Prediction of immunochemotherapy response for diffuse large B-cell lymphoma using artificial intelligence digital pathology.

Diffuse large B-cell lymphoma (DLBCL) is a heterogeneous and prevalent subtype of aggressive non-Hodgkin lymphoma that poses diagnostic and prognostic challenges, particularly in predicting drug responsiveness. In this study, we used digital pathology and deep learning to predict responses to immunochemotherapy in patients with DLBCL. We retrospectively collected 251 slide images from 216 DLBCL patients treated with rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP), with their immunochemotherapy response labels. The digital pathology images were processed using contrastive learning for feature extraction. A multi-modal prediction model was developed by integrating clinical data and pathology image features. Knowledge distillation was employed to mitigate overfitting on gigapixel histopathology images to create a model that predicts responses based solely on pathology images. Based on the importance derived from the attention mechanism of the model, we extracted histological features that were considered key textures associated with drug responsiveness. The multi-modal prediction model achieved an impressive area under the ROC curve of 0.856, demonstrating significant associations with clinical variables such as Ann Arbor stage, International Prognostic Index, and bulky disease. Survival analyses indicated their effectiveness in predicting relapse-free survival. External validation using TCGA datasets supported the model's ability to predict survival differences. Additionally, pathology-based predictions show promise as independent prognostic indicators. Histopathological analysis identified centroblastic and immunoblastic features to be associated with treatment response, aligning with previous morphological classifications and highlighting the objectivity and reproducibility of artificial intelligence-based diagnosis. This study introduces a novel approach that combines digital pathology and clinical data to predict the response to immunochemotherapy in patients with DLBCL. This model shows great promise as a diagnostic and prognostic tool for clinical management of DLBCL. Further research and genomic data integration hold the potential to enhance its impact on clinical practice, ultimately improving patient outcomes.

Ensemble Deep Learning Model to Predict Lymphovascular Invasion in Gastric Cancer.

Lymphovascular invasion (LVI) is one of the most important prognostic factors in gastric cancer as it indicates a higher likelihood of lymph node metastasis and poorer overall outcome for the patient. Despite its importance, the detection of LVI(+) in histopathology specimens of gastric cancer can be a challenging task for pathologists as invasion can be subtle and difficult to discern. Herein, we propose a deep learning-based LVI(+) detection method using H&E-stained whole-slide images. The ConViT model showed the best performance in terms of both AUROC and AURPC among the classification models (AUROC: 0.9796; AUPRC: 0.9648). The AUROC and AUPRC of YOLOX computed based on the augmented patch-level confidence score were slightly lower (AUROC: -0.0094; AUPRC: -0.0225) than those of the ConViT classification model. With weighted averaging of the patch-level confidence scores, the ensemble model exhibited the best AUROC, AUPRC, and F1 scores of 0.9880, 0.9769, and 0.9280, respectively. The proposed model is expected to contribute to precision medicine by potentially saving examination-related time and labor and reducing disagreements among pathologists.

Low gamma-butyrobetaine dioxygenase (BBOX1) expression as a prognostic biomarker in patients with clear cell renal cell carcinoma: a machine learning approach.

Gamma-butyrobetaine dioxygenase (BBOX1) is a catalyst for the conversion of gamma-butyrobetaine to l-carnitine, which is detected in normal renal tubules. The purpose of this study was to analyze the prognosis, immune response, and genetic alterations associated with low BBOX1 expression in patients with clear cell renal cell carcinoma (RCC). We analyzed the relative influence of BBOX1 on survival using machine learning and investigated drugs that can inhibit renal cancer cells with low BBOX1 expression. We analyzed clinicopathologic factors, survival rates, immune profiles, and gene sets according to BBOX1 expression in a total of 857 patients with kidney cancer from the Hanyang University Hospital cohort (247 cases) and The Cancer Genome Atlas (610 cases). We employed immunohistochemical staining, gene set enrichment analysis, in silico cytometry, pathway network analyses, in vitro drug screening, and gradient boosting machines. BBOX1 expression in RCC was decreased compared with that in normal tissues. Low BBOX1 expression was associated with poor prognosis, decreased CD8+ T cells, and increased neutrophils. In gene set enrichment analyses, low BBOX1 expression was related to gene sets with oncogenic activity and a weak immune response. In pathway network analysis, BBOX1 was linked to regulation of various T cells and programmed death-ligand 1. In vitro drug screening showed that midostaurin, BAY-61-3606, GSK690693, and linifanib inhibited the growth of RCC cells with low BBOX1 expression. Low BBOX1 expression in patients with RCC is related to short survival time and reduced CD8+ T cells; midostaurin, among other drugs, may have enhanced therapeutic effects in this context.

Deep learning model to predict Epstein-Barr virus associated gastric cancer in histology.

The detection of Epstein-Barr virus (EBV) in gastric cancer patients is crucial for clinical decision making, as it is related with specific treatment responses and prognoses. Despite its importance, the limited medical resources preclude universal EBV testing. Herein, we propose a deep learning-based EBV prediction method from H&E-stained whole-slide images (WSI). Our model was developed using 319 H&E stained WSI (26 EBV positive; TCGA dataset) from the Cancer Genome Atlas, and 108 WSI (8 EBV positive; ISH dataset) from an independent institution. Our deep learning model, EBVNet consists of two sequential components: a tumor classifier and an EBV classifier. We visualized the learned representation by the classifiers using UMAP. We externally validated the model using 60 additional WSI (7 being EBV positive; HGH dataset). We compared the model's performance with those of four pathologists. EBVNet achieved an AUPRC of 0.65, whereas the four pathologists yielded a mean AUPRC of 0.41. Moreover, EBVNet achieved an negative predictive value, sensitivity, specificity, precision, and F1-score of 0.98, 0.86, 0.92, 0.60, and 0.71, respectively. Our proposed model is expected to contribute to prescreen patients for confirmatory testing, potentially to save test-related cost and labor.

How Many Private Data Are Needed for Deep Learning in Lung Nodule Detection on CT Scans? A Retrospective Multicenter Study.

Early detection of lung nodules is essential for preventing lung cancer. However, the number of radiologists who can diagnose lung nodules is limited, and considerable effort and time are required. To address this problem, researchers are investigating the automation of deep-learning-based lung nodule detection. However, deep learning requires large amounts of data, which can be difficult to collect. Therefore, data collection should be optimized to facilitate experiments at the beginning of lung nodule detection studies. We collected chest computed tomography scans from 515 patients with lung nodules from three hospitals and high-quality lung nodule annotations reviewed by radiologists. We conducted several experiments using the collected datasets and publicly available data from LUNA16. The object detection model, YOLOX was used in the lung nodule detection experiment. Similar or better performance was obtained when training the model with the collected data rather than LUNA16 with large amounts of data. We also show that weight transfer learning from pre-trained open data is very useful when it is difficult to collect large amounts of data. Good performance can otherwise be expected when reaching more than 100 patients. This study offers valuable insights for guiding data collection in lung nodules studies in the future.

Machine Learning Models for Predicting Mortality in 7472 Very Low Birth Weight Infants Using Data from a Nationwide Neonatal Network.

Statistical and analytical methods using artificial intelligence approaches such as machine learning (ML) are increasingly being applied to the field of pediatrics, particularly to neonatology. This study compared the representative ML analysis and the logistic regression (LR), which is a traditional statistical analysis method, using them to predict mortality of very low birth weight infants (VLBWI). We included 7472 VLBWI data from a nationwide Korean neonatal network. Eleven predictor variables (neonatal factors: male sex, gestational age, 5 min Apgar scores, body temperature, and resuscitation at birth; maternal factors: diabetes mellitus, hypertension, chorioamnionitis, premature rupture of membranes, antenatal steroid, and cesarean delivery) were selected based on clinical impact and statistical analysis. We compared the predicted mortality between ML methods­such as artificial neural network (ANN), random forest (RF), and support vector machine (SVM)­and LR with a randomly selected training set (80%) and a test set (20%). The model performances of area under the receiver operating curve (95% confidence interval) equaled LR 0.841 (0.811−0.872), ANN 0.845 (0.815−0.875), and RF 0.826 (0.795−0.858). The exception was SVM 0.631 (0.578−0.683). No statistically significant differences were observed between the performance of LR, ANN, and RF (i.e., p > 0.05). However, the SVM model was lower (p < 0.01). We suggest that VLBWI mortality prediction using ML methods would yield the same prediction rate as the traditional statistical LR method and may be suitable for predicting mortality. However, low prediction rates are observed in certain ML methods; hence, further research is needed on these limitations and selecting an appropriate method.

High SLC2A1 expression associated with suppressing CD8 T cells and B cells promoted cancer survival in gastric cancer.

High expression of glucose transporter family members, which augment glucose uptake and glycolytic flux, has been shown to play a pivotal role in the proliferation and survival of tumor cells, contributing to the energy supply, biosynthesis and homeostasis of cancer cells. Among the many members, solute carrier family 2 member 1 (SLC2A1) encodes a glucose transporter, GLUT1, that is critical in the metabolism of glucose, which is an energy source for cell growth that contributes to cancer progression and development. The aim of this study was to analyze the survival and genetic changes/immune profiles in patients with gastric cancer with high SLC2A1 expression and to provide treatment for improving prognosis. This study investigated the clinicopathologic parameters, the proportion of immune cells and gene sets affecting SLC2A1 expression in 279 and 415 patients with gastric cancer from the Eulji Hospital cohort and The Cancer Genome Atlas, respectively. We assessed the response to conventional chemotherapy drugs, including fluorouracil, a compound of fluoropyrimidine S-1, oxaliplatin, and all-trans-retinoic acid (ATRA), in gastric cancer cell lines with high SLC2A1 expression. High SLC2A1 expression was associated with poor prognosis, cancer cell proliferation, decreased immune cells, including CD8 T cells and B cells, and a low prognostic nutrition index, representing body nutrition-related status. In pathway network analysis, SLC2A1 was indirectly linked to the retinoic signaling pathway and negatively regulated immune cells/receptors. In the drug response analysis, the drug ATRA inhibited gastric cancer cell lines with high SLC2A1 expression. Treatment involving the use of SLC2A1 could contribute to better clinical management/research for patients with gastric cancer.

Exercise-Induced Myokines can Explain the Importance of Physical Activity in the Elderly: An Overview.

Physical activity has been found to aid the maintenance of health in the elderly. Exercise-induced skeletal muscle contractions lead to the production and secretion of many small proteins and proteoglycan peptides called myokines. Thus, studies on myokines are necessary for ensuring the maintenance of skeletal muscle health in the elderly. This review summarizes 13 myokines regulated by physical activity that are affected by aging and aims to understand their potential roles in metabolic diseases. We categorized myokines into two groups based on regulation by aerobic and anaerobic exercise. With aging, the secretion of apelin, ß-aminoisobutyric acid (BAIBA), bone morphogenetic protein 7 (BMP-7), decorin, insulin-like growth factor 1 (IGF-1), interleukin-15 (IL-15), irisin, stromal cell-derived factor 1 (SDF-1), sestrin, secreted protein acidic rich in cysteine (SPARC), and vascular endothelial growth factor A (VEGF-A) decreased, while that of IL-6 and myostatin increased. Aerobic exercise upregulates apelin, BAIBA, IL-15, IL-6, irisin, SDF-1, sestrin, SPARC, and VEGF-A expression, while anaerobic exercise upregulates BMP-7, decorin, IGF-1, IL-15, IL-6, irisin, and VEGF-A expression. Myostatin is downregulated by both aerobic and anaerobic exercise. This review provides a rationale for developing exercise programs or interventions that maintain a balance between aerobic and anaerobic exercise in the elderly.

Clinical Characteristics of and Risk Factors for Fever after Endobronchial Ultrasound-Guided Transbronchial Needle Aspiration: A Retrospective Study Involving 6336 Patients.

Endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) is a minimally invasive diagnostic for mediastinal and hilar lymphadenopathy/mass. This study investigated fever incidence and associated risk factors after EBUS-TBNA in 6336 patients who underwent EBUS-TBNA at Asan Medical Center from October 2008 to February 2018. Bronchoscopists evaluated participants' medical records for fever the 24 h following EBUS-TBNA. Patients were placed in either a Fever group (n = 665) or a non-Fever group (n = 5671). Fever developed in 665 of 6336 patients (10.5%) with a mean peak body temperature of 38.3 °C (range, 37.8-40.6 °C). Multivariate analysis revealed that fever-associated risk factors after EBUS-TBNA are older age (adjusted OR 0.015, 95% CI (0.969-0.997), p = 0.015), bronchoscopic washing (adjusted OR 1.624, 95% CI (1.114-2.368), p = 0.012), more than four samples of EBUS-TBNA (adjusted OR 2.472, 95% CI (1.288-4.745), p = 0.007), hemoglobin levels before EBUS-TBNA (adjusted OR 0.876, 95% CI (0.822-0.933), p < 0.001), CRP levels before EBUS-TBNA (adjusted OR 1.115, 95% CI (1.075-1.157), p < 0.001), and a diagnosis of EBUS-TBNA tuberculosis (adjusted OR 3.409, 95% CI (1.870-6.217), p < 0.001). Clinicians should be aware of the possibility of fever after EBUS-TBNA because it is common. Additional, prospective, large-scale research should assess the need for prophylactic antibiotics for EBUS-TBNA.

Expression of UTX Indicates Poor Prognosis in Patients With Luminal Breast Cancer and is Associated With MMP-11 Expression.

Ubiquitously transcribed tetratricopeptide repeat, X chromosome (UTX) is involved in the epigenetic regulation. A previous mouse xenograft study revealed that UTX knockdown is associated with downregulated expression of matrix metalloproteinase-11 (MMP-11). The authors investigated 224 cases of breast cancer from Kangbuk Samsung Medical Center between 2000 and 2005. Nuclear UTX and cytoplasmic MMP-11 expressions were assessed using immunohistochemistry of tumor tissue microarray specimens. The relationships between the expression of UTX, MMP-11, and patients' outcomes were analyzed. UTX expression was significantly associated with high histologic grade, lymphatic invasion, vascular invasion, and tumoral expression of MMP-11. Survival analysis revealed that patients with UTX expression had a poorer overall survival rate (P=0.010) as well as diminished disease-free survival rate (P=0.001). The prognostic power of UTX expression was significant in patients with luminal-type breast cancer (P=0.027, overall survival; P=0.008, disease-free survival). Validation of UTX can provide further prognostic information beyond traditional indicators and represents a potential therapeutic target for breast cancer.

Resident Working Hour Restrictions Increased the Workload of the Medical Emergency Team: A Retrospective Observational Study.

BACKGROUND: Restrictions to residents' working hours have been shown to increase the workload of other medical resources; few studies have measured the effects on medical emergency teams (METs). OBJECTIVES: This study evaluated how limiting residents' working hours affected the workload of MET in a pulmonology unit. METHODS: This retrospective observational study analyzed MET activity during periods before and after we limited the working hours of residents in our pulmonary unit to 88 h/wk: Period 1, March 2014 to February 2015; and Period 2, March 2015 to February 2016. Medical emergency team activities, dose (activations/1000 admissions), intensive care unit transfers, and mortality were compared between the two periods for weekdays and for weekends and holidays. RESULTS: There were no significant differences between the two periods in MET dose (85.0 in Period 1 versus 91.3 in Period 2, P = 0.675), intensive care unit transfers (P = 0.828), 30-day mortality (P = 0.701), and 60-day mortality (P = 0.531). However, some activities increased significantly or near significantly in Period 2, including portable echocardiography (P < 0.001), arterial line insertion (P = 0.034), mechanical ventilation (P = 0.063), and fluid therapy (P = 0.220). These increases were greater for weekends and holidays than for weekdays. CONCLUSIONS: Since December 2017, a specific law for improving the training environment and status of residents has been implemented and applied at all hospitals in Korea. This legal restriction to working hours raises concerns regarding other medical personnel and system improvements to ensure patient safety and care continuity.

Higher Acid-Base Imbalance Associated with Respiratory Failure Could Decrease the Survival of Patients with Scrub Typhus during Intensive Care Unit Stay: A Gene Set Enrichment Analysis.

Ninety percent of patients with scrub typhus (SC) with vasculitis-like syndrome recover after mild symptoms; however, 10% can suffer serious complications, such as acute respiratory failure (ARF) and admission to the intensive care unit (ICU). Predictors for the progression of SC have not yet been established, and conventional scoring systems for ICU patients are insufficient to predict severity. We aimed to identify simple and robust indicators to predict aggressive behaviors of SC. We evaluated 91 patients with SC and 81 non-SC patients who were admitted to the ICU, and 32 cases from the public functional genomics data repository for gene expression analysis. We analyzed the relationships between several predictors and clinicopathological characteristics in patients with SC. We performed gene set enrichment analysis (GSEA) to identify SC-specific gene sets. The acid-base imbalance (ABI), measured 24 h before serious complications, was higher in patients with SC than in non-SC patients. A high ABI was associated with an increased incidence of ARF, leading to mechanical ventilation and worse survival. GSEA revealed that SC correlated to gene sets reflecting inflammation/apoptotic response and airway inflammation. ABI can be used to indicate ARF in patients with SC and assist with early detection.

Author Correction: Dual-organ invasion is associated with a lower survival rate than single-organ invasion in distal bile duct cancer: A multicenter study.

A correction to this article has been published and is linked from the HTML and PDF versions of this paper. The error has been fixed in the paper.

High TNFRSF12A level associated with MMP-9 overexpression is linked to poor prognosis in breast cancer: Gene set enrichment analysis and validation in large-scale cohorts.

BACKGROUND: Matrix metalloproteinase-9 (MMP-9) is associated with remodelling of the extracellular matrix and invasion in various cancers. Identifying proteins connected to high MMP-9 expression is important in explaining its mechanisms. Our study aims to shed light on genes associated with high MMP-9 expression and to discuss their clinical impact in breast cancer. METHODS: We evaluated 173 breast cancer cases from the Kangbuk Samsung Hospital, with 1964 cases from the Molecular Taxonomy of Breast Cancer International Consortium database serving as a validation cohort. We investigated relationships between MMP-9 expression and clinicopathological characteristics. We then used gene set enrichment analyses to detect the association of genes with MMP-9 overexpression, and performed survival analyses to determine the significance of the gene in three independent cohorts. RESULTS: High MMP-9 expression correlated with poor prognosis in univariate and multivariate analyses. Using gene set enrichment analysis, we found that tumour necrosis factor receptor superfamily member 12A (TNFRSF12A) was linked to high MMP-9 expression. In the survival analysis of three published data sets (METABRIC, GSE1456, GSE20685), high TNFRSF12A was relevant to a poor survival rate. CONCLUSIONS: High levels of TNFRSF12A associated with MMP-9 overexpression may be important to explain the progression of breast cancer, and survival could be improved using therapy targeting TNFRSF12A.

Dual-organ invasion is associated with a lower survival rate than single-organ invasion distal bile duct cancer: A multicenter study.

The revised criteria of the 8th American Joint Committee on Cancer (AJCC) cancer staging system consider depth of invasion as one of the factors that determine stage in distal bile duct (DBD) cancer, but exclude adjacent organ invasion. The aims were to evaluate the association between adjacent organ invasion and relapse-free survival (RFS) and overall survival (OS) after curative surgical resection of DBD cancer and to propose optimal criteria for predicting clinical outcomes. In this retrospective cohort study, 378 patients with DBD cancer treated in multi-institutions between 1996 and 2013 were investigated. This study evaluated the relationship between clinicopathologic parameters and adjacent organ invasion and used organ invasion to compare the survival times of each group. Among 204 patients with adjacent organ invasion, 152 were in the single-organ invasion group and 52 were in the dual-organ invasion group based on a review of microscopic slides. In univariate and multivariate analyses, patients with dual-organ invasion had a shorter RFS and OS time than those with single-organ invasion. Organ invasion should be included as one of the factors that determine the AJCC stage; this might ultimately help to predict better the survival rate of patients with DBD cancer.

IgG4-Related Lung Disease without Elevation of Serum IgG4 Level: A Case Report.

Since IgG4-related pancreatitis was first reported in 2001, IgG4-related disease has been identified in other organs such as salivary gland, gallbladder, thyroid, retroperitoneum and kidney; but lung invasion is rare. A 63-year-old man presented with hemoptysis at the pulmonary clinic and chest computed tomography revealed about 4.1 cm irregular shaped mass with spiculated margin at the left upper lobe. Despite no elevation of serum IgG4 level, he was finally diagnosed as IgG4-related lung disease by transthoracic needle biopsy. After treatment with oral glucocorticoids, hemoptysis disappeared and the size of lung mass was decreased.

High Ki67/BCL2 index is associated with worse outcome in early stage breast cancer.

AIMS: Breast cancers are heterogeneous, making it essential to recognise several biomarkers for cancer outcome predictions. Ki67 proliferation index and B cell lymphoma 2 (BCL2) proteins are widely used as prognostic indicators in many types of malignancies. While Ki67 is a marker of normal or tumour cell proliferation, BCL2 plays a central role in antiproliferative activities. A combination of these two biomarkers with contrary purposes can provide enhanced prognostic accuracy than an analysis using a single biomarker. METHODS: We evaluated Ki67 and BCL2 expression with 203 cases of breast cancer. The relative expression of each biomarker named as Ki67/BCL2 index was divided into two groups (low vs high) with the use of area under receiver operating characteristic curves. RESULTS: There were significant correlations between Ki67/BCL2 index and clinicopathological findings such as age, tumour stage, size and necrosis, histological grade, extensive intraductal component, lymphatic and vascular invasion, oestrogen receptor, progesterone receptor, human epithelial growth factor receptor 2 and p53 expression (all p<0.05). In univariate and multivariate analyses, high Ki67/BCL2 index correlated with shorter disease-free survival and overall survival in patients with early stage invasive ductal carcinoma (all p<0.05). CONCLUSIONS: The Ki67/BCL2 index should be considered as a prognostic predictor in patients with early stage invasive ductal carcinoma.

Transfusion Related Acute Lung Injury after Cesarean Section in a Patient with HELLP Syndrome.

Transfusion-related acute lung injury (TRALI) is a serious adverse reaction of transfusion, and presents as hypoxemia and non-cardiogenic pulmonary edema within 6 hours of transfusion. A 14-year-old primigravida woman at 34 weeks of gestation presented with upper abdominal pain without dyspnea. Because she showed the syndrome of HELLP (hemolysis, elevated liver enzymes, and low platelet count), an emergency cesarean section delivery was performed, and blood was transfused. In the case of such patients, clinicians should closely observe the patient's condition at least during the 6 hours while the patient receives blood transfusion, and should suspect TRALI if the patient complains of respiratory symptoms such as dyspnea. Furthermore, echocardiography should be performed to distinguish between the different types of transfusion-related adverse reactions.

Negative association between GATA3 and fascin could predict relapse-free and overall survival in patients with breast cancer.

GATA3 and fascin proteins are known prognostic markers in several cancers. GATA3 is a key regulator of mammary gland morphogenesis and luminal cell differentiation, whereas fascin is a pro-metastatic actin-bundling protein. In this study, we analyzed and compared the predictive abilities of GATA3 and fascin for clinical outcomes of patients with breast cancer. The combined expression pattern based on GATA3-/+ and fascin-/+ was evaluated by immunostaining using a tissue microarray, and relationships between protein expression and several clinicopathological parameters were analyzed. GATA3 expression was associated with good prognostic parameters, but fascin was correlated with poor prognostic parameters. On comparing GATA3 and fascin, we found an inverse relationship between fascin and GATA3 expressions. On analysis of combined markers, GATA3+/fascin- was correlated with improved clinical outcomes compared to GATA3-/fascin+. Univariate and multivariate analyses revealed significant differences in relapse-free and overall survival between GATA3+/fascin- and GATA3-/fascin+. Combined marker analysis of GATA3/fascin showed an inverse association and improved prognostic information for patients with breast cancer.