转录组
基因签名
癌变
肿瘤科
生物
胰腺导管腺癌
内科学
基因表达谱
基因
腺癌
生存分析
癌症研究
胰腺癌
医学
基因表达
癌症
遗传学
作者
Sophia G. Kisling,Pranita Atri,Ashu Shah,Jesse L. Cox,Sunandini Sharma,Lynette M. Smith,Dario Ghersi,Surinder K. Batra
标识
DOI:10.1158/1078-0432.ccr-23-0825
摘要
Abstract Purpose: Despite the significant association of molecular subtypes with poor prognosis in patients with pancreatic ductal adenocarcinoma (PDAC), few efforts have been made to identify the underlying pathway(s) responsible for this prognosis. Identifying a clinically relevant prognosis-based gene signature may be the key to improving patient outcomes. Experimental Design: We analyzed the transcriptomic profiles of treatment-naïve surgically resected short-term survivor (STS) and long-term survivor (LTS) tumors (GSE62452) for expression and survival, followed by validation in several datasets. These results were corroborated by IHC analysis of PDAC-resected STS and LTS tumors. The mechanism of this differential survival was investigated using CIBERSORT and pathway analyses. Results: We identified a short-surviving prognostic subtype of PDAC with a high degree of significance (P = 0.018). One hundred thirty genes in this novel subtype were found to be regulated by a master regulator, homeobox gene HOXA10, and a 5-gene signature derived from these genes, including BANF1, EIF4G1, MRPS10, PDIA4, and TYMS, exhibited differential expression in STSs and a strong association with poor survival. This signature was further associated with the proportion of T cells and macrophages found in STSs and LTSs, demonstrating a potential role in PDAC immunosuppression. Pathway analyses corroborated these findings, revealing that this HOXA10-driven prognostic signature is associated with immune suppression and enhanced tumorigenesis. Conclusions: Overall, these findings reveal the presence of a HOXA10-associated prognostic subtype that can be used to differentiate between STS and LTS patients of PDAC and inform on the molecular interactions that play a role in this poor prognosis.
科研通智能强力驱动
Strongly Powered by AbleSci AI