肿瘤微环境细胞亚群基因标签预测复发转移性头颈鳞状细胞癌EGFR单抗疗效

Gene Signatures of Cell Subpopulations in Tumor Microenviroment Predict Response to EGFR Monoclonal Antibodies in Recurrent and Metastatic Squamous Cell Carcinoma of Head and Neck

目的:利用复发和转移性头颈部鳞癌肿瘤微环境中细胞亚群特征表达基因构建抗EGFR单抗疗效预测模型。方法:通过Gene Expression Omnibus数据库搜集有抗EGFR单抗无进展生存时间(progression free survival,PFS)记录和治疗前肿瘤组织转录组微阵列芯片表达值数据集,并从文献获取肿瘤微环境细胞亚群特征表达基因。发现集中,在长PFS组和短PFS组间分析差异表达基因并计算各个细胞亚群敏感指数。最小绝对值收敛和选择算子(least absolute shrinkage and selection operator,LOSSA)算法筛选与长PFS显著相关的细胞亚群。线性权重法构建抗EGFR单抗敏感评分(anti-EGFR antibodies sensitive score,EASS),并在独立验证集中验证。结果:在GSE65021发现集中,9个细胞亚群中计算得到EASS中位值为30.73(范围:26.24~41.29),ROC曲线表明EASS可完全区别长PFS组人群与短PFS人群(AUC:1.000,95%CI:1.000~1.000,P<0.001)。在GSE102995验证集中,对分期(IV vs II^III)和PS评分校正后,EASS依然是PFS的独立预后因素(HR=0.647,95%CI:0.503~0.831,P=0.001)。EASS对PFS预测准确性的C指数为0.755(95%CI:0.719~0.791),显著高于分期C指数0.542(95%CI:0.336~0.748,P=0.032)。结论:EASS预测模型可很好地预测复发和转移性头颈部鳞癌抗EGFR抗体疗效,值得进一步研究和验证。

Objective:Objective:To develop a model for predicting response to anti-epidermal growth factor receptor(anti-EGFR)antibody by using feature genes of cell subpopulations in the tumor microenvironment of recurrent and metastatic squamous cell carcinoma of the head and neck.Methods:Gene Expression Omnibus datasets with records of progression free survival after treated with anti-EGFR antibody and gene expression profiles evaluated by microarray assay were retrieved.The feature genes peculiar to each of cell subpopulations in the tumor microenvironment was obtained from previously published articles.Differentially expressed genes were firstly identified between the long progression free survival(PFS)group and the short PFS group in the discovery set and used to calculate sensitive indexes of each cell subpopulation.Least absolute shrinkage and selection operator was used to establish the anti-EGFR antibody sensitivity score(EASS)which was further verified in the validation set.Results:The median EASS deduced from the nine cell subpopulations was 30.73(26.24-41.29)in the discovery set GSE65021.The long PFS group was significantly discriminated from the short PFS group through EASS revealed by ROC analysis(AUC:1.000,95%CI:1.000-1.000,P<0.001).In the validation set GSE102995,the EASS remained an independent prognostic factor for PFS after adjusted for stage(IV vs II-III)and performance status score(HR=0.647,95%CI:0.503-0.831,P=0.001).The consensus index of EASS for predicting PFS was 0.755(95%CI:0.719-0.791),which was significantly higher than that of stage 0.542(95%CI:0.336-0.748)(P=0.032).Conclusion:The EASS is efficient to predict the response to anti-EGFR antibody in the recurrence and metastasis of squamous cell carcinoma of the head and neck,which is worthy of further study and validation.