Although pieces of evidence have indicated that lncRNAs play regulatory roles in oncogenesis and tumor progression of various cancers, the potential mechanisms of how lncRNAs exert their regulatory roles in PC remain undefined [25]. Recently, a great deal of research has suggested that lncRNAs can interact with miRNAs and further regulate downstream mRNA expression [7]. Notably, lnRNAs, miRNAs, and mRNAs are inclined to function as a unit, not simply one to one interactions in PC. For instance, Hai Gao et al. reported that lncRNA ZEB2-AS1 was involved in encouraging tumor growth and invasion of PC through the ZEB2-AS1/miR-204/HMGB1 ceRNA network [26]; Xiong et al. indicated that GSTM3TV2 functioned as a ceRNA and negatively regulated let-7 expression, thereby leading to the PC progression and chemo-resistance through upsetting the expression of LAT2 and OLR1 [27]; Li et al. demonstrated that NORAD may function as a ceRNA to regulate the expression of the small GTP binding protein Rho A through competition for hsa-miR-125a-3p, thereby promoting EMT [28]. More importantly, it is well known that cancer-related pathways play significant roles in the progression of PC, and genes with similar functions cluster together to form a regulatory pathway [3]. However, it is unclear about the relationship between the ceRNA network and cancer-related pathways. Therefore, it's necessary to elucidate the role of pathway-related ceRNA network in tumor origin and progress.
In the present study, we successfully identified a novel pathway-related PVT1/miR-20b/CCND1 ceRNA network involved in PC progression through multi-omics analysis. Firstly, we divided eight GEO datasets into three groups according to their platform, and combined TCGA and GTEx databases as a new group. Next, we screened out DEGs and carried out KEGG enrichment analysis in each group, and looked for all the genes in the most enriched pathway. Then, those DEGs in KEGG were visualized in the PPI network and identified as hub genes according to their node degrees calculated by cytoHubba tool. The expression role and survival value of hub genes were validated using GEPIA and KM plotter databases separately. Eleven qualified genes met the criteria of expression validation and survival analyses. Notably, their oncogenic roles were also detected in PC progression. For example, there was an 8.1% actionable alterations of CCND1 in PC, which means that treatments for CCND1 could be implemented to diminish its impact on the cell cycle [29, 30]. The MET/HGF axis is involved in the complex crosstalk between tumor and stroma, especially in the interaction between cancer cells and activated PSCs, thereby favors the progression and metastasis of pancreatic cancer [31]. MMP-9 promoted disease progression to PDAC, while the deficiency of MMP-9 resulted in more invasive tumors and an increase in the desmoplastic stroma [32].
Additionally, the upstream miRNAs of hub genes were predicted and validated by relative databases motioned above. four qualified miRNAs were screened out as key miRNAs. Similarly, some of the miRNAs were also involved in the development of PC. For instance, Sandhu et al. demonstrate that MiR-139-5p/SLC7A11 served as a tumor suppressor was able to inhibit invasion and metastasis of PC by direct-acting with PI3K/Akt Signaling Pathway [33]. Sureban et al. proved that miR-144 could mediate the low expression of Notch-1, which interacted with DCAMKL-1 to regulates the epithelial-mesenchymal transition of PC [34]. Zhao et al. illuminated that exosome-encapsulated miR-451a functioned as a high-sensitive liquid biomarker, which could predict the overall survival of high-risk PC patients [35]. Moreover, the upstream lncRNAs of key miRNAs were identified. But only two lncRNAs conformed to expression and prognostic standards. LncRNAs PVT1 was reported to facilitate PC proliferation and migration via sponge with miR-448, thereby regulating the expression of SERBP1 [36]. Also, You et al. demonstrate that gemcitabine therapy inhibited PVT1 expression but promote its encoded miRNAs miR-1207 level, and over-expression of miR-1207 enhanced the chemo-sensitivity of PC cells to gemcitabine [37]. The cancer-related pathway is an important biological process, which not only carries a variety of biological functions but also is closely related to the development and occurrence of many disease processes, especially in cancer. In this work, we established and validated a pathway-related PVT1/miR-20b/CCND1 ceRNA network in the progression of PC, which perfectly satisfying all the conditions of the ceRNA hypothesis.
Although a few studies have already assessed the role of ceRNA in the progression of PC [38, 39], few studies focus on the pathway-related ceRNA axis using a multi-omics analysis. Also, to our knowledge, it is the first study of PC that constructed a pathway-related ceRNA network in the progression of PC. Inevitably, there are some limitations that can be found in our study. First, we didn’t stratify the samples based on their clinical characteristics, such as sex. A recent study has indicated that incorporating sex as a biological variable is rewarding to better know cancer mechanisms [40]. Second, we verify the expression and prognostic value using online databases rather than the date from clinical samples, which will undermine our work’s credibility. That’s the reason why we constructed the ceRNA axis through comprehensive analysis and validated the data with the same conditions. Last, we identified a ceRNA network mainly based on their expression and prognostic value, which might overlook some valuable information.