COAD is one of the most frequent malignant tumors of the digestive tract, with the majority of cases stemming from colorectal epithelial cells. The role of m7G in cancer is complicated. The involvement of m7G in the formation and progression of cancer has been discovered in a growing number of studies, although there are few investigations on its significance in cancer prognosis. It has not been reported that building an m7G-related lncRNA prediction signature can predict the prognosis of COAD patients.
lncRNAs have been demonstrated to play a crucial role in COAD in numerous studies. As a result, identifying an m7G-related lncRNA prediction characteristic in COAD patients is critical.
In this study, we used univariate Cox regression analysis to obtain 37 lncRNAs associated with the prognosis of COAD patients. Then, six m7 G-related lncRNAs (LINC01063, ARRDC1-AS1, AL354993.2, ZEB1-AS1, SNHG16, LINC02474) were obtained by multivariate Cox regression analysis to construct the prognosis model. Five m7G-related lncRNAs (LINC01063, ARRDC1-AS1, AL354993.2, ZEB1-AS1, SNHG16, and LINC02474) were reported to be associated with cancer. In melanoma, LINC01063 acts as an oncogene by regulating SOX12 expression via the miR-5194 pathway[16]. STAT1 activates the long noncoding RNA ARRDC1-AS1, which has oncogenic characteristics in glioma through sponging the miR-432-5p/PRMT5 axis[17]. lncRNA ZEB1-AS1Positive Reciprocal Feedback of and α Contributes to Hypoxia-Promoted Tumorigenesis and Metastasis of Pancreatic Cancer[18]. In human gastric cancer, SNHG16 lncRNAs are overexpressed and may be carcinogenic via influencing cell cycle progression[19]. Long noncoding RNA LINC02474 affects colorectal cancer metastasis and apoptosis by inhibiting GZMB expression[20]. We also found mRNA (DXO, EIF4E, RNMT, NCBP1, RNGTT, CCNH, GTF2H1, POLR2A, POLR2D, POLR2E, POLR2J, PSMC4, METTL1, MNAT1, SUPT5H, GTF2H3, and IFIT5) significantly co-expressed with these lncRNAs. Among them, by preventing DXO destabilization of cyclin D1 mRNA, NPL4 upregulation increases bladder cancer cell proliferation[21]. Upregulation of the eukaryotic translation initiation factor 4E is linked to a poor prognosis in gallbladder cancer patients and enhances cell proliferation both in vitro and in vivo[22]. The interaction between the methyl-7-guanosine cap maturation enzyme RNMT and the cap-binding protein eIF4E has been identified and characterized[23]. Through up-regulation of CUL4B, NCBP1 promotes the progression of lung adenocarcinoma[24]. As a carcinoma inducer, cyclin H regulates lung cancer progression[25]. POLR2A promotes gastric cancer cell proliferation by advancing cell cycle progression overall[26]. SUPT5H post-transcriptional silencing modulates PIN1 expression in human breast cancer cells, inhibits tumorigenicity, and induces apoptosis[27]. Micro-RNA378a-3p induces apoptosis in sarcomatoid renal cell carcinoma and regulates POLR2A and RUNX2 expression[28]. The roles and mechanism of IFIT5 in bladder cancer epithelial-mesenchymal transition and progression[29].
We calculated the risk score for each patient using the formula. According to the median value, the patients were separated into high-risk and low-risk groups. The OS of the high-risk group was shorter than the low-risk group. The predictive signature has a strong predictive performance, according to the ROC curve. In predicting the prognosis of COAD patients, the predictive signature was more reliable than clinicopathological factors. The predictive signature has strong predictive performance, according to internal verification.
GSEA showed that compared with the low group, the high-risk group was mainly enriched in snare interactions in vesicular transport, citrate cycle TCA cycle and gluconeogenesis. The literature suggests that high PLOD3 expression in COAD leads to a poor prognosis by reducing immune cell infiltration and enhancing tumor-promoting pathways such as gluconeogenesis and TGF-beta signaling in the epithelial-mesenchymal transition (EMT)[30].
We calculated ssGSEA enrichment scores for different immune cell subgroups, immune functions, and pathways to see if there was a link between risk scores and immune cells and functions. The results showed that the T helper type 1 (Th1) cell content was significantly different between the high-risk and low-risk groups and that the immune function scores for T cell co-inhibition and T cell co-stimulation were higher in the high-risk group than in the low-risk group. These findings suggest that the immune system may be more active in the high-risk group. Axitinib, cytarabine, sorafenib, parthenolide, and vorinostat are presumably sensitive to high-risk patients, but they are resistant to lapatinib, according to our findings. This suggests that patients in the high-risk group can obtain better outcomes from immunotherapy combined with chemotherapy, which gives a theoretical foundation for COAD patients to receive tailored treatment.
However, our study has several limits. First, this study just used data from the TCGA database for internal testing; external validation of the predictive signature's applicability still requires data from other databases. Second, to confirm the risk score model's clinical utility, it must be further validated in clinical studies. Furthermore, the functions and mechanisms of the six m7G-related lncRNAs must be investigated further.
To summarize, the m7G-related lncRNA signature can independently predict COAD patients’ prognosis. In addition, it can provide the foundation for a putative mechanism of m7G-related lncRNAs in COAD and the efficacy of treatment.