Abstract：
BACKGROUND: Few studies have focused on DNA methylation in endometrial cancer. The aim of our study is identify its role in endometrial cancer prognosis.
METHODS: A publicly available dataset was retrieved from The Cancer Genome Atlas. For validation of expression alteration due to methylation, RNA sequencing data were obtained from other independent cohorts. MethSurv was used to search for candidate CpG probes, which were then filtered by least absolute shrinkage and selection operator Cox regression and multivariate Cox regression analyses to identify final set of CpG probes for overall survival. A methylation-based risk model was developed and receiver operating characteristic analysis with area under curve was used for evaluation. Patients were divided into high- and low-risk groups using an optimal cut-off point. Comprehensive bioinformatic analyses were conducted to identify hub genes, key transcription factors, and enriched cancer-related pathways. Kaplan-Meier curve was used for survival analysis.
RESULTS: A 5-CpG signature score was established. Its predictive value for 5-year overall survival was high, with area under curve of 0.828, 0.835 and 0.816 for the training, testing and entire cohorts. cg27487839 and cg12885678 had strong correlation with their gene expression, XKR6 and PTPRN2, and lower PTPRN2 expression was associated with poorer survival in both The Cancer Genome Atlas and the validation datasets. Low-risk group was associated with significantly better survival. Low-risk group harboured more mutations in hub genes and key transcription factors, and mutations in SP1 and MECP2 represented favourable outcome.
CONCLUSIONS: We developed a methylation-based prognostic stratification system for endometrial cancer. Low-risk group was associated with better survival and harboured more mutations in the key regulatory genes.
METHODS: A publicly available dataset was retrieved from The Cancer Genome Atlas. For validation of expression alteration due to methylation, RNA sequencing data were obtained from other independent cohorts. MethSurv was used to search for candidate CpG probes, which were then filtered by least absolute shrinkage and selection operator Cox regression and multivariate Cox regression analyses to identify final set of CpG probes for overall survival. A methylation-based risk model was developed and receiver operating characteristic analysis with area under curve was used for evaluation. Patients were divided into high- and low-risk groups using an optimal cut-off point. Comprehensive bioinformatic analyses were conducted to identify hub genes, key transcription factors, and enriched cancer-related pathways. Kaplan-Meier curve was used for survival analysis.
RESULTS: A 5-CpG signature score was established. Its predictive value for 5-year overall survival was high, with area under curve of 0.828, 0.835 and 0.816 for the training, testing and entire cohorts. cg27487839 and cg12885678 had strong correlation with their gene expression, XKR6 and PTPRN2, and lower PTPRN2 expression was associated with poorer survival in both The Cancer Genome Atlas and the validation datasets. Low-risk group was associated with significantly better survival. Low-risk group harboured more mutations in hub genes and key transcription factors, and mutations in SP1 and MECP2 represented favourable outcome.
CONCLUSIONS: We developed a methylation-based prognostic stratification system for endometrial cancer. Low-risk group was associated with better survival and harboured more mutations in the key regulatory genes.
摘要：
背景：很少有研究关注子宫内膜癌中的DNA甲基化。我们研究的目的是确定其在子宫内膜癌预后中的作用。
方法：从癌症基因组图谱中检索到公开可用的数据集。为了验证甲基化引起的表达改变，RNA测序数据从其他独立队列获得。MethSurv用于搜索候选CpG探针，然后通过最小绝对收缩率和选择算子Cox回归和多变量Cox回归分析进行过滤，以确定最终的CpG探针集用于总生存期。建立了基于甲基化的风险模型，并使用曲线下面积的接受者操作特征分析进行评估。使用最佳截止点将患者分为高危组和低危组。进行了全面的生物信息学分析，以确定枢纽基因，关键转录因子，和丰富的癌症相关途径。Kaplan-Meier曲线用于生存分析。
结果：建立了5-CpG签名评分。它对5年总生存率的预测价值很高，训练的曲线下面积分别为0.828、0.835和0.816，测试和整个队列。cg27487839和cg12885678与其基因表达有很强的相关性,在癌症基因组图谱和验证数据集中,XKR6和PTPRN2以及较低的PTPRN2表达与较差的存活相关。低风险组的生存率显著提高。低风险组有更多的hub基因和关键转录因子突变，SP1和MECP2的突变代表了有利的结果。
结论：我们开发了一种基于甲基化的子宫内膜癌预后分层系统。低风险组与更好的生存率相关，并且在关键调节基因中具有更多的突变。
方法：从癌症基因组图谱中检索到公开可用的数据集。为了验证甲基化引起的表达改变，RNA测序数据从其他独立队列获得。MethSurv用于搜索候选CpG探针，然后通过最小绝对收缩率和选择算子Cox回归和多变量Cox回归分析进行过滤，以确定最终的CpG探针集用于总生存期。建立了基于甲基化的风险模型，并使用曲线下面积的接受者操作特征分析进行评估。使用最佳截止点将患者分为高危组和低危组。进行了全面的生物信息学分析，以确定枢纽基因，关键转录因子，和丰富的癌症相关途径。Kaplan-Meier曲线用于生存分析。
结果：建立了5-CpG签名评分。它对5年总生存率的预测价值很高，训练的曲线下面积分别为0.828、0.835和0.816，测试和整个队列。cg27487839和cg12885678与其基因表达有很强的相关性,在癌症基因组图谱和验证数据集中,XKR6和PTPRN2以及较低的PTPRN2表达与较差的存活相关。低风险组的生存率显著提高。低风险组有更多的hub基因和关键转录因子突变，SP1和MECP2的突变代表了有利的结果。
结论：我们开发了一种基于甲基化的子宫内膜癌预后分层系统。低风险组与更好的生存率相关，并且在关键调节基因中具有更多的突变。
