甲状腺乳头状癌转移相关基因分类器预后模型的建立与验证

A novel metastasis-related genes-based signature for predicting progression-free interval of papillary thyroid carcinoma

目的甲状腺乳头状癌(papillary thyroid carcinoma,PTC)的发病率居高不下,本研究着眼于建立新型的转移相关基因分类器及相关的预后模型,以有效预测PTC的无进展生存期(progression free interval,PFI),并识别复发高危风险的患者。方法利用转移相关基因(metastasis-related genes,MTGs)与人类癌症基因图谱(The Cancer Genome Atlas,TCGA)合并分析,并使用套索回归分析及受试者曲线(receiver operating characteristic curve,ROC)分析来建立相应的新型基因分类器,并与相关的临床特征一起建立预后模型,以预测PTC术后的PFI。最后,在外部数据集和不同细胞系上验证基因分类器的有效性。本研究使用R 3.6.3和GraphPad Prism 8软件进行统计学分析。结果通过差异整合分析提取了155预后相关的MTGs,从中建立并优化了新型的10基因分类器,预后分析证实分类器评分是PTC的重要独立预后因素。最后,建立了包括新型分类器的回归模型。新型分类器的曲线下面积(area under curve,AUC)为0.76,回归模型的AUC为0.80,同PTC的各种进展期临床特征高度符合,此外,新型分类器与关键临床特征和细胞系的侵袭性密切相关。结论基于转移相关基因建立的新型分类器和回归模型与PTC的预后密切相关,将有助于临床实践中对PTC患者预后进行精准的个体化预测。

Objective We aimed to build a novel model with metastasis-related genes(MTGs)signature for predicting progression-free interval(PFI)of papillary thyroid carcinoma(PTC).Methods We integrated PTC datasets with the MTGs to identify differentially expressed MTGs(DE-MTGs),then we established a novel MTGs based signature and validated it in external datasets and cell lines.Finally,we established a signature and clinical parameters-based nomogram for predicting the PFI of PTC.Results We identified 155 DE-MTGs related to PFI in PTC.The functional enrichment analysis showed that the DE-MTGs were associated with essential oncogenic processes.Consequently,we established and optimized a novel 10-gene signature.The novel signature had a C-index of 0.76 and the relevant nomogram had a C-index of 0.80.Also,it was closely related to pivotal clinical characters of multiple datasets and invasiveness of PTC cell lines.And the signature was an independent prognostic factor in PTC.Finally,we built a nomogram including the signature and relevant clinical factors.The efficacy was satisfying in predicting PTC’s PFI.Conclusions The MTG signature and nomogram were closely associated with PTC prognosis and may help clinicians improve the individualized prediction of PFI.