前列腺癌
转录组
生物信息学
亚型
医学
计算生物学
生物
生物信息学
肿瘤科
癌症
内科学
基因
基因表达
遗传学
计算机科学
程序设计语言
作者
Yingchun Liang,Enlin Rong,Jin Qian,Chenkai Ma,Jimeng Hu
标识
DOI:10.1038/s41391-022-00495-9
摘要
Men with metastatic prostate cancer who are treated with androgen deprivation therapy (ADT) typically develop therapeutic resistance eventually and have a dismal outcome. Moreover, the limited survival benefit observed in only a proportion of patients receiving novel therapeutics including immune checkpoint blockade and PARP inhibitors suggests the biological heterogeneity of mCRPCs.To understand the heterogeneity of mCRPC, we analyzed the transcriptome of 231 mCRPCs and identified four disparate biological subtypes (Basal, Homologous Recombination Repair (HRR), Neuroendocrine and Luminal). The package "randomForest" was used to construct a Random Forest model. Circular plots were used for visualization of TMPRSS2-ERG fusion in each subtype.The Luminal subtype of mCRPCs has higher Androgen Receptor (AR) expression and copy number alterations as compared with the other subtypes. Genes in HRR pathway are relatively downregulated in most subtypes regardless of the genetic alterations, except for the HRR and NE subtypes, suggesting potential resistance of the HRR and NE mCRPCs to PARP inhibitor treatment. The HRR subtype has relatively more immune cell infiltration and higher expression of immune checkpoints, highlighting that the efficacy of immunotherapy should be evaluated in this particular subtype. TMPRSS2-ERG fusion is the most frequent gene fusion in all mCRPCs, and the Basal subtype has a higher frequency of this fusion than the other subtypes.Our results reveal that the stratification of mCRPC according to transcriptome is informative of personalized therapeutics in the treatment of mCRPCs. The predictive capacity of the transcriptome subtyping of mCRPC warrants further exploration in the future.
科研通智能强力驱动
Strongly Powered by AbleSci AI