Stratification of HPV-associated and HPV-negative oropharyngeal squamous cell carcinomas based on DNA methylation epigenotypes

Stratification of HPV-associated and HPV-negative oropharyngeal squamous cell carcinomas based on DNA methylation epigenotypes
复制标题

DOI:
10.1002/ijc.32890
复制
发表时间:
2020-02-11
影响因子:
6.4
通讯作者:
Kaneda, Atsushi
Kaneda, Atsushi
中科院分区:
医学1区
文献类型:
--
作者:
Nakagawa, Takuya;Matsusaka, Keisuke;Kaneda, Atsushi

文献摘要

被引文献

相似文献

近二十年来,口咽鳞状细胞癌(OPSCC)的发病率一直在上升,主要是由于人类乳头瘤病毒(HPV),但OPSCC的分子亚群表现出不同的临床病理特征还没有得到充分的研究。我们使用Infinium450k对170例OPSCC进行了DNA甲基组分析,其中包括我们队列中的89例和癌症基因组图谱报道的81例,并进行了靶向外显子测序分析。我们利用甲基组数据进行系统聚类分析,对OPSCC进行分层。用焦磷酸测序定量验证分类标记的甲基化水平,并计算受试者工作特征(ROC)曲线的曲线下面积(AUC)值。OPSCC可分为四种表型:HPV(+)高甲基化(OP1)、HPV(+)中甲基化(OP2)、HPV(-)中甲基化(OP3)和HPV(-)低甲基化(OP4)。产生了10个甲基化标记基因:5个将HPV(+)病例分类为OP1和OP2,5个将HPV(-)病例分类为OP3和OP4。两个标记板的ROC曲线AUC值分别为0.969和0.952。虽然HPV(-)组的TP53突变和CCND1拷贝数增量显著高于HPV(+)组(P<0.01),但OP1和OP2、OP3和OP4之间的基因组畸变率没有显著差异。四种表型的预后有显著差异(P=0.0006),在一般有利的HPV(+)病例中区分最有利的OPSCC亚群(OP1),在一般不有利的HPV(-)病例中区分最不利的OPSCC亚群(OP3)。HPV(+)和HPV(-)OPSCC进一步分为不同的DNA甲基化表型,预后有显著差异。
While the incidence of oropharyngeal squamous cell carcinoma (OPSCC) has been increasing in these two decades, primarily due to human papillomavirus (HPV), stratification of OPSCC into molecular subgroups showing different clinicopathological features has not been fully investigated. We performed DNA methylome analysis using Infinium 450k for 170 OPSCC cases, including 89 cases in our cohort and 81 cases reported by The Cancer Genome Atlas, together with targeted exon sequencing analysis. We stratified OPSCC by hierarchical clustering analysis using methylome data. Methylation levels of classifier markers were validated quantitatively using pyrosequencing, and area under the curve (AUC) values of receiver operating characteristics (ROC) curves were calculated. OPSCC was stratified into four epigenotypes: HPV(+) high-methylation (OP1), HPV(+) intermediate-methylation (OP2), HPV(-) intermediate-methylation (OP3) and HPV(-) low-methylation (OP4). Ten methylation marker genes were generated: five to classify HPV(+) cases into OP1 and OP2, and five to classify HPV(-) cases into OP3 and OP4. AUC values of ROC curves were 0.969 and 0.952 for the two marker panels, respectively. While significantly higher TP53 mutation and CCND1 copy number gains were observed in HPV(-) than in HPV(+) groups (p < 0.01), no significant difference of genomic aberrations was observed between OP1 and OP2, or OP3 and OP4. The four epigenotypes showed significantly different prognosis (p = 0.0006), distinguishing the most favorable OPSCC subgroup (OP1) among generally favorable HPV(+) cases, and the most unfavorable OPSCC subgroup (OP3) among generally unfavorable HPV(-) cases. HPV(+) and HPV(-) OPSCC are further divided into distinct DNA methylation epigenotypes, showing significantly different prognosis.