Skip to main content
NCBI home page
Journal of Cancer Research and Clinical Oncology logoLink to Journal of Cancer Research and Clinical Oncology
. 2023 Jun 24;149(13):11207–11221. doi: 10.1007/s00432-023-05013-5

Construction and validation of a cuproptosis-related lncRNA prognosis signature in bladder carcinoma

Jinbo Song 1,✉, Xiaoke Sun 1, Ting Wang 1, Li Yan 1, Pengxiao Su 1, Leihong Yuan 1
PMCID: PMC11797366  PMID: 37354222

Abstract

Background

Bladder cancer (BLCA) is a prevalent urological tumor with high morbidity and mortality. However, BLCA treatment remains challenging due to a lack of effective biomarkers. Long non-coding RNAs (lncRNAs), as active participants in tumor progression are involved in multiple biological regulatory mechanisms, and cuproptosis-related genes participate in the development of cancer. It is important to discover cuproptosis- related lncRNAs for BLCA diagnosis and treatment.

Methods

A predictive signature was constructed based on least absolute shrinkage and selection operator regression (LASSO) and Cox regression analyses of the 9 cuproptosis-related lncRNAs. Samples were divided into high-risk group and low-risk group based on their median risk scores to explore their prognosis.

Results

This signature is well predictive, as evidenced by the receiver operating characteristic curves (ROC curves) and K-M curves. Based on the nomogram, we were able to visually forecast the survival rates of patients with BLCA at 1-, 3-, and 5-year, and the calibration plots displayed that the actual results were well matched with the predicted 1-, 3-, and 5-year survival rates. Furthermore, BLCA patients in the high-risk group had a higher Tumor Immune Dysfunction and Exclusion (TIDE) score and lower TMB. Finally, we investigated the response of antitumor drugs for BLCA patients in different risk groups, and a statistically significant difference was observed in the sensitivity of those drugs between low- and the high-risk groups.

Conclusion

According to the 9 cuproptosis-related lncRNAs, we constructed a signature which can be served as a promising prognostic biomarker for BLCA patients.

Keywords: Bladder cancer, Long non-coding RNA, Cuproptosis, Prognostic, Targeted therapy, Biomarkers

Introduction

Bladder cancer (BLCA) is a prevalent urological tumor with high morbidity and mortality (Siegel et al. 2022). Clinically, according to the extent of tumor invasion, it can be classified into muscle-invasive bladder cancer (MIBC) and non-muscle invasive bladder cancer (NMIBC) (Thompson 2006). NMIBC compose around 75% of newly diagnosed BLCA (Burger et al. 2013).Cystoscopy and its biopsy were currently suggested as the main method for the diagnosis of BLCA (Babjuk et al. 2017; Ritch et al. 2019). Once diagnosed, Transurethral resection or radical cystectomy is the most popular therapy if contraindications have been excluded (Xie et al. 2021). However, the recurrence rate of primary BLCA is up to 50% even after surgery, the prognosis is far from satisfactory (Roupret et al. 2021; Cambier et al. 2016). Given the high morbidity and mortality of BLCA, it is imperative to further explore biomarkers that can predict the effectiveness of BLCA immunotherapy or targeted therapy, discover new prognosis related biomarkers, and identify new therapeutic targets to prolong the survival time of patients with BLCA.

Transition metals such as Fe, Cu, Zn and Mn are essential cytokines in various biological processes (Zygiel and Nolan 2018; Festa and Thiele 2011; Bafaro et al. 2017). Copper (Cu), an indispensable trace metal element, is essential to cell growth and development, aerobic respiration, and oxidative phosphorylation (Robinson and Winge 2010; Babak and Ahn 2021). However, either insufficient or excessive Cu can cause irreversible damage and trigger cell death. According to recent studies, the copper levels in patients’ serum samples are significantly elevated in prostate cancer (Saleh et al. 2020), breast cancer (Feng et al. 2020), and bladder cancer. (Mortada et al. 2020). “Cuproptosis” is a novel form of cell death published in the journal Science by Tsvetkov and colleagues (Tsvetkov et al. 2022). Unlike apoptosis, ferroptosis, or necroptosis, copper directly binds to lipoylated proteins of the tricarboxylic acid (TCA) cycle to trigger programmed cell death. Treatment with mitochondrial antioxidants, fatty acids, and inhibitors of mitochondrial function had a very distinct effect on the sensitivity to copper ionophores as compared with sensitivity to the ferroptosis-inducing GPX4 inhibitor ML162 (Tsvetkov et al. 2022). Several key genes (FDX1, DLAT, ATP7A, ATP7B, PDHB and so on) were confirmed to be closely related to protein lipoylation, which patients with BLCA may benefit from by providing novel strategies for predicting their prognosis.

According to the length, shape and location, non-coding RNAs (ncRNAs) have been divided into different classes. Among them, microRNA (miRNA), long ncRNA (lncRNA), circular RNA (circRNA) and PIWI interacting RNA (piRNA) are the four major ncRNA types with distinct functions in cancers (Yan and Bu 2021). Recent studies have begun to reveal the crucial roles of lncRNAs in cellular processes (differentiation, development and tumorigenesis) (Toden et al. 2021). Notably, recent studies have revealed that a subset of lncRNAs act as guides or scaffolds for epigenetic modifiers and recruit them to specific genomic loci. Thus, aberrant expression of such lncRNAs could contribute to tumorigenesis by inducing aberrant epigenetic gene regulation (Saw et al. 2021). Long non-coding RNAs (lncRNAs) are characterized by a lack of complete open reading frames (ORFs) and more than 200 nucleotides (Quinn and Chang 2016). With the deepening of research, an increasing number of lncRNAs have been proved to be involved in biological regulatory mechanisms, which play vital roles in tumor occurrence and development (Gao et al. 2020). In bladder cancer, the lncRNA RP11-89 has been shown to promote cell multiplication by the miR-129-5p/PROM2 axis to facilitate tumorigenesis (Luo et al. 2021). Furthermore, one recent study demonstrated that lncRNA HAGLROS can regulate the miR-330-5p/SPRR1B axis to accelerate the malignant progression of BLCA (Xiao et al. 2022). Our previous studies have shown that some Cuproptosis-Related LncRNAs were prognostic factors in many cancers including clear cell renal cell Carcinoma (Bian et al. 2022) and acute myeloid leukemia (Li et al. 2022). However, the role of Cuproptosis-Related LncRNAs in BLCA has not fully understood and merit further work.

In this study, we analyzed BLCA genes and screened 9 cuproptosis-related IncRNAs. Then, cuproptosis-related IncRNAs were used to construct a prognostic signature with strong prediction accuracy. The identification of 9 cuproptosis-related IncRNAs could serve as diagnostic and prognostic biomarkers.

Materials and methods

Data acquisition

The TCGA database was searched for gene expression data and relevant clinical data (normal and tumor) of BLCA, including 406 samples with BLCA and 19 samples with adjacent normal tissues. To obtain cuproptosis-related lncRNAs, the “limma” package was used for analyzing the relationship between cuproptosis-related genes and lncRNAs. The correlation coefficients of |R2|> 0.4 and p < 0.001 were set as the screening standards.

Enrichment functional analysis

Using the “ClusterProfiler” package, the biological functions of cuproptosis-related lncRNAs were examined according to Gene Ontology (GO) (Gene Ontology Consortium 2015) and Kyoto Encyclopedia of Genes and Genomes (KEGG) (Kanehisa and Goto 2000). The screening conditions were FDR < 0.05 and p < 0.05. Then, significant biological pathways and functions were outlined, which may be involved in the regulation of cuproptosis-related lncRNAs.

Construction of the prognostic signature

Previous studies have identified several cuproptosis-related genes (Tsvetkov et al. 2022; Polishchuk et al. 2019; Aubert et al. 2020; Dong et al. 2021; Ren et al. 2021). A total of 406 BLCA specimens were randomly divided into training and testing groups. The clinical characteristics of the two groups did not differ statistically (Table 1). Then, a cuproptosis-related IncRNA prognostic signature was constructed using the training dataset. Univariate Cox regression analysis was performed to identify co-expressed cuproptosis-related IncRNAs. Next, a least absolute shrinkage and selection operator (LASSO) regression analysis was applied to screen lncRNAs and construct the prognostic signature using 1000 times tenfold cross-validation by the “glmnet” package. Then, we obtained 9 prognostic cuproptosis-related lncRNAs by multivariate Cox regression analysis. The risk score profile for BLCA was calculated using the corresponding coefficients of the 9 lncRNAs corresponding coefficients. For each sample, the following formula was used for calculating the risk score. Risk score = (coef lncRNA1 × exp lncRNA1) + (coef lncRNA2 × exp lncRNA2) + ∙∙∙ + (coef lncRNAn × exp lncRNAn).

Table 1.

Clinical characteristics of 406 patients of bladder cancer

Character Entire dataset Training dataset Testing dataset P OS (days) P
n = 406 n = 203 n = 203
Age (%) 0.363 0.320
 ≤ 65 160 (39.41%) 75 (36.95%) 85 (41.87%) 695.28
 > 65 246 (60.59%) 128 (63.05%) 118 (58.13%) 873.24
Gender (%) 0.910 0.379
 Female 106 (26.11%) 54 (26.60%) 52 (25.62%) 734.58
 Male 300 (73.89%) 149 (73.40%) 151 (74.38%) 824.89
Grade (%) 0.168 0.637
 High grade 382 (94.09%) 186 (91.63%) 196 (96.55%) 799.85
 Low grade 21 (5.17%) 14 (6.90%) 7 (3.45%) 918.71
 Unknown 3 (0.74%) 3 (1.48%) 0 (0%) 471.67
Stage (%) 0.057
 Stage I–II 131 (37.27%) 69 (33.99%) 62 (30.54%) 720.11 0.232
 Stage III–IV 273 (67.24%) 132 (65.02%) 141 (69.45%) 847.04
 Unknown 2 (0.49%) 2 (0.99%) 0 (0%) 282.00
T (%) 0.210
 T1–T2 122 (30.05%) 62 (30.54%) 60 (29.56%) 734.56 0.526
 T3–T4 251 (61.82%) 126 (62.07%) 125 (61.86%) 829.75
 Unknown 33 (8.13%) 15 (7.39%) 18 (8.87%) 861.75
M (%) 0.192
 M0 195 (48.03%) 102 (50.25%) 93 (45.81%) 742.31 0.145
 M1 11 (2.71%) 3 (1.48%) 8 (3.94%) 1184.36
 Unknown 200 (49.26%) 98 (48.28%) 102 (50.25%) 841.32
N (%) 0.106
 N0 215 (52.96%) 117 (57.64%) 98 (48.28%) 765.70 0.476
 N1–N3 161 (39.65%) 67 (33.00%) 94 (46.30%) 874.81
 Unknown 30 (7.39%) 19 (9.36%) 11 (5.42%) 795.52
Open in a new tab

Validation of the prognostic signature

To assess the prognosis of the signature, the samples were categorized into low- and high-risk groups according to the median risk score. The “survival” package was used to calculate PFS and OS for BLCA patients. Then, we assessed the independent prognostic value of the signature by univariate and multivariate Cox regression analysis of the risk scores and the clinical characteristics (grade, stage, age and gender). Based on risk scores, a lncRNA expression heatmap and patient survival status were drawn using the “pheatmap” package. We calculated the 1-, 3-, and 5-year survival rates using the “surviminer” and “timeROC” packages and the C-index curves were used to validate whether the signature accurately predicted the prognosis of BLCA patients.

Construction of the nomogram and analysis of clinical subgroups

The “rms”, “regplot” and “survival” packages were used for constructing nomogram to estimate the prognosis of BLCA patients at 1-, 3- and 5-year, which integrated the prognostic signatures, age, gender, grade and stage. Finally, the consistency index (C-index) and calibration curves were plotted to verify the accuracy and stability of the nomogram.

Principal component analysis

Principal component analysis (PCA) was established by using “limma” and “scatterplot3d” packages to explore the sample distribution according to risk scores.

Immune-related functional and tumor mutation burden analysis

We explored the relationship between the risk score and tumor mutation burden (TMB) using the "maftools" package, and the “survival” package was used for visualizing the difference between patient survival and TMB. In addition, we divided samples into four subgroups according to risk score and the median value of TMB. K-M analysis was performed for different subgroups. The “GSVA” and “limma” packages were used for analyzing the immune-related functions of BLCA patients, and the analysis results were visualized by the “Pheatmap” package.

Pharmaceutical sensitivity analysis

Tumor treatment has been hindered by drug resistance for years. To investigate the relationship between the sensitivity of therapeutic drugs and the risk score, the “pRRophetic”, “ggplot2” and “ggpubr” packages were used to determine the 50% inhibiting concentration (IC50) with pFilter = 0.001 and corPvalue = 0.001.

Statistical analysis

The clinical data and RNA-seq transcriptome data were preprocessed using the PERL programming language (version 5.32.1) for screening prognosis-related genes. All data visualizations and statistical analyses were performed using R software (version 4.2.0) and a variety of R packages. p < 0.05 was suggested statistically significant.

Results

Acquirement of cuproptosis-related lncRNAs in BLCA patients

Previous studies have identified 19 cuproptosis-related genes (Tsvetkov et al. 2022; Polishchuk et al. 2019; Aubert et al. 2020; Dong et al. 2021; Ren et al. 2021), and 16,876 lncRNAs were extracted from the bladder urothelial carcinoma (BLCA) project of the TCGA database. There were 1426 cuproptosis-related lncRNAs identified among those genes and lncRNAs (19 cuproptosis-related genes and 16,876 lncRNAs), based on p value < 0.001 and Pearson’s correlation coefficient |R2|> 0.4.The co-expression relationships between cuproptosis-related lncRNAs and cuproptosis-related genes are illustrated in the Sankey diagram (Fig. 1).

Fig. 1.

Fig. 1

Open in a new tab

Sankey diagram for the cuproptosis-related genes and cuproptosis-related IncRNA

Enrichment functional analysis

The GO enrichment pathways mainly included epidermis development, skin development, negative regulation of transport, collagen−containing, endopeptidase activity (Fig. 2A, B). The KEGG analysis demonstrated that the cuproptosis-related IncRNAs mainly concerned with the PI3K−Akt signaling pathway, focal adhesion, rap1 signaling pathway, cytokine−cytokine receptor interaction, human papillomavirus infection, calcium signaling pathway, and so on (Fig. 2C, D). Enrichment functional analysis suggested that these IncRNAs contributed to tumor development.

Fig. 2.

Fig. 2

Open in a new tab

Results of Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses. GO (A, B) and KEGG (C, D) analyses based on the cuproptosis -related differentially expressed genes

Construction of prognostic signature

LASSO Cox regression analysis was performed using the “glmnet” package to construct the predictive score model. In the training group, 22 differentially expressed cuproptosis-related IncRNAs were found by univariate Cox regression analysis (Fig. 3A–C). Subsequently, multivariate Cox regression analysis found 9 cuproptosis-related IncRNAs related to BLCA OS as independent prognostic factors. Based on the expression levels of the 9 cuproptosis-related lncRNAs, the following formula was used for calculating the risk score of each sample. Risk score = (0.843 × AL356019.2) + (0.491 × AC010328.1) + (1.322 × LINC02773) + (1.935 × AC022405.1) + (− 1.148 × AC125494.1) + (1.106 × AC105001.1) + ( − 0.979 × AC009831.3) + (0.317 × LINCADL) + ( − 0.777 × UBE2Q1 − AS1). In addition, the relationship between 9 prognostic lncRNAs and 19 cuproptosis-related genes was highlighted through a correlation heatmap (Fig. 3D).

Fig. 3.

Fig. 3

Open in a new tab

Identification of cuproptosis-prognosis-related lncRNAs. A, B LASSO regression analysis screened cuproptosis -prognosis–related lncRNAs. C Forest plot showing 22 lncRNAs related to BLCA clinical survival prognosis obtained by univariate Cox regression analysis, with red representing high-risk IncRNAs and green representing low-risk IncRNAs. D Correlation heatmap showing the relationship between 9 cuproptosis-related IncRNAs with independent prognosis and 19 cuproptosis-related genes

Survival analysis of the signature

A total of 406 BLCA samples were divided into training and testing groups randomly, and there was no difference between clinical characteristics and the three groups (Table 1). The training group included 203 samples, which were divided into low-risk group (n = 102) and high-risk group (n = 101) based on the calculated median value of the risk scores as the cutoff value. Compared to the low-risk group, the high-risk group overall survival rate was significantly lower (p < 0.001; Fig. 4A). The reliability and accuracy of the prognostic signature were further validated using the entire dataset and testing group. Similar to training group, the entire dataset and testing group were divided into low- and high-risk groups. Similarly, K–M analysis of the testing group (p = 0.016; Fig. 4B) and entire dataset (p < 0.001; Fig. 4C) indicated that patients in the high-risk group had a lower OS than those in the low-risk group. Furthermore, progression-free survival (PFS) in the low-risk group was significantly higher than that in the high-risk group in the entire dataset (p < 0.001; Fig. 4D).

Fig. 4.

Fig. 4

Open in a new tab

Kaplan–Meier survival analyses of BLCA patients. OS of patients in the training group (A), testing group (B), and entire dataset (C). D PFS of patients in the entire dataset

The survival status and distribution of the risk score of BLCA patients were further explored between the two risk groups (Fig. 5). We found that there were less mortality and more years of survival time in the low-risk group. In the high-risk group, heatmap revealed that patients had high expression of AL356019.2, AC010328.1, LINC02773, AC022405.1, AC105001.1, and LINCADL but low expression of AC125494.1, AC009831.3 and UBE2Q1-AS1.

Fig. 5.

Fig. 5

Open in a new tab

Distribution of risk scores and survival status and expression heatmaps of the 9 cuproptosis-prognosis-related lncRNAs in the training group (A), testing group (B), and all group (C)

Validation of signature

Univariate and multivariate Cox regression analyses were used for evaluating whether the signature of 9 cuproptosis-related lncRNAs could serve as an independent prognostic factor in BLCA patients. Univariate Cox regression analyses indicated that stage, age and risk score were markedly related to OS (Fig. 6A). After correction for additional confounding variables, multivariate Cox regression analyses continued to show that the risk score was an independent predictor of OS (Fig. 6B), indicating that the cuproptosis-related IncRNAs prognostic signature was better than the clinical characteristics. Then, we assessed the specificity and sensitivity of our prognostic signature using receiver operating characteristic (ROC) curves. The results revealed that the areas under curve (AUCs) for 1-, 3- and 5-year were 0.722, 0.686 and 0.675, respectively (Fig. 6C). Meanwhile, the risk model AUC was 0.722, which was higher than stage (0.673), grade (0.533), gender (0.489) and age (0.669) (Fig. 6D), indicating that the signature was reasonably dependable in predicting BLCA prognosis. Compared with the low-risk group, patients in the high-risk group had significantly shorter OS in both stages III-IV (advanced stage) and I-II (early stage) (p < 0.001; Fig. 6E, F), illustrating that our prognostic signature was applied to all stages of BLCA.

Fig. 6.

Fig. 6

Open in a new tab

Validation of the signature in the training group. A Univariate B multivariate Cox regression analysis showed that the risk score was independently associated with OS. C ROC analysis of the 1-, 3-, and 5-year survival rates of BLCA patients. D ROC analyses showed the prognostic accuracy of the risk score and other clinical features. OS of BLCA patients in stages I-II (E) and III-IV (F)

Construction of the nomogram and principal component analysis

We established a predictive nomogram to forecast the OS of BLCA patients at 1-, 3-, and 5-year by integrating the TNM stage, grade, age, gender and risk score (Fig. 7A) and used the calibration curve for verification. The results demonstrated that the nomogram-predicted OS and observed OS had excellent consistency (Fig. 7B). Then, we plotted C-index curves to compare the risk score with clinical characteristics (stage, grade, gender, age) (Fig. 7C). The above results indicated that the prognostic signature we established could serve as an independent prognostic factor for BLCA patients. Finally, we used PCA to compare the distributions of BLCA patients in the low- and high-risk groups based on risk IncRNAs, cuproptosis–related IncRNAs, all genes and cuproptosis-related genes. The results indicated that the two risk groups displayed clear distributions in view of the risk IncRNAs (Fig. 8).

Fig. 7.

Fig. 7

Open in a new tab

Construction and assessment of a predictive nomogram. A Prognostic nomogram for predicting the 1-, 3-, and 5-year survival rates of BLCA patients. B Calibration curve of the nomogram for 1-, 3-, and 5-year survival rates. C C-index curve of the risk score

Fig. 8.

Fig. 8

Open in a new tab

Principal component analysis (PCA) observed the distribution of BLCA patients in different risk groups based on A all genes, B 19 cuproptosis-related genes, C 1426 cuproptosis–related IncRNAs, and D 9 risk IncRNAs

Immune-Related function and tumor mutation burden analysis

Recently, more and more studies have discovered that TMB was closely related to OS after immunotherapy for multiple tumor types (Klempner et al. 2020; Jardim et al. 2021). We analyzed mutations in the two risk groups to better observe the biology of the characteristic lncRNAs. The waterfall plot displayed the top 15 genes with the highest mutation rates between the high- and low-risk groups (Fig. 9A, B). TMB was higher in the low-risk group than in the high-risk group. (TTN: high risk, 45%, low risk, 57%; KMT2D: high risk, 9%, low risk, 23%; MUC16: high risk, 25%, low risk, 36%; ARID1A: high risk, 18%, low risk, 33%). According to the mutation effect predictor, all samples were divided into low- and high-TMB groups using the best cutoff value. As shown in Fig. 9C, patients had higher OS in the high-TMB group (p < 0.001). Subsequently, samples were divided into four subgroups combining the TMB groups and risk scores. The patients with high TMB and low risk scores had the highest OS, and patients with low TMB and high risk scores performed the worst survival rate (Fig. 9D). Figure 9E shows the results of the immune status in the two different risk groups, which revealed that most immune functions (type-I-IFN-response, APC functions, T functions, HLA, and so on) were significantly more active in the high-risk group.

Fig. 9.

Fig. 9

Open in a new tab

Tumor mutation burden (TMB) and immune-related functional analysis. The waterfall plot displays the top 15 genes with the highest mutation rates in the A high-risk group and B low-risk group. C OS of BLCA patients in the high- and low-TMB groups. D OS of BLCA patients in the four subgroups combining the TMB groups and risk scores. E Immune-related functions of the 9 cuproptosis-related IncRNAs

Drug sensitivity analysis

Moreover, the TIDE algorithm was used to assess the difference in sensitivity to immunotherapy in the high- and low-risk groups. Our results indicated that the TIDE score in the high-risk group was higher than that in the low-risk group (Fig. 10A). Finally, we investigated the response to antitumor drugs for BLCA patients of different risk scores using the “pRRophetic” package. As shown in Fig. 10, the sensitivity of these drugs was significantly different between the different risk groups. Most drugs (sunitinib, saracatinib, ruxolitinib, bexarotene, and cyclopamine) had lower IC50 values in the high-risk group, indicating that they were more sensitive to those drugs in the high-risk group (Fig. 10B–F). Nevertheless, FH535, WZ3105, and TAK-715 were more sensitive in the low-risk patients than in the high-risk groups (Fig. 10G–I).

Fig. 10.

Fig. 10

Open in a new tab

Comparison of two risk groups in immune escape and drug sensitivity. A Compared with the low-risk group, the high-risk group had a higher Tumor Immune Dysfunction and Exclusion (TIDE) score and greater potential for immune escape. Drug sensitivity analysis of B sunitinib, C saracatinib, D ruxolitinib, E bexarotene, F cyclopamine, G FH535, H WZ3105, and I TAK-715

Discussion

In China, bladder cancer (BLCA) has high morbidity and mortality rates among urothelial tumors, and its overall incidence rate continues to increase (Lobo et al. 2020). Previous studies reported that various clinical parameters (age, tumor size, grade, in situ cancer, and TNM stage) were closely linked to BLCA prognosis (Sylvester et al. 2596; He et al. 2021). This study examined whether cuproptosis-related lncRNAs contribute to bladder cancer development and therapy. LncRNAs have been discovered in many studies as active participants in cancer progression and have become new biomarkers in cancer diagnosis and treatment. The TCGA dataset was used to determine 1426 cuproptosis-related lncRNAs differentially expressed in BLCA patients. LncRNAs may participate in the development and occurrence of BLCA via cuproptosis. A recent study revealed that cancer development was regulated by lncRNAs in many ways, including tumor invasion, recurrence and metastasis. (Statello et al. 2021). A lncRNA-miRNA-mRNA axis was formed by the lncRNA LINC01140, FGF9 and miR-140-5p, which could affect macrophage M2 polarization through the tumor microenvironment and then regulate bladder cancer cell proliferation (Wu et al. 2020). Upregulated lncRNA HIF1A-AS2 promoted resistance to cisplatin through inhibition of the p53-dependent apoptotic pathway in BLCA patients (Chen et al. 2019). Although lncRNAs have been researched extensively in tumors, further studies will be needed to better explore the predictive biomarkers and biological mechanisms in BLCA to improve patient survival.

First, KEGG and GO analyses demonstrated that cuproptosis-related lncRNAs were mainly concerned with the PI3K–Akt signaling pathway, focal adhesion, rap1 signaling pathway, human papillomavirus infection, calcium signaling pathway, and cytokine−cytokine receptor interaction, which participate in the process of cancer progression. The PI3K pathway, as a potential therapeutic target, is activated in more than 40% of BLCA (Network 2014). Leupaxin may promote BLCA progression by upregulating the expression of S100P via the PI3K−Akt signaling pathway (Hou et al. 2018). In addition, leflunomide significantly induced apoptosis in BLCA cells via inhibition of the PI3K−Akt signaling pathway, which plays a key role in antitumor therapeutics (Cheng et al. 2020).

To explore the prognostic factors in BLCA, we identified 1426 cuproptosis-related lncRNAs from TCGA. 22 cuproptosis-related lncRNAs were confirmed to be differentially expressed, 9 of which were used to establish a signature to predict OS in BLCA patients. These 9 cuproptosis-related lncRNAs contained 6 risk factors: AL356019.2, AC010328.1, LINC02773, AC022405.1, AC105001.1, and LINCADL; 3 protection factors: AC125494.1, AC009831.3 and UBE2Q1-AS1. Studies have shown that lncRNA AL356019.2 is hypermethylated in glioma. It can serve as a potential immunotherapy target and is a promising biomarker for the prognosis of glioma (Shao et al. 2022). Meanwhile, high expression of lncRNA LINC02773 demonstrated a worse prognosis in stomach adenocarcinoma patients (Luo et al. 2022). It is worth noting that gastric cancer with high expression of lncRNA UBE2Q1-AS1 was related to its progression, indicating poor prognosis (Zhang et al. 2020). However, either univariate or multivariate Cox regression analysis indicated that lncRNA UBE2Q1-AS1 is a low-risk factor for BLCA prognosis in this study. Future researches are needed to further investigate the specific mechanism of UBE2Q1-AS1. Notably, the remaining lncRNAs have not been clarified in tumors, future research directions may be provided by our findings.

We then constructed a BLCA prognosis signature according to 9 cuproptosis-related lncRNAs. Patients were divided into two different risk groups based on median score, and the high-risk group was found to have shorter survival time. In a multivariate Cox analysis, the prognostic signature was an independent factor of survival in BLCA patients. ROC analysis showed that the risk score AUC was higher than that of clinical features, suggesting that the prognostic signature is more accurate in predicting prognosis for BLCA. The nomogram was constructed to visually forecast the patients’ survival rates at 1-, 3-, and 5-year, and calibration plots revealed the predicted 1-, 3-, and 5-year OS rates matched well with the actual results. Overall, our findings revealed that cuproptosis-related lncRNAs were independently and accurately predictive of OS in BLCA patients.

Subsequently, we explored the relationship between TMB, immune-related function and prognosis signature in BLCA patients. The results showed that most immune functions were obviously more active in the high-risk group. TMB has been regarded as a powerful predictor to predict immunotherapy response in multiple cancer types (Samstein et al. 2019). PD1-stimulated T cells with a higher TMB have stronger killing effects, which lead to better clinical outcomes. (Allgauer et al. 2018). Lv et al. found that BLCA patients with high TMB showed a better prognosis (Lv et al. 2020). At the same time, a report showed that mutation of EP300 upregulated immune system signaling pathways to promote antitumor immunity, and its mutation was linked to an increase in TMB (Zhu et al. 2020). As with prior studies, our findings showed the prognosis of BLCA patients with high TMB was better than those with low TMB. In addition, patients were divided into four subgroups combining TMB groups and risk scores. Low -risk patients with high TMB had the highest OS, and high-risk patients with low TMB performed the worst survival rate. The results clearly revealed the importance of TMB in predicting prognosis. We then used the TIDE algorithm to assess whether the two risk groups responded differently to immunotherapy. Immune evasion is more likely to occur in patients with a higher TIDE prediction score, resulting in a poor therapeutic effect. We found that TIDE scores were lower for the low-risk patients than for the high-risk patients, suggesting that immunotherapy might be more beneficial to them. Finally, we investigated the response to antitumor drugs for BLCA patients of different risk scores, and the sensitivity of these drugs was significantly different between the two risk groups. The results demonstrated that most drugs (sunitinib, saracatinib, ruxolitinib, bexarotene, and cyclopamine) had lower IC50 values in the high-risk patients, indicating that they were more sensitive to those drugs. Nevertheless, FH535, WZ3105, and TAK-715 were more sensitive in low-risk patients. Chemotherapy and targeted therapy could significantly prolong the progression-free survival and the overall survival of advanced bladder cancer patients (Patel et al. 2020; Tran et al. 2021; Liu et al. 2022). The difference between the two risk groups in the treatment response of commonly used chemotherapeutic agents for BLCA suggests that this model may be helpful in the selection of chemotherapy regimen and the judgment of curative effect. The results corroborate that this signature may be helpful to prevent drug resistance and select promising chemotherapy or targeted drugs in patients with BLCA.

In a word, we identified some novel biomarkers closely related to survival rate of patients with BLCA. Based on 9 cuproptosis-related lncRNAs, we established a prognostic signature and analyzed the relationship between immunotherapy, TMB, drug sensitivity and risk score-based groups. Although the results of our study could be used for improving the prognosis and clinical treatment of BLCA patients, further validation is needed for the prognostic signature developed in our study.

Acknowledgements

We thank the TCGA database for free.

Author contributions

Conception and design: JS and XS. Acquisition, analysis, and interpretation of the data: TW, LY, PS, and LY. Drafting and writing: JS. Final approval of the article: JS, XS, TW, LY, PS, and LY.

Funding

Not applicable.

Availability of data and materials

Data supporting the findings of this study are available from the respective authors upon reasonable request.

Declarations

Conflict of interest

The authors declare that have no competing interests.

Ethical approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Footnotes

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  1. Allgauer M, Budczies J, Christopoulos P et al (2018) Implementing tumor mutational burden (TMB) analysis in routine diagnostics-a primer for molecular pathologists and clinicians. Transl Lung Cancer Res 7(6):703–715 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Aubert L, Nandagopal N, Steinhart Z et al (2020) Copper bioavailability is a KRAS-specific vulnerability in colorectal cancer. Nat Commun 11(1):3701 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Babak MV, Ahn D (2021) Modulation of intracellular copper levels as the mechanism of action of anticancer copper complexes: clinical relevance. Biomedicines 9(8):852 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Babjuk M, Bohle A, Burger M et al (2017) EAU guidelines on non-muscle-invasive urothelial carcinoma of the bladder: update 2016. Eur Urol 71(3):447–461 [DOI] [PubMed] [Google Scholar]
  5. Bafaro E, Liu Y, Xu Y et al (2017) The emerging role of zinc transporters in cellular homeostasis and cancer. Signal Transduct Target Ther 2:1–12 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Bian Z, Fan R, Xie L (2022) A novel cuproptosis-related prognostic gene signature and validation of differential expression in clear cell renal cell carcinoma. Genes (basel) 13(5):851 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Burger M, Catto JW, Dalbagni G et al (2013) Epidemiology and risk factors of urothelial bladder cancer. Eur Urol 63(2):234–241 [DOI] [PubMed] [Google Scholar]
  8. Cambier S, Sylvester RJ, Collette L et al (2016) EORTC nomograms and risk groups for predicting recurrence, progression, and disease-specific and overall survival in non-muscle-invasive stage Ta-T1 urothelial bladder cancer patients treated with 1–3 years of maintenance bacillus Calmette-Guerin. Eur Urol 69(1):60–69 [DOI] [PubMed] [Google Scholar]
  9. Chen X, Liu M, Meng F et al (2019) The long noncoding RNA HIF1A-AS2 facilitates cisplatin resistance in bladder cancer. J Cell Biochem 120(1):243–252 [DOI] [PubMed] [Google Scholar]
  10. Cheng L, Wang H, Wang Z et al (2020) Leflunomide inhibits proliferation and induces apoptosis via suppressing autophagy and PI3K/Akt signaling pathway in human bladder cancer cells. Drug Des Devel Ther 14:1897–1908 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Dong J, Wang X, Xu C et al (2021) Inhibiting NLRP3 inflammasome activation prevents copper-induced neuropathology in a murine model of Wilson’s disease. Cell Death Dis 12(1):87 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Feng Y, Zeng JW, Ma Q et al (2020) Serum copper and zinc levels in breast cancer: a meta-analysis. J Trace Elem Med Biol 62:126629 [DOI] [PubMed] [Google Scholar]
  13. Festa RA, Thiele DJ (2011) Copper: an essential metal in biology. Curr Biol 21(21):R877–R883 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Gao N, Li Y, Li J et al (2020) Long non-coding RNAs: the regulatory mechanisms, research strategies, and future directions in cancers. Front Oncol 10:598817 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Gene Ontology Consortium (2015) Gene Ontology Consortium: going forward. Nucleic Acids Res 43:D1049–D1056 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. He H, Liu T, Han D et al (2021) Incidence trends and survival prediction of urothelial cancer of the bladder: a population-based study. World J Surg Oncol 19(1):221 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Hou T, Zhou L, Wang L et al (2018) Leupaxin promotes bladder cancer proliferation, metastasis, and angiogenesis through the PI3K/AKT pathway. Cell Physiol Biochem 47(6):2250–2260 [DOI] [PubMed] [Google Scholar]
  18. Jardim DL, Goodman A, de Melo GD et al (2021) The challenges of tumor mutational burden as an immunotherapy biomarker. Cancer Cell 39(2):154–173 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kanehisa M, Goto S (2000) KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Res 28(1):27–30 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Klempner SJ, Fabrizio D, Bane S et al (2020) Tumor mutational burden as a predictive biomarker for response to immune checkpoint inhibitors: a review of current evidence. Oncologist 25(1):e147–e159 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Li P, Li J, Wen F et al (2022) A novel cuproptosis-related LncRNA signature: prognostic and therapeutic value for acute myeloid leukemia. Front Oncol 12:966920 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Liu S, Chen X, Lin T (2022) Emerging strategies for the improvement of chemotherapy in bladder cancer: current knowledge and future perspectives. J Adv Res 39:187–202 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lobo N, Shariat SF, Guo CC et al (2020) What is the significance of variant histology in urothelial carcinoma? Eur Urol Focus 6(4):653–663 [DOI] [PubMed] [Google Scholar]
  24. Luo W, Wang J, Xu W et al (2021) LncRNA RP11-89 facilitates tumorigenesis and ferroptosis resistance through PROM2-activated iron export by sponging miR-129-5p in bladder cancer. Cell Death Dis 12(11):1043 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Luo L, Li L, Liu L et al (2022) A necroptosis-related lncRNA-based signature to predict prognosis and probe molecular characteristics of stomach adenocarcinoma. Front Genet 13:833928 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Lv J, Zhu Y, Ji A et al (2020) Mining TCGA database for tumor mutation burden and their clinical significance in bladder cancer. Biosci Rep 40(4):BSR20194337 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Mortada WI, Awadalla A, Khater S et al (2020) Copper and zinc levels in plasma and cancerous tissues and their relation with expression of VEGF and HIF-1 in the pathogenesis of muscle invasive urothelial bladder cancer: a case-controlled clinical study. Environ Sci Pollut Res Int 27(13):15835–15841 [DOI] [PubMed] [Google Scholar]
  28. Network CGAR (2014) Comprehensive molecular characterization of urothelial bladder carcinoma. Nature 507(7492):315–322 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Patel VG, Oh WK, Galsky MD (2020) Treatment of muscle-invasive and advanced bladder cancer in 2020. CA Cancer J Clin 70(5):404–423 [DOI] [PubMed] [Google Scholar]
  30. Polishchuk EV, Merolla A, Lichtmannegger J et al (2019) Activation of autophagy, observed in liver tissues from patients with Wilson disease and from ATP7B-deficient animals, protects hepatocytes from copper-induced apoptosis. Gastroenterology 156(4):1173–1189 [DOI] [PubMed] [Google Scholar]
  31. Quinn JJ, Chang HY (2016) Unique features of long non-coding RNA biogenesis and function. Nat Rev Genet 17(1):47–62 [DOI] [PubMed] [Google Scholar]
  32. Ren X, Li Y, Zhou Y et al (2021) Overcoming the compensatory elevation of NRF2 renders hepatocellular carcinoma cells more vulnerable to disulfiram/copper-induced ferroptosis. Redox Biol 46:102122 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Ritch CR, Cookson MS, Clark PE et al (2019) Perioperative oral nutrition supplementation reduces prevalence of sarcopenia following radical cystectomy: results of a prospective randomized controlled trial. J Urol 201(3):470–477 [DOI] [PubMed] [Google Scholar]
  34. Robinson NJ, Winge DR (2010) Copper metallochaperones. Annu Rev Biochem 79:537–562 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Roupret M, Babjuk M, Burger M et al (2021) European Association of Urology Guidelines on upper urinary tract urothelial carcinoma: 2020 update. Eur Urol 79(1):62–79 [DOI] [PubMed] [Google Scholar]
  36. Saleh S, Adly HM, Abdelkhaliq AA et al (2020) Serum levels of selenium, zinc, copper, manganese, and iron in prostate cancer patients. Curr Urol 14(1):44–49 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Samstein RM, Lee CH, Shoushtari AN et al (2019) Tumor mutational load predicts survival after immunotherapy across multiple cancer types. Nat Genet 51(2):202–206 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Saw PE, Xu X, Chen J et al (2021) Non-coding RNAs: the new central dogma of cancer biology. Sci China Life Sci 64(1):22–50 [DOI] [PubMed] [Google Scholar]
  39. Shao D, Li Y, Wu J et al (2022) An m6A/m5C/m1A/m7G-related long non-coding RNA signature to predict prognosis and immune features of glioma. Front Genet 13:903117 [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Siegel RL, Miller KD, Fuchs HE et al (2022) Cancer statistics, 2022. CA Cancer J Clin 72(1):7–33 [DOI] [PubMed] [Google Scholar]
  41. Statello L, Guo CJ, Chen LL et al (2021) Gene regulation by long non-coding RNAs and its biological functions. Nat Rev Mol Cell Biol 22(2):96–118 [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Sylvester RJ, van der Meijden AP, Oosterlinck W et al (2006) Predicting recurrence and progression in individual patients with stage Ta T1 bladder cancer using EORTC risk tables: a combined analysis of 2596 patients from seven EORTC trials. Eur Urol 49(3):465–466 [DOI] [PubMed] [Google Scholar]
  43. Thompson L (2006) World Health Organization classification of tumours: pathology and genetics of head and neck tumours. Ear Nose Throat J 85(2):74 [PubMed] [Google Scholar]
  44. Toden S, Zumwalt TJ, Goel A (2021) Non-coding RNAs and potential therapeutic targeting in cancer. Biochim Biophys Acta Rev Cancer 1875(1):188491 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Tran L, Xiao JF, Agarwal N et al (2021) Advances in bladder cancer biology and therapy. Nat Rev Cancer 21(2):104–121 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Tsvetkov P, Coy S, Petrova B et al (2022) Copper induces cell death by targeting lipoylated TCA cycle proteins. Science 375(6586):1254–1261 [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Wu S, Xu R, Zhu X et al (2020) The long noncoding RNA LINC01140/miR-140–5p/FGF9 axis modulates bladder cancer cell aggressiveness and macrophage M2 polarization. Aging 12:25845 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Xiao S, Zuo Y, Li Y et al (2022) Long noncoding RNA HAGLROS promotes the malignant progression of bladder cancer by regulating the miR-330-5p/SPRR1B axis. Front Oncol 12:876090 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Xie S, Zhao Z, Feng B et al (2021) A comparative study of perioperative and survival outcomes of robot-assisted radical cystectomy in patients over 80 and under 80 years old. World J Surg Oncol 19(1):202 [DOI] [PMC free article] [PubMed] [Google Scholar]
  50. Yan H, Bu P (2021) Non-coding RNA in cancer. Essays Biochem 65(4):625–639 [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Zhang X, Jiang Y, Xie Y et al (2020) Comprehensive analysis of lncRNAs associated with the pathogenesis and prognosis of gastric cancer. DNA Cell Biol 39(2):299–309 [DOI] [PubMed] [Google Scholar]
  52. Zhu G, Pei L, Li Y et al (2020) EP300 mutation is associated with tumor mutation burden and promotes antitumor immunity in bladder cancer patients. Aging (albany NY) 12(3):2132–2141 [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Zygiel EM, Nolan EM (2018) Transition metal sequestration by the host-defense protein calprotectin. Annu Rev Biochem 87:621–643 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

Data supporting the findings of this study are available from the respective authors upon reasonable request.

Articles from Journal of Cancer Research and Clinical Oncology are provided here courtesy of Springer

RESOURCES