First Clinical Medical College of Southern Medical University, Nanfang Hospital of Southern Medical University, Guangzhou 510515, China.
Department of Hematology, Beijing Tongren Hospital, Capital Medical University, Beijing 100730, China.
J Immunol Res. 2021 Jun 8;2021:5564568. doi: 10.1155/2021/5564568. eCollection 2021.
BACKGROUND: Diffuse large B cell lymphoma (DLBCL) is a life-threatening malignant tumor characterized by heterogeneous clinical, phenotypic, and molecular manifestations. Given the association between immunity and tumors, identifying a suitable immune biomarker could improve DLBCL diagnosis. METHODS: We systematically searched for DLBCL gene expression microarray datasets from the GEO database. Immune-related genes (IRGs) were obtained from the ImmPort database, and 318 transcription factor (TF) targets in cancer were retrieved from the Cistrome Cancer database. An immune-related classifier for DLBCL prognosis was constructed using Cox regression and LASSO analysis. To assess differences in overall survival between the low- and high-risk groups, we analyzed the tumor microenvironment (TME) and immune infiltration in DLBCL using the ESTIMATE and CIBERSORT algorithms. WGCNA was applied to study the molecular mechanisms explaining the clinical significance of our immune-related classifier and TFs. RESULTS: Eighteen IRGs were selected to construct the classifier. The multi-IRG classifier showed powerful predictive ability. Patients with a high-risk score had poor survival. Based on the AUC for three- and five-year survival, the classifier exhibited better predictive power than clinical data. Discrepancies in overall survival between the low- and high-risk score groups might be explained by differences in immune infiltration, TME, and transcriptional regulation. CONCLUSIONS: Our study describes a novel prognostic IRG classifier with strong predictive power in DLBCL. Our findings provide valuable guidance for further analysis of DLBCL pathogenesis and clinical treatment.
背景：弥漫性大 B 细胞淋巴瘤（DLBCL）是一种危及生命的恶性肿瘤，其临床、表型和分子表现具有异质性。鉴于免疫与肿瘤之间的关联，确定合适的免疫生物标志物可能有助于改善 DLBCL 的诊断。
方法：我们从 GEO 数据库中系统地搜索了 DLBCL 的基因表达微阵列数据集。从 ImmPort 数据库中获取免疫相关基因（IRGs），并从 Cistrome Cancer 数据库中检索了 318 个癌症转录因子（TF）靶标。使用 Cox 回归和 LASSO 分析构建了用于 DLBCL 预后的免疫相关分类器。为了评估低风险和高风险组之间总生存率的差异，我们使用 ESTIMATE 和 CIBERSORT 算法分析了 DLBCL 的肿瘤微环境（TME）和免疫浸润。应用 WGCNA 研究了解释我们的免疫相关分类器和 TF 临床意义的分子机制。
结果：选择了 18 个 IRG 来构建分类器。多 IRG 分类器具有强大的预测能力。高风险评分的患者生存率较差。基于三年和五年生存率的 AUC，该分类器的预测能力优于临床数据。低风险和高风险评分组之间总生存率的差异可能与免疫浸润、TME 和转录调控的差异有关。
结论：本研究描述了一种新的预后性 IRG 分类器，在 DLBCL 中具有强大的预测能力。我们的研究结果为进一步分析 DLBCL 的发病机制和临床治疗提供了有价值的指导。
Zhonghua Xue Ye Xue Za Zhi. 2019-9-14
Zotero 插件
