software.

There is a growing interest in three-dimensional computed tomography (3D-CT) as a research tool for the study of bone, joint anatomy, and kinematics. However, when CT data are processed and handled manually using image processing programs to yield 3D image and coordinate value, systematic and random errors should be validated. We evaluated the accuracy and reliability of length measurement on CT with OsiriX software. 3D-CT scans were made of 14 frozen pig knees with five transosseous holes in the metaphyseal portion of femur. The lengths between tunnel orifices were measured using Mitutoyo Digimatic digital calipers to establish the gold standard, and with the OsiriX program in 3D multi-planar reformatting mode for comparison. All measurements were recorded by a principal (replicate 1, trial 1) and a secondary observer (replicate 2, trial 1) and were repeated once by each observer (trial 2). The mean differences between OsiriX and real measurements were less than 0.1 mm in both replicates, and maximum differences were less than 0.3 mm. There were no significant differences between the replicates and real measurements (p=0.544 and 0.622 for replicates 1 and 2, respectively). The intraclass correlation coefficients (ICC) were very high between trials and between replicates (ICC=0.998 and 0.999, respectively). For kinematic analysis of the knees, length measurements on 3D-CT using OsiriX program can be used as alternatives to real measurements with less than 0.3-mm accuracy and very high reliability.

Staphylococcus epidermidis WF2R11 Suppresses PM2.5-Mediated Activation of the Aryl Hydrocarbon Receptor in HaCaT Keratinocytes.

The skin supports a diverse microbiome whose imbalance is related to skin inflammation and diseases. Exposure to fine particulate matter (PM2.5), a major air pollutant, can adversely affect the skin microbiota equilibrium. In this study, the effect and mechanism of PM2.5 exposure in HaCaT keratinocytes were investigated. PM2.5 stimulated the aryl hydrocarbon receptor (AhR) to produce reactive oxygen species (ROS) in HaCaT cells, leading to mitochondrial dysfunction and intrinsic mitochondrial apoptosis. We observed that the culture medium derived from a particular skin microbe, Staphylococcus epidermidis WF2R11, remarkably reduced oxidative stress in HaCaT cells caused by PM2.5-mediated activation of the AhR pathway. Staphylococcus epidermidis WF2R11 also exhibited inhibition of ROS-induced inflammatory cytokine secretion. Herein, we demonstrated that S. epidermidis WF2R11 could act as a suppressor of AhRs, affect cell proliferation, and inhibit apoptosis. Our results highlight the importance of the clinical application of skin microbiome interventions in the treatment of inflammatory skin diseases.

Whole-Transcriptome Sequencing Reveals Characteristics of Cancer Microbiome in Korean Patients with GI Tract Cancer: Fusobacterium nucleatum as a Therapeutic Target.

Remarkable progress has occurred over the past two decades in identifying microbiomes affecting the human body in numerous ways. The microbiome is linked to gastrointestinal (GI) tract cancer. The purpose of this study was to determine if there is a common microbiome among GI tract cancers and how the microbiome affects the disease. To ensure ethnic consistency, Korean patients with GI tract cancer were selected. Fusobacterium nucleatum is an enriched bacteria in all cancer tissues. F. nucleatum is a Gram-negative obligate anaerobe that promotes colorectal cancer. Through Gene Set Enrichment Analysis (GSEA) and Differentially Expressed Genes (DEG) analyses, the upregulation of the G2M checkpoint pathway was identified in the F. nucleatum-high group. Cell viability and G2M checkpoint pathway genes were examined in MC 38 cells treated with F. nucleatum. F. nucleatum upregulated the expression of G2M checkpoint pathway genes and the cell proliferation of MC 38 cells. F. nucleatum facilitated cancer's use of G2M checkpoint pathways and F. nucleatum could be a therapeutic target in Korean GI tract cancer.

Gut microbiome alterations in preclinical Alzheimer's disease.

BACKGROUND: Although some human studies have reported gut microbiome changes in individuals with Alzheimer's disease (AD) dementia or mild cognitive impairment (MCI), gut microbiome alterations in preclinical AD, i.e., cerebral amyloidosis without cognitive impairment, is largely unknown. OBJECTIVE: We aimed to identify gut microbial alterations associated with preclinical AD by comparing cognitively normal (CN) older adults with cerebral Aß deposition (Aß+ CN) and those without cerebral Aß deposition (Aß- CN). METHODS: Seventy-eight CN older participants (18 Aß+ CN and 60 Aß- CN) were included, and all participants underwent clinical assessment and Pittsburg compound B-positron emission tomography. The V3-V4 region of the 16S rRNA gene of genomic DNA extracted from feces was amplified and sequenced to establish the microbial community. RESULTS: Generalized linear model analysis revealed that the genera Megamonas (B = 3.399, q<0.001), Serratia (B = 3.044, q = 0.005), Leptotrichia (B = 5.862, q = 0.024) and Clostridium (family Clostridiaceae) (B = 0.788, q = 0.034) were more abundant in the Aß+ CN group than the Aß- CN group. In contrast, genera CF231 (B = -3.237, q< 0.001), Victivallis (B = -3.447, q = 0.004) Enterococcus (B = -2.044, q = 0.042), Mitsuokella (B = -2.119, q = 0.042) and Clostridium (family Erysipelotrichaceae) (B = -2.222, q = 0.043) were decreased in Aß+ CN compared to Aß- CN. Notably, the classification model including the differently abundant genera could effectively distinguish Aß+ CN from Aß- CN (AUC = 0.823). CONCLUSION: Our findings suggest that specific alterations of gut bacterial taxa are related to preclinical AD, which means these changes may precede cognitive decline. Therefore, examining changes in the microbiome may be helpful in preclinical AD screening.

Bifidobacterial carbohydrate/nucleoside metabolism enhances oxidative phosphorylation in white adipose tissue to protect against diet-induced obesity.

BACKGROUND: Comparisons of the gut microbiome of lean and obese humans have revealed that obesity is associated with the gut microbiome plus changes in numerous environmental factors, including high-fat diet (HFD). Here, we report that two species of Bifidobacterium are crucial to controlling metabolic parameters in the Korean population. RESULTS: Based on gut microbial analysis from 99 Korean individuals, we observed the abundance of Bifidobacterium longum and Bifidobacterium bifidum was markedly reduced in individuals with increased visceral adipose tissue (VAT), body mass index (BMI), blood triglyceride (TG), and fatty liver. Bacterial transcriptomic analysis revealed that carbohydrate/nucleoside metabolic processes of Bifidobacterium longum and Bifidobacterium bifidum were associated with protecting against diet-induced obesity. Oral treatment of specific commercial Bifidobacterium longum and Bifidobacterium bifidum enhanced bile acid signaling contributing to potentiate oxidative phosphorylation (OXPHOS) in adipose tissues, leading to reduction of body weight gain and improvement in hepatic steatosis and glucose homeostasis. Bifidobacterium longum or Bifidobacterium bifidum manipulated intestinal sterol biosynthetic processes to protect against diet-induced obesity in germ-free mice. CONCLUSIONS: Our findings support the notion that treatment of carbohydrate/nucleoside metabolic processes-enriched Bifidobacterium longum and Bifidobacterium bifidum would be a novel therapeutic strategy for reprograming the host metabolic homeostasis to protect against metabolic syndromes, including diet-induced obesity. Video Abstract.

Bifidobacterium Strain-Specific Enhances the Efficacy of Cancer Therapeutics in Tumor-Bearing Mice.

Colorectal cancer (CRC) is among the leading causes of cancer-related death in the world. The development of CRC is associated with smoking, diet, and microbial exposure. Previous studies have shown that dysbiosis of the gut microbiome affects cancer development, because it leads to inflammation and genotoxicity. Supplementation with specific microbiota induces anti-tumor effects by enhancing of anti-tumor immunity. Here, we observed that supplementation with either of two B. breve strains reduces tumor growth in MC38 colon carcinoma-bearing mice. Interestingly, only one B. breve strain boosted the efficacy of cancer therapeutics, including oxaliplatin and PD-1 blockade. Extensive immune profiling and transcriptomic analysis revealed that the boosting B. breve strain augments lymphocyte-mediated anti-cancer immunity. Our results suggest that supplementation with B. breve strains could potentially be used as a strategy to enhance the efficacy of CRC therapeutics.

Practical prediction model of the clinical response to programmed death-ligand 1 inhibitors in advanced gastric cancer.

The identification of predictive biomarkers or models is necessary for the selection of patients who might benefit the most from immunotherapy. Seven histological features (signet ring cell [SRC], fibrous stroma, myxoid stroma, tumor-infiltrating lymphocytes [TILs], necrosis, tertiary lymphoid follicles, and ulceration) detected in surgically resected tissues (N = 44) were used to train a model. The presence of SRC became an optimal decision parameter for pathology alone (AUC = 0.78). Analysis of differentially expressed genes (DEGs) for the prediction of genomic markers showed that C-X-C motif chemokine ligand 11 (CXCL11) was high in responders (P < 0.001). Immunohistochemistry (IHC) was performed to verify its potential as a biomarker. IHC revealed that the expression of CXCL11 was associated with responsiveness (P = 0.003). The response prediction model was trained by integrating the results of the analysis of pathological factors and RNA sequencing (RNA-seq). When trained with the C5.0 decision tree model, the categorical level of the expression of CXCL11, a single variable, was shown to be the best model (AUC = 0.812). The AUC of the model trained with the random forest was 0.944. Survival analysis revealed that the C5.0-trained model (log-rank P = 0.01 for progression-free survival [PFS]; log-rank P = 0.012 for overall survival [OS]) and the random forest-trained model (log-rank P < 0.001 for PFS; log-rank P = 0.001 for OS) predicted prognosis more accurately than the PD-L1 test (log-rank P = 0.031 for PFS; log-rank P = 0.107 for OS).