Artificial intelligence-based prognostic model accurately predicts the survival of patients with diffuse large B-cell lymphomas: analysis of a large cohort in China DOI Creative Commons

Huilin Peng,

M. Su,

Xiang Guo

и другие.

BMC Cancer, Год журнала: 2024, Номер 24(1)

Опубликована: Май 22, 2024

Abstract Background Diffuse large B-cell lymphomas (DLBCLs) display high molecular heterogeneity, but the International Prognostic Index (IPI) considers only clinical indicators and has not been updated to include data. Therefore, we developed a widely applicable novel scoring system with screened by artificial intelligence (AI) that achieves accurate prognostic stratification promotes individualized treatments. Methods We retrospectively enrolled cohort of 401 patients DLBCL from our hospital, covering period January 2015 2019. included 22 variables in analysis assigned them weights using random survival forest method establish new predictive model combining bidirectional long-short term memory (Bi-LSTM) logistic hazard techniques. compared performance “molecular-contained model” (McPM) IPI. In addition, simplified version McPM (sMcPM) enhance its practical applicability settings. also demonstrated improved risk capabilities sMcPM. Results Our showed superior accuracy, as indicated C-index low integrated Brier score (IBS), for both overall (OS) progression-free (PFS). The was better than IPI based on receiver operating characteristic (ROC) curve fitting. selected five key indicators, including extranodal involvement sites, lactate dehydrogenase (LDH), MYC gene status, absolute monocyte count (AMC), platelet (PLT) sMcPM, which is more suitable applications. sMcPM similar OS results ( P < 0.0001 both) significantly PFS vs. = 0.44 IPI). Conclusions McPM, variables, IPI, rendering it era. Moreover, may become used effective tool guide individual precision treatments drive drug development.

Язык: Английский

Prediction of Diabetic Kidney Disease in Newly Diagnosed Type 2 Diabetes Mellitus DOI Creative Commons
Xiaodie Mu, Aihua Wu, Huiyue Hu

и другие.

Diabetes Metabolic Syndrome and Obesity, Год журнала: 2023, Номер Volume 16, С. 2061 - 2075

Опубликована: Июль 1, 2023

Background: Diabetic kidney disease (DKD), a common microvascular complication of diabetes mellitus (DM), is always asymptomatic until it develops to the advanced stage. Thus, we aim develop nomogram prediction model for progression DKD in newly diagnosed type 2 (T2DM). Methods: This was single-center analysis prospective data collected from 521 patients with T2DM. All related clinical records were incorporated, including triglyceride-glucose index (TyG index). The least absolute shrinkage and selection operator (LASSO) used build model. In addition, discrimination, calibration, practicality evaluated. Results: this study, 156 participants incorporated as validation set, while remaining 365 into training set. predictive factors included individualized 5 variables. area under curve (AUC) 0.826 (95% CI 0.775 0.876), indicating excellent discrimination performance. performed exceptionally well terms accuracy applicability, according calibration curves decision analysis. Conclusion: risk T2DM had outstanding which could help practice. Keywords: diabetic disease, mellitus, nomogram, model, assessment

Язык: Английский

Процитировано

7

Identification of ACAA1 and HADHB as potential prognostic biomarkers based on a novel fatty acid oxidation-related gene model in head and neck squamous cell carcinoma: A retrospective study DOI
Wanli Wang,

Hong-Jie Jiang,

Ze-Liang Shen

и другие.

Archives of Oral Biology, Год журнала: 2024, Номер 163, С. 105982 - 105982

Опубликована: Апрель 24, 2024

Язык: Английский

Процитировано

2

SLC27A2 is a potential immune biomarker for hematological tumors and significantly regulates the cell cycle progression of diffuse large B-cell lymphoma DOI Creative Commons
Yi Wang,

Xue Chen,

Yun Li

и другие.

BMC Medical Genomics, Год журнала: 2024, Номер 17(1)

Опубликована: Апрель 25, 2024

Abstract Background Research on the fatty acid metabolism related gene SLC27A2 is currently mainly focused solid tumors, and its mechanism of action in hematological tumors has not been reported. Method This study aims to explore pathological immune mechanisms verify functional role through cell experiments improve treatment decisions clinical outcomes tumors. Result identified as a common differentially expressed between DLBCL AML. Immune microenvironment analysis showed that was significantly positively correlated with T CD4 + , CD8 endothelial cells, macrophages, NK cells DLBCL. In AML, there significant negative correlation B macrophages. participates process The GESA results indicate high expression involved pathway, cycle pathway low Therefore, pathways, have also confirmed regulation cells. Conclusion summary, our research comprehensively report for first time apoptotic effects experiments. can help AML patients.

Язык: Английский

Процитировано

2

Identification and validation of a novel robust glioblastoma prognosis model based on bioinformatics DOI Creative Commons
Le Zhang, Xiaoling Yan, Yahong Wang

и другие.

Heliyon, Год журнала: 2024, Номер 10(18), С. e37374 - e37374

Опубликована: Сен. 1, 2024

Glioblastoma (GBM) is a very common primary malignant tumor of the central nervous system (CNS). Aging, macrophage, autophagy, and methylation related genes are hypothesized to be crucial its pathogenesis. In this study, we aimed explore role these in prognosis GBM.

Язык: Английский

Процитировано

2

Downregulation of T-cell cytotoxic marker IL18R1 promotes cancer proliferation and migration and is associated with dismal prognosis and immunity in lung squamous cell carcinoma DOI Creative Commons

Qiang Guo,

Chuangyan Wu,

Ni Jiang

и другие.

Frontiers in Immunology, Год журнала: 2022, Номер 13

Опубликована: Дек. 5, 2022

Immunotherapy can improve the survival of patients with advanced lung squamous cell carcinoma (LUSC). T cytotoxic cells are one main members immune microenvironment. Herein, we aimed to identify roles T-cell markers interleukin 18 (IL18) receptor 1 (IL18R1) in LUSC progression using bioinformatics, clinical tissue specimen, and experiment. We assessed association between IL18R1 expression infiltration IL18R1-related competing RNA network. The was downregulated tissues. downregulation associated diagnosis short overall disease-specific survival, it also an independent risk factor for dismal time LUSC. nomograms predicted overexpression inhibited proliferation, migration, invasion cells. significantly microenvironment (stromal, immune, estimate scores), (such as cells, CD8 cells), CD8A, PD-1, CTLA4) AC091563.1 RBPMS-AS1 positively expression, negatively miR-128-3p In conclusion, prognosis inhibits growth migration cancer Furthermore, might compete bind participating progression. These results showed that is a biomarker evaluating

Язык: Английский

Процитировано

12

Comprehensive analysis of the prognostic implication and immune infiltration of CISD2 in diffuse large B-cell lymphoma DOI Creative Commons

ChaoFeng Zhang,

Qi Lin,

Chuntuan Li

и другие.

Frontiers in Immunology, Год журнала: 2023, Номер 14

Опубликована: Дек. 12, 2023

Background Diffuse large B-cell lymphoma (DLBCL) is the most common in adults. CDGSH iron sulfur domain 2 (CISD2) an iron–sulfur protein and plays a critical role of cell proliferation. The aberrant expression CISD2 associated with progression multiple cancers. However, its DLBCL remains unclear. Methods differential was identified via public databases, quantitative real-time PCR (qRT-PCR) western blot were used to identifed CISD2. We estimated impact on clinical prognosis using Kaplan-Meier plotter. Meanwhile, drug sensitivity assessed CellMiner database. 100 CISD2-related genes from STRING obtained analyzed LASSO Cox regression. A related signature for risk model (CISD2Risk) established. PPI network CISD2Risk performed, functional enrichment conducted through DAVID impacts features analyzed. ESTIMATE, CIBERSORT, MCP-counter algorithm identify immune infiltration. Subsequently, Univariate multivariate regression analysis applied, prognostic nomogram, accompanied by calibration curve, constructed predict 1-, 3-, 5-years survival probabilities. Results upregulated patients comparing normal controls datasets, similarly, highly expressed lines. Overexpression poor based GSE31312, GSE32918, GSE93984 datasets (P&lt;0.05). Nine drugs considered as potential therapeutic agents By cox regression, twenty seven construct CISD2Risk, biological functions these might be involved apoptosis P53 signaling pathway. high value had worse effect higher stromal score, ESTIMATE score lowe value, negatively correlated several infiltrating cells (macrophages M0 M1, CD8 T cells, CD4 naïve NK cell, etc) that better prognosis. Additionally, independent factor both univariate nomogram produced accurate predictions curves good agreement. Conclusion Our study demonstrates have successfully validated prediction efficacy monitoring included 27 genes. may promising evaluator infiltration serve reference decision-making patients.

Язык: Английский

Процитировано

6

A novel angiogenesis-related scoring model predicts prognosis risk and treatment responsiveness in diffuse large B-cell lymphoma DOI
Yu Liu, Jinhua Wang,

Xiaochen Shen

и другие.

Clinical and Experimental Medicine, Год журнала: 2023, Номер 23(7), С. 3781 - 3797

Опубликована: Июль 4, 2023

Язык: Английский

Процитировано

6

A Novel and Validated Inflammation-Based Prognosis Score (IBPS) Predicts Outcomes in Patients with Diffuse Large B-Cell Lymphoma DOI Creative Commons
Yajiao Liu, Sheng Li, Haiying Hua

и другие.

Cancer Management and Research, Год журнала: 2023, Номер Volume 15, С. 651 - 666

Опубликована: Июль 1, 2023

We aimed to create a novel prognostic score, the inflammation-based prognosis score (IBPS). In addition, we attempted establish and validate nomogram predict overall survival (OS) of patients with DLBCL.We retrospectively investigated data 213 DLBCL diagnosed treated in Affiliated Hospital Jiangnan University used these develop nomograms. At same time, 89 Wuxi People's Nanjing Medical from January 2015 June 2021 were collected as an external validation cohort. developed IBPS through least absolute shrinkage selection operator (LASSO) Cox regression. The univariate multivariate regression method was nomogram. concordance index (C-index), calibration chart, time-dependent receiver operating characteristic (ROC) analysis, decision curve analysis (DCA), Kaplan-Meier assess nomogram.The systemic immune inflammation (SII), nutrition (PNI), modified Glasgow (mGPS) construct IBPS. Eastern Cooperative Oncology Group performance status (ECOG PS), IBPS, response treatment, whether accept surgery OS patients. C-index training cohorts 0.844 0.828, respectively. According ROC DCA, has good predictive accuracy clinical net benefit. showed that according could be classified into three risk groups.In DLBCL, baseline reliable predictor OS. probability can precisely predicted using based on

Язык: Английский

Процитировано

5

A Novel Lipid Metabolism and Endoplasmic Reticulum Stress-Related Risk Model for Predicting Immune Infiltration and Prognosis in Colorectal Cancer DOI Open Access
Hao-Ran Jin, Bihan Xia, Jin Wang

и другие.

International Journal of Molecular Sciences, Год журнала: 2023, Номер 24(18), С. 13854 - 13854

Опубликована: Сен. 8, 2023

Lipid metabolism and endoplasmic reticulum stress exhibit crosstalk in various cancer types, which are closely associated with the progression of colorectal (CRC). This study constructs a prognostic signature based on lipid stress-related genes (LERGs) for CRC patients, aiming to predict prognosis immune response. RNA sequencing clinical data from TCGA GEO databases were analyzed identify differentially expressed LERGs relevance using univariate Cox regression. Subsequently, risk model was developed LASSO patients stratified into low-risk high-risk groups scores, cohort demonstrating poorer multiple databases. The showed robust correlations features, gene mutations, treatment sensitivity. Significant differences cell infiltration expression immune-related factors also detected between groups, elevated scores cytokines failure single-cell analysis. research indicates that correlated tumor progression, an immunosuppressive landscape, alterations drug can serve as powerful tool, offering critical insights refining management optimizing patients.

Язык: Английский

Процитировано

5

Identification of lactate regulation pattern on tumor immune infiltration, therapy response, and DNA methylation in diffuse large B-cell lymphoma DOI Creative Commons
Jinghua Wang, Yanjun Wang, Li Wan

и другие.

Frontiers in Immunology, Год журнала: 2023, Номер 14

Опубликована: Сен. 18, 2023

Lactate, produced through glycolytic metabolism in the tumor microenvironment (TME), is implicated tumorigenesis and progression diverse cancers. However, impact of lactate on remodeling TME diffuse large B-cell lymphoma (DLBCL) its implications for therapy options remain unclear.A lactate-related (LAR) scoring model was constructed DLBCL patients using bioinformatic methods. CIBERSORT, XCELL, ssGSEA algorithms were used to determine correlation between LAR score immune cell infiltration. Tumor Immune Dysfunction Exclusion (TIDE), rituximab, cyclophosphamide, adriamycin, vincristine, prednisone (R-CHOP) cohorts, Genomics Drug Sensitivity Cancer (GDSC) utilized predict therapeutic response patients. The hub gene STAT4 biological behavior DNA methylation experimentally validated or accessed by TSIDE database.The developed based 20 prognosis-related genes, which enabled division into high- low-risk groups median score. Patients with high-risk exhibited significantly worse survival outcomes both training cohorts (GSE181063) validation (GSE10846, GSE32918, GSE69053), as indicated statistically significant differences (all P<0.05) area under curve (AUC) values exceeding 0.6. analyses revealed that had higher levels infiltration antitumor activation compared Furthermore, high scores associated a lower TIDE value poor efficacy R-CHOP regimen. GDSC analysis identified 18 drugs sensitivity Moreover, vitro experiments demonstrated overexpression key could suppress proliferation migration, induce cycle arrest, promote apoptosis cells. Transcriptional expression found be immunomodulators chemokines.The lactate-based signature effectively predicts prognosis regulates DLBCL. Our study underscores role gene, STAT4, an important suppressor Modulating promising strategy clinical practice.

Язык: Английский

Процитировано

5