Table 4_Identification of risk factors for high-risk dedifferentiation in papillary thyroid carcinoma and construction of discriminative model.docx

ObjectiveWe initially found that the thyroid differentiation score (TDS) was associated with the prognosis of papillary thyroid carcinoma (PTC) patients. Therefore, this study aimed to investigate the influencing factors and construct a discriminative model of high-risk dedifferentiation, and to explore the possible mechanisms.
MethodsData were sourced from the TCGA database. The influences of the TDS, tumor mutation burden, and immune score on the progression-free interval (PFI) were assessed by the Kaplan-Meier method and multivariable Cox regression. Then, logistic regression analyses were utilized to explore the factors of dedifferentiation and a nomogram model was conducted. Additionally, differentially expressed genes (DEGs) were identified using RNA sequencing data, while their regulatory pathways were determined by the Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis. Finally, the differential expression of key genes of major pathways was explored.
ResultsThis study included 391 PTC patients. After analyzing the influences of the three indicators on survival, only TDS showed an association with PFI. Multivariable logistic analysis revealed that the disease duration and PTC subtypes influenced dedifferentiation. The nomogram model based on these two variables showed improved discriminative capability. The study identified 17 overlapping DEGs associated with the dedifferentiation and three primary enrichment pathways, with complement and coagulation cascade pathways being the most significant (P<0.001). The central gene was CD55, which showed high expression in high-risk dedifferentiated and tall cell PTC, and the expression level increased as the disease progressed.
ConclusionThis research may contribute to promising identifying high-risk dedifferentiated PTC and also provide a potential therapeutic target.

研究目的：本研究最初发现甲状腺分化评分（thyroid differentiation score, TDS）与乳头状甲状腺癌（papillary thyroid carcinoma, PTC）患者的预后相关，因此旨在探究其影响因素，构建高危去分化鉴别模型，并探讨潜在的作用机制。方法：本研究数据来源于癌症基因组图谱（The Cancer Genome Atlas, TCGA）数据库。采用卡普兰-迈耶（Kaplan-Meier）法及多变量Cox回归分析，评估TDS、肿瘤突变负荷（tumor mutation burden, TMB）与免疫评分对无进展间隔期（progression-free interval, PFI）的影响；随后通过logistic回归分析探索去分化相关影响因素，并构建列线图（nomogram）模型。此外，利用RNA测序（RNA sequencing）数据鉴定差异表达基因（differentially expressed genes, DEGs），通过京都基因与基因组百科全书（Kyoto Encyclopedia of Genes and Genomes, KEGG）富集分析明确其调控通路；最终探究核心通路关键基因的差异表达情况。结果：本研究共纳入391例PTC患者。在分析上述三项指标对生存的影响后，仅TDS与PFI存在相关性。多变量logistic回归分析显示，病程时长与PTC亚型可影响肿瘤去分化。基于这两项变量构建的列线图模型具备更优异的鉴别能力。本研究共鉴定出17个与去分化相关的重叠差异表达基因，富集得到3条主要调控通路，其中补体与凝血级联通路最为显著（P<0.001）。核心基因为CD55，该基因在高危去分化型及高细胞型PTC中呈高表达，且其表达水平随疾病进展而升高。结论：本研究可为高危去分化型PTC的精准识别提供潜在参考，同时也可为该疾病的治疗提供潜在靶点。