Exploration of biomarkers for predicting the prognosis of patients with diffuse large B-cell lymphoma by machine-learning analysis
Abstract Background As one distinct origin of hematological malignancies, diffuse large B-cell lymphoma (DLBCL) has caused a major public health problem. However, the molecular mechanisms that underlie this association have not been clearly elucidated. To improve this situation, it is urgent to expl...
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
BMC
2025-08-01
|
| Series: | BMC Immunology |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s12865-025-00738-z |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849226529468317696 |
|---|---|
| author | Shifen Wang Hong Tao Xingyun Zhao Siwen Wu Chunwei Yang Yuanfei Shi Zhenshu Xu Dawei Cui |
| author_facet | Shifen Wang Hong Tao Xingyun Zhao Siwen Wu Chunwei Yang Yuanfei Shi Zhenshu Xu Dawei Cui |
| author_sort | Shifen Wang |
| collection | DOAJ |
| description | Abstract Background As one distinct origin of hematological malignancies, diffuse large B-cell lymphoma (DLBCL) has caused a major public health problem. However, the molecular mechanisms that underlie this association have not been clearly elucidated. To improve this situation, it is urgent to explore disease-specific diagnostic biomarkers and mechanisms. Methods Three microarray datasets (GSE25638, GSE12195 and GSE12453) were downloaded from the Gene Expression Omnibus (GEO) database. The key genes in DLBCL patients were screened by differentially expression gene (DEG) and weighted gene co-expression network analysis (WGCNA). Functional enrichment analysis and protein-protein interaction (PPI) network construction were employed to reveal DLBCL-related pathogenic molecules and underlying mechanisms. Candidate biomarkers were screened using random forest (RF) analysis. A diagnostic nomogram and Kaplan-Meier (KM) survival analysis were constructed to predict the risk of patients. Single-sample gene set enrichment analysis (ssGSEA) was used for exploring immune cell infiltration in lymphoma. The validation of the hub genes expressions was confirmed by quantitative real-time polymerase chain reaction (qRT-PCR) and immunohistochemistry (IHC) tests. Results A total of 95 key genes were acquired from three datasets of DLBCL patients by DEG analysis and WGCNA. DEGs were significantly enriched in pathways associated with inflammatory response, biological process involved in interspecies interaction between organisms, C-X-C chemokine receptor binding as well as chemokine activity. This was determined by Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) analyses. Moreover, four hub genes (CXCL9, CCL18, C1QA and CTSC) were significantly screened from the three datasets using RF algorithms. They were closely correlated with the overall survival of DLBCL patients. The dysregulated infiltration of immune cells, including natural killer (NK) cells and T cells, were positively linked to the expression levels of the four hub genes. The receiver operating characteristic (ROC) results were promising via the construction of a nomogram model. Additionally, the increased expression of the four key genes was further verified in DLBCL patients. Conclusion Four crucial hub genes (CXCL9, CCL18, C1QA and CTSC) that could predict the risk of DLBCL were systematically identified. In particular, CXCL9 may be the most important potential biomarker for the progression of DLBCL patients. |
| format | Article |
| id | doaj-art-cf5571d9751a4bc99a6048e00b61e501 |
| institution | Kabale University |
| issn | 1471-2172 |
| language | English |
| publishDate | 2025-08-01 |
| publisher | BMC |
| record_format | Article |
| series | BMC Immunology |
| spelling | doaj-art-cf5571d9751a4bc99a6048e00b61e5012025-08-24T11:16:36ZengBMCBMC Immunology1471-21722025-08-0126111710.1186/s12865-025-00738-zExploration of biomarkers for predicting the prognosis of patients with diffuse large B-cell lymphoma by machine-learning analysisShifen Wang0Hong Tao1Xingyun Zhao2Siwen Wu3Chunwei Yang4Yuanfei Shi5Zhenshu Xu6Dawei Cui7Department of Blood Transfusion, The First Affiliated Hospital, Zhejiang University School of MedicineCollege of Public Health, Suzhou Vocational Health CollegeDepartment of Blood Transfusion, The First Affiliated Hospital, Zhejiang University School of MedicineDepartment of Blood Transfusion, The First Affiliated Hospital, Zhejiang University School of MedicineDepartment of Pathology, The First Affiliated Hospital, Zhejiang University School of MedicineDepartment of Hematology, The First Affiliated Hospital, Zhejiang University School of MedicineDepartment of Hematology, Fujian Provincial Key Laboratory on Hematology, Fujian Institute of Hematology, Fujian Medical University Union HospitalDepartment of Blood Transfusion, The First Affiliated Hospital, Zhejiang University School of MedicineAbstract Background As one distinct origin of hematological malignancies, diffuse large B-cell lymphoma (DLBCL) has caused a major public health problem. However, the molecular mechanisms that underlie this association have not been clearly elucidated. To improve this situation, it is urgent to explore disease-specific diagnostic biomarkers and mechanisms. Methods Three microarray datasets (GSE25638, GSE12195 and GSE12453) were downloaded from the Gene Expression Omnibus (GEO) database. The key genes in DLBCL patients were screened by differentially expression gene (DEG) and weighted gene co-expression network analysis (WGCNA). Functional enrichment analysis and protein-protein interaction (PPI) network construction were employed to reveal DLBCL-related pathogenic molecules and underlying mechanisms. Candidate biomarkers were screened using random forest (RF) analysis. A diagnostic nomogram and Kaplan-Meier (KM) survival analysis were constructed to predict the risk of patients. Single-sample gene set enrichment analysis (ssGSEA) was used for exploring immune cell infiltration in lymphoma. The validation of the hub genes expressions was confirmed by quantitative real-time polymerase chain reaction (qRT-PCR) and immunohistochemistry (IHC) tests. Results A total of 95 key genes were acquired from three datasets of DLBCL patients by DEG analysis and WGCNA. DEGs were significantly enriched in pathways associated with inflammatory response, biological process involved in interspecies interaction between organisms, C-X-C chemokine receptor binding as well as chemokine activity. This was determined by Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) analyses. Moreover, four hub genes (CXCL9, CCL18, C1QA and CTSC) were significantly screened from the three datasets using RF algorithms. They were closely correlated with the overall survival of DLBCL patients. The dysregulated infiltration of immune cells, including natural killer (NK) cells and T cells, were positively linked to the expression levels of the four hub genes. The receiver operating characteristic (ROC) results were promising via the construction of a nomogram model. Additionally, the increased expression of the four key genes was further verified in DLBCL patients. Conclusion Four crucial hub genes (CXCL9, CCL18, C1QA and CTSC) that could predict the risk of DLBCL were systematically identified. In particular, CXCL9 may be the most important potential biomarker for the progression of DLBCL patients.https://doi.org/10.1186/s12865-025-00738-zDLBCLDEGsHub genesImmune cell infiltration |
| spellingShingle | Shifen Wang Hong Tao Xingyun Zhao Siwen Wu Chunwei Yang Yuanfei Shi Zhenshu Xu Dawei Cui Exploration of biomarkers for predicting the prognosis of patients with diffuse large B-cell lymphoma by machine-learning analysis BMC Immunology DLBCL DEGs Hub genes Immune cell infiltration |
| title | Exploration of biomarkers for predicting the prognosis of patients with diffuse large B-cell lymphoma by machine-learning analysis |
| title_full | Exploration of biomarkers for predicting the prognosis of patients with diffuse large B-cell lymphoma by machine-learning analysis |
| title_fullStr | Exploration of biomarkers for predicting the prognosis of patients with diffuse large B-cell lymphoma by machine-learning analysis |
| title_full_unstemmed | Exploration of biomarkers for predicting the prognosis of patients with diffuse large B-cell lymphoma by machine-learning analysis |
| title_short | Exploration of biomarkers for predicting the prognosis of patients with diffuse large B-cell lymphoma by machine-learning analysis |
| title_sort | exploration of biomarkers for predicting the prognosis of patients with diffuse large b cell lymphoma by machine learning analysis |
| topic | DLBCL DEGs Hub genes Immune cell infiltration |
| url | https://doi.org/10.1186/s12865-025-00738-z |
| work_keys_str_mv | AT shifenwang explorationofbiomarkersforpredictingtheprognosisofpatientswithdiffuselargebcelllymphomabymachinelearninganalysis AT hongtao explorationofbiomarkersforpredictingtheprognosisofpatientswithdiffuselargebcelllymphomabymachinelearninganalysis AT xingyunzhao explorationofbiomarkersforpredictingtheprognosisofpatientswithdiffuselargebcelllymphomabymachinelearninganalysis AT siwenwu explorationofbiomarkersforpredictingtheprognosisofpatientswithdiffuselargebcelllymphomabymachinelearninganalysis AT chunweiyang explorationofbiomarkersforpredictingtheprognosisofpatientswithdiffuselargebcelllymphomabymachinelearninganalysis AT yuanfeishi explorationofbiomarkersforpredictingtheprognosisofpatientswithdiffuselargebcelllymphomabymachinelearninganalysis AT zhenshuxu explorationofbiomarkersforpredictingtheprognosisofpatientswithdiffuselargebcelllymphomabymachinelearninganalysis AT daweicui explorationofbiomarkersforpredictingtheprognosisofpatientswithdiffuselargebcelllymphomabymachinelearninganalysis |