Co-expressed functional module-related genes in ovarian cancer stem cells represent novel prognostic biomarkers in ovarian cancer

Ovarian cancer is the leading cause of death from gynecologic malignancies. Cancer stem cells (CSC) seem to play a crucial role in tumor metastasis, recurrence, and chemoresistance. Therefore, CSCs offer significant potential for developing therapeutic targets and to understand tumor recurrence and chemoresistance mechanisms. In the present study, our aim was the identification of the gene group in ovarian CSCs (O-CSCs) and the potential of the resultant gene group in ovarian cancer prognosis. Two different microarray data sets were analyzed by comparing gene expression levels between O-CSCs and cancer samples. The O-CSC co-expression network was reconstructed and its modules were identified. According to the analysis results, 74 mutual DEGs were identified. The O-CSC-specific co-expression network included 32 nodes and 95 edges (network density: 19%), while the co-expression network in cancer samples was reconstructed with 74 nodes and 1066 edges (network density: 39%). Understanding of the molecular mechanism and signatures of O-CSCs should provide valuable insight into chemotherapy resistance and recurrence of ovarian tumors. A highly connected 12 gene module in O-CSC samples of BAMB1, NFKB12, EZR, TNFAIP3, C1orf86, PMAIP1, GEM, KHDRBS3, FILIP1, FGFR2, TGFBR3 and PEG10, (network density: 67%) was identified. Prognostic performance of these genes was evaluated independently using six ovarian cancer datasets (n = 1933 patient samples) via survival analysis. These co-expressed genes were determined as prognostic targets in ovarian cancer. Through literature search validation, five genes (C1orf86, PMAIP1, FILIP1, NFKB12 and PEG10) suggested as novel molecular targets in ovarian cancer. The presented prognostic biomarkers here provide a resource for the understanding of tumor recurrence and chemoresistance and may facilitate critical research directions and development of new prognostic and therapeutic strategies for ovarian cancer. CSCs: cancer stem cells; O-CSCs: ovarian CSCs; FACS: fluorescence-activated cell sorting; SP: side population; MP: main population; TFs: transcription factors.

卵巢癌是妇科恶性肿瘤致死的首要病因。癌干细胞（cancer stem cells, CSC）在肿瘤转移、复发及化疗耐药过程中发挥关键作用。因此，癌干细胞可为开发肿瘤治疗靶点、解析肿瘤复发与化疗耐药机制提供重要研究价值。本研究旨在鉴定卵巢癌干细胞（ovarian CSCs, O-CSCs）相关基因集，并探究该基因集在卵巢癌预后评估中的应用潜力。本研究通过比较卵巢癌干细胞与癌组织样本的基因表达水平，对两套独立的微阵列数据集进行了分析。研究人员重构了卵巢癌干细胞的共表达网络，并对其功能模块进行了鉴定。分析结果显示，共筛选得到74个共有差异表达基因（differentially expressed genes, DEGs）。卵巢癌干细胞特异性共表达网络包含32个节点与95条连接边（网络密度：19%），而癌组织样本的共表达网络则包含74个节点与1066条连接边（网络密度：39%）。解析卵巢癌干细胞的分子机制与特征，可为理解卵巢肿瘤的化疗耐药与复发机制提供重要参考。本研究在卵巢癌干细胞样本中鉴定出一个高度连通的12基因模块，其基因成员包括BAMB1、NFKB12、EZR、TNFAIP3、C1orf86、PMAIP1、GEM、KHDRBS3、FILIP1、FGFR2、TGFBR3及PEG10，该模块的网络密度为67%。研究人员利用6套卵巢癌数据集（共包含1933例患者样本），通过生存分析独立评估了这些基因的预后性能。最终确定这些共表达基因为卵巢癌的潜在预后靶点。经文献检索验证，其中5个基因（C1orf86、PMAIP1、FILIP1、NFKB12及PEG10）被提示为卵巢癌新型分子靶点。本研究提出的预后生物标志物可为理解肿瘤复发与化疗耐药机制提供研究资源，并有望为卵巢癌新型预后策略与治疗方案的开发指明关键研究方向。CSC：癌干细胞（cancer stem cells）；O-CSCs：卵巢癌干细胞（ovarian CSCs）；FACS：荧光激活细胞分选术（fluorescence-activated cell sorting）；SP：侧群（side population）；MP：主群（main population）；TFs：转录因子（transcription factors）。