...
  • 主题
高级搜索  >
历史搜索 清空历史
首页> 外文期刊>Clinical and Translational Medicine >Integrated multi‐omics data reveals the molecular subtypes and guides the androgen receptor signalling inhibitor treatment of prostate cancer
【24h】

Integrated multi‐omics data reveals the molecular subtypes and guides the androgen receptor signalling inhibitor treatment of prostate cancer

机译：集成的多OMICS数据显示了分子亚型并引导雄激素受体信号传导抑制剂治疗前列腺癌

获取原文
           
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

Dear editor, Prostate cancer (PCa) is the most frequent malignant tumour in males, ~(1) it is essential to precisely identify the specific molecular features and judge potential clinical outcomes from the multi‐omics aspect. Recently, we developed an R package “ MOVICS ” ( https://xlucpu.github.io/MOVICS/MOVICS‐VIGNETTE.html ) for multi‐omics integration and clustering, aim to stratify tumour molecular subtypes and facilitate precision medicine. ~(2) In the current study, we firstly proposed the PCa multi‐omics classification (PMOC) system derived from mRNA, microRNA, long noncoding RNA, DNA methylation, and somatic mutation, using 10 leading‐edge clustering algorithms. We collected a total of 1192 PCa patients from five independent cohorts and an external AHMU‐PC cohort from our own institute ~(3) (Tables S1 and S2 ). The technical details are listed in Supporting Information. We identified three clusters independently from ten multi‐omics integrative clustering algorithms (Figure? 1A ) referring to the clustering prediction index, Gaps‐statistics analysis (Figure S1 ) and predefined PAM50 system, ~(4) and further combined the clustering results via a consensus ensemble approach (Figure? 1B ). Multi‐omics data in PMOCs was visualized in Figure? 1C . Significantly, diverse clinical recurrence‐free survival (RFS) outcomes were observed (all p ?&?.001, Figure? 1D ). Most PMOC2 patients had a higher Gleason score than PMOC1 and PMOC3 (61.8%?vs. 23.0%?vs. 9.7%, p ?=?.012), as well as the proportion of advanced pathology T stage (86.8%?vs. 54.9%?vs. 52.6%, p ?&?0.001, Table S3 ). The top 100 subtype‐specific markers for each PMOC were selected for the reproduction of PMOCs in external validation cohorts (Table S4 ). FIGURE 1 Recognition of the prostate cancer multi‐omics classification (PMOC) system in the TCGA‐PRAD cohort. (A) Clustering of prostate cancer (PCa) patients via 10 leading‐edge clustering methods. (B) Consensus matrix for three clusters based on the 10 algorithms. (C)?Visualization of multi‐omics data for 1526 mRNAs, 242 lncRNAs, 30 miRNAs, 1073 DNA CpG methylation sites and 23 mutant genes. (D)?Differential recurrence‐free survival outcome in three PMOCs, log‐rank test We observed the significant activation of the G2M checkpoint, E2F target pathways in PMOC2 (Figure? 2A ). Specifically, the decreased phosphorylated protein levels of p‐CHK1 and p‐CHK2 in PMOC2 may weaken the inhibitory function of CDC25 components and increase CDK1 activation (Figure? 2B ). For PMOC1, we observed the activation of TNF‐α signalling, IL6/JAK/STAT3 and IL2/STAT5 signalling, which were immune response relevant. PMOC3 presented activation of both immune‐associated and oncogenic pathways; the phosphorylated mTOR and mTOR levels were activated and could further promote cell growth (Figure? 2C ). FIGURE 2 Differential activity of tumour‐associated pathways across three prostate cancer multi‐omics classifications (PMOCs). (A)?Heatmap of 50 differentially activated HALLMARK pathways. (B) G2M pathways activated in the PMOC2 at both the mRNA and protein levels. (C) PI3K/AKT pathways activated in the PMOC3 at both the mRNA and protein levels. (D) Heatmap of subtype‐specific metabolism signalling pathways. (E) Differential infiltration of 13 immunocytes among three subtypes, Kruskal‐Wallis test. (F) Expression patterns of four immune checkpoints across three PCa subtypes, Kruskal‐Wallis test. ns, not significant; * p ?&?.05; ** p ?&?.01; *** p ?&?.001; **** p ?&?.0001. We further compared the activation status of metabolic pathways. The nicotinamide adenine dinucleotide biosynthesis and cyclooxygenase arachidonic acid metabolism pathways were activated in PMOC1, which were reported to be associated with tumour inflammation and immune‐metabolic circuits. ~(5) In PMOC2, we observed the activated pyrimidine metabolism and biosynthesis, biotin metabolism, and oxidative phosphorylation. Glycogen metabolism and amino acid metabolism‐associated pathways were highly activated in PMOC3 (Figure? 2D ). PMOC3 had higher infiltration of immune‐suppressed components, while PMOC1 tended to exhibit immune activated components, and higher expression of PD1, PDL1 and CTLA4 (Figure? 2E,F ), and was also associated with immune activated molecular subtype ~(3) (Figure S2A ). Genetic alteration contributed dramatically to shaping the subtypes. Specifically, the total tumour mutant burden was highest in PMOC2 ( p ?&?.001, Figure S3A ). PMOC2 contained most patients with TP53 mutation (PMOC2: 23.6%, PMOC1: 8.6%, PMOC3: 5.8%, p ?&?.001), and SPOP (PMOC2: 18.7%, PMOC1: 9.6%, PMOC3: 7.8%, p ?=?.0138, Figure S3B , Table S5 ). The tumour suppressor APC protein is an antagonist of the Wnt signalling pathway. ~(6) Mutant APC resulted in lower expression ( p ?=?.0026, Figure S3C ) and led to an unfavourable clinical outcome ( p ?=?.013, Figure S3D ). Both the lost and gained copy numbers were significant
机译：亲爱的编辑，前列腺癌（PCA）是雄性中最常见的恶性肿瘤，〜（1）必须精确鉴定特定的分子特征和判断来自多OMICS方面的潜在临床结果。最近，我们开发了一个R包“Movics”（https://xlucpu.github.io/movics/movics -vignette.html），用于多OMICS集成和聚类，旨在分层肿瘤分子亚型并促进精密药物。 〜（2）在目前的研究中，首先提出了使用10个前沿聚类算法的MRNA，MicroRNA，长度非编码RNA，DNA甲基化和体细胞突变的PCA多OMICS分类（PMOC）系统。我们共收集了来自五个独立队列的1192名PCA患者，并从我们自己的研究所（3）（表S1和S2）的外部Ahmu-PC队列。技术详细信息列在支持信息中。我们独立于十个多OMICS综合聚类算法（图1A）识别了三个集群，参考群集预测索引，间隙统计分析（图S1）和预定义的PAM50系统，〜（4）并通过A进一步组合聚类结果共识融合方法（图？1B）。 PMOC中的多OMICS数据在图中可视化？ 1C。显着地，观察到不同的临床复发存活（RFS）结果（所有p？＆amp; Δ001，图1d）。大多数PMOC2患者的Gleason分数高于PMoc1和PMoc3（61.8％？vs。23.0％？vs.9.7％，p？= 012），以及晚期病理T阶段的比例（86.8％？VS。 54.9％？vs。52.6％，p？＆amp; ？0.001，表S3）。选择每个PMOC的前100个亚型特异性标记用于在外部验证队列中的PMOC中的再现（表S4）。图1识别TCGA-PRAD队列中的前列腺癌多OMICS分类（PMOC）系统。 （a）通过10个前缘聚类方法聚类前列腺癌（PCA）患者。 （b）基于10算法的三集群共识矩阵。 （c）？为1526 mRNA，242克朗，30 miRNA，1073dNA CpG甲基化位点和23个突变基因的多OMICS数据的可视化。 （d）？在三个PMOC中的差异复发存活结果，对数级试验我们观察到PMOC2中的G2M检查点，E2F靶途径的显着激活（图？2A）。具体地，P-CHK1和P-CHK2中的降低的磷酸化蛋白水平可能削弱CDC25组分的抑制作用，并增加CDK1激活（图β2B）。对于PMOC1，我们观察了TNF-α信令，IL6 / JAK / Stat3和IL2 / Stat5信号传导的激活，其免疫应答相关。 PMOC3呈现免疫相关和致癌途径的激活;激活磷酸化的MTOR和MTOR水平，并进一步促进细胞生长（图2℃）。图2肿瘤相关途径跨越三个前列腺癌多OMICS分类（PMOC）的差异活性。 （a）？50个差异激活的标志途径的热线图。 （b）在mRNA和蛋白质水平的PMOC2中激活的G2M途径。 （c）在mRNA和蛋白质水平的PMOC3中激活的PI3K / AKT途径。 （d）亚型特异性代谢信号通路的热法。 （e）三个亚型，Kruskal-Wallis测试中13个免疫细胞的差异渗透。 （f）四种PCA亚型的四种免疫检查点的表达模式，Kruskal-Wallis试验。 ns，不重要; * p？＆amp; ？。05; ** p？＆amp; ？。01; *** p？＆amp; 001; **** p？＆amp; 0001。我们进一步比较了代谢途径的激活状态。在PMOC1中激活烟酰胺腺嘌呤二核苷酸生物合成和环氧氧酶活性族酸代谢途径，据报道，据报道与肿瘤炎症和免疫代谢电路相关。 〜（5）在PMOC2中，我们观察到活化的嘧啶代谢和生物合成，生物素代谢和氧化磷酸化。在PMOC3（图2D）中高度活化糖原代谢和氨基酸代谢相关途径。 PMOC3具有更高的免疫抑制成分浸润，而PMoc1倾向于表现出免疫活化的组分，并且PD1，PDL1和CTLA4的更高表达（图？2E，F），并且也与免疫活化的分子亚型〜（3）相关（图S2A）。遗传改变贡献造成塑造亚型。具体地，PMOC 2（P→＆amp;×。001，图S3A）中总肿瘤突变体重最高。 PMOC2含有大多数TP53突变的患者（PMOC2：23.6％，PMOC1：8.6％，PMOC 3：5.8％，P？＆amp; LT;？。001）和血管（PMOC2：18.7％，PMOC1：9.6％，PMOC3：7.8 ％，p？= ?. 0138，图S3B，表S5）。肿瘤抑制剂APC蛋白是WNT信号传导途径的拮抗剂。 〜（6）突变体APC导致较低的表达（p？= 0026，图S3c）并导致不利的临床结果（p？= 013，图S3D）。丢失和获得的副本数字都很重要

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号