Department of Thyroid Surgery, Guangzhou First People's Hospital, Guangzhou Medical University, Guangzhou, China.
Department of Thyroid Surgery, School of Medicine, Guangzhou First People's Hospital, South China University of Technology, Guangzhou, Guangdong, China.
J Endocrinol Invest. 2021 Oct;44(10):2153-2163. doi: 10.1007/s40618-021-01514-7. Epub 2021 Feb 23.
Increasing evidence indicates that there is a correlation between papillary thyroid carcinoma (PTC) prognosis and the immune signature. Our goal was to construct a new prognostic tool based on immune genes to achieve more accurate prognosis predictions and earlier diagnoses of PTC.
The 493 PTCs samples and 58 tumor-adjacent normal tissues were obtained from The Cancer Genome Atlas database (TCGA). Immune genes were obtained from the ImmPort database. First, this cohort was randomly divided into training cohort and testing cohort. Second, the differentially expressed (DE) immune genes from the training set were used to construct the prognostic model. Then, the testing and entire data cohorts were used to validate the model, and the data were analyzed to determine the correlation of the clinical prognostic model with immune cell infiltration and expression profiles of human leukocyte antigen (HLA) genes. Finally, an analysis of the gene ontology (GO) annotation was performed.
A total of 189 upregulated and 128 downregulated DE immune genes were identified. We developed and validated a three-immune gene model for PTC that includes Hsp70, NOX5, and FGF23. This model was demonstrated to be an independent prognostic variable. In addition, the overall immune activity of the high-risk group was higher than that of the low-risk group.
We developed and validated a three-immune gene model for PTC that includes HSPA1A, NOX5, and FGF23. This model can be used as a validated tool to predict outcomes in PTC.
越来越多的证据表明，甲状腺乳头状癌（PTC）的预后与免疫特征之间存在相关性。我们的目标是构建一个基于免疫基因的新预后工具，以实现更准确的 PTC 预后预测和早期诊断。
从癌症基因组图谱数据库（TCGA）获得 493 例 PTC 样本和 58 例肿瘤旁正常组织。免疫基因从 ImmPort 数据库中获得。首先，将该队列随机分为训练队列和测试队列。其次，使用训练集中的差异表达（DE）免疫基因构建预后模型。然后，使用测试集和整个数据队列验证模型，并分析数据以确定临床预后模型与免疫细胞浸润和人类白细胞抗原（HLA）基因表达谱的相关性。最后，进行基因本体（GO）注释分析。
共鉴定出 189 个上调和 128 个下调的 DE 免疫基因。我们开发并验证了一个包含 Hsp70、NOX5 和 FGF23 的 PTC 三免疫基因模型。该模型被证明是一个独立的预后变量。此外，高危组的整体免疫活性高于低危组。
我们开发并验证了一个包含 HSPA1A、NOX5 和 FGF23 的 PTC 三免疫基因模型。该模型可作为验证工具，用于预测 PTC 的结局。
