It is well known that the expression of lncRNAs has crucial biological functions. Some of the DE-ir-lncRNAs detected in this study, such as TRPM2-AS, LINC01605, AC104041.1, and UCA1, have been confirmed to play significant roles in the progression of BLCA. Avgeris et al.[20] reported that the downregulation of UCA1 was correlated with a higher risk of short-term relapse in BLCA. Tian et al.[21] reported that TRPM2-AS promoted BLCA by targeting miR-22-3p and regulating the expression of GINS2. Qin et al.[22] revealed that high LINC01605 expression promoted the progression of BLCA by upregulating MMP9. Moreover, Lian et al.[23] established an 8-lncRNA signature, comprising APCDD1L-AS1, FAM225B, LINC00626, LINC00958, LOC100996694, LOC441601, LOC101928111, and ZSWIM8-AS1, as candidate prognostic markers for BLCA. Although various functions of lncRNAs have been proposed, single lncRNAs may have a bias to predict the prognosis of patients with BLCA. Previous studies[24, 25] have shown that the combinations of two genetic markers are more accurate than single genes in establishing prognostic models for cancers. To date, few studies have confirmed the prognostic value of lncRNA pairs in this setting[26–28]. In the present study, we established a prognostic risk model by pairing immune-related genes and constructed a risk model with two lncRNA pairs, without adopting their exact expression value. First, we screened the lncRNAs within the TCGA-BLCA dataset, selected the DE-lncRNAs, conducted a coexpression analysis to identify the DE-ir-lncRNAs, and validated the obtained DE-ir-lncRNA-pairs using a 0-or-1 matrix. Second, we applied a modified Lasso penalized regression model, including the procedures of the cross, multiple repetitions of validation, and random stimulations to determine DE-ir-lncRNA pairs. Third, we delineated ROC curves and calculated the AUC values to acquire the optimized model. In addition, we calculated the AIC value of each point on the AUC to detect the best cutoff value to differentiate the high- and low-risk groups among patients with BLCA. Finally, we assessed the relationship between this novel risk model and different clinical parameters.
Preclinical studies have confirmed that increased infiltration of CD4+ or CD8+ immune cells[29–31] leads to a better response to ICIs. In the present study, we used various online tools, including CIBERSORT, XCELL, CIBERSORT-ABS, QUANTISEQ, MCPcounter, EPIC, and TIMER, to estimate the tumor-infiltrating cells in patients with BLCA, and analyzed their association with the predicted risk scores. Our results showed that CD4+ T-cells, monocytes, macrophages, cancer-associated fibroblasts, and myeloid dendritic cells were enriched in the high-risk group, which may explain why the high-risk group was related to poor prognosis. In addition, correlation analysis demonstrated that the high-risk group was positively correlated with the expression of some immune microenvironmental inhibitory genes, such as HAVCR2, DDR2, and a positive correlation trend with LAG3.
LINC00665 and some other lncRNAs have been shown to enhance the efficacy of immunotherapy in BLCA[32–34]. In addition, Zhang et al.[35] found that the lncRNA HOTAIR can inhibit 5-fluorouracil sensitivity by mediating MTHFR methylation, and Gu et al. reported that NONHSAT141924 was associated with paclitaxel chemotherapy resistance[36]. Overall, these findings demonstrate that lncRNAs may be related to chemotherapy resistance. Based on this, herein, we explored the relationship between the identified risk group and chemotherapy. Our risk model suggested that the high-risk group was more sensitive to methotrexate and metformin, whereas the low-risk group was more sensitive to cisplatin, docetaxel, and paclitaxel, which was consistent with previous studies.
There were some limitations to our study. First, the raw data obtained from the TCGA database were relatively insufficient for an initial analysis. Second, external validation was necessary to verify the efficiency of the risk model herein established. To overcome these limitations, we screened lncRNA pairs using a 0-or-1 matrix, which was optimal in this study. Further studies comprising more clinical samples are underway for further verification of the proposed model. In summary, we defined a novel risk predictive model comprising ir-lncRNAs that does not requires the exact expression of the lncRNAs, which may help clinicians identify patients who can benefit from immunotherapy.