Bladder cancer (BC) is one of the top 10 cancer types in terms of estimated cases and deaths for both men and women worldwide, responsible for an estimated 573,278 new cases and 212,536 deaths in 2020 (1). In developed regions and areas with a high prevalence of tobacco smoking, the incidence of BC is increased (2). Non-muscle invasive bladder cancer (NMIBC) accounts for more than 70% of BC cases, with a recurrence rate of about 50%-70%, and may progress to muscle-invasive bladder cancer (MIBC) in about 10%-20% of cases (3). The preferred treatment for BC is the removal of the tumor, typically done by inserting a transurethral device into the bladder (transurethral resection of the bladder tumor). Based on the depth of tumor invasion into the bladder wall and the tumor’s aggressiveness, chemotherapy, radiotherapy, or immunotherapy can be selected singly or in combination (4). Most BC cases are NMIBC, where tumors have breached the epithelial basement membrane of the bladder, and these patients have a 5-year survival rate of up to 90% (5). NMIBCs are prone to recurrence (50%-70%) but progress slowly. However, MIBC patients have a worse overall survival rate, with less than 50% surviving at five years. Distant metastases are more likely when BC cells breach the muscle layer, underscoring the importance of early diagnosis in improving survival rates (6). Therefore, early screening and diagnosis of BC before metastasis is critical to enhancing patient survival. One significant problem related to the diagnosis of BC is that standard diagnostic approaches, such as ultrasound, cystoscopy, and urinary cytology, are unsatisfactory due to poor precision or high invasiveness (7). For instance, urinary cytology, which is minimally invasive, has a specificity of 90%-100% but a sensitivity of 40%-60%, whereas cystoscopy, despite being more accurate, carries a 10% risk of complications like urinary tract infection even when using a flexible cystoscope (8). Thus, developing new diagnostic tools with higher specificity and sensitivity for early screening and detection of BC is of great significance.

MicroRNAs (miRNAs), which inhibit or degrade mRNA translation, are a type of endogenous noncoding small RNA that bind to the 3’ untranslated regions of related mRNAs to regulate gene expression. Previous studies have discovered that miRNAs participate in the occurrence and development of various malignant tumors and could serve as diagnostic biomarkers. In recent years, many studies have shown that highly stable circulating miRNAs primarily exist in lipid or lipoprotein complexes, such as exosomes, microvesicles, or apoptotic bodies (9, 10). Additionally, acquiring serum is relatively easy, and detecting RNA can be readily accomplished using quantitative reverse transcription-polymerase chain reaction (qRT-PCR). Therefore, some specific serum miRNAs show promise as potential noninvasive diagnostic biomarkers for BC.

In our study, based on qRT-PCR, we evaluated potential serum biomarkers by examining microRNA expression profiles in BC patients. The purpose of our research was to identify crucial miRNAs and develop a microRNA panel with excellent diagnostic value in three phases. Firstly, in the training phase, 10 serum miRNAs were screened using BC pools and healthy control (HC) pools. Secondly, abnormally expressed miRNAs identified in the training phase were further validated in the validation phase. Finally, the miRNAs with validated expression differences were analyzed using backward stepwise logistic regression to develop an efficient panel for BC diagnosis. Additionally, we used bioinformatic analysis to investigate the potential biological functions of the crucial miRNAs.

Studies have revealed that miRNAs play critical roles in cell metabolic pathways, intercellular communication, and modifying the tumor microenvironment. Using serum miRNAs for disease screening offers a non-invasive and convenient alternative to traditional inspection methods, making it potentially valuable in clinical practice (14, 15). Recently, several studies including Singh et al. (16), Qureshi et al. (17) and Dip et al. (18) revealed a potential relationship between miRNAs and bladder cancer, while their studies only showed the role of a specific miRNA in bladder cancer. Other studies including Li et al. (19) and Wen et al. (20) elucidated the potential of combined miRNA panel for diagnosis in bladder cancer or urothelial carcinoma. In our study, we designed a three-phase test—screening, training, and validation stages—to identify serum miRNAs that could serve as potential biomarkers for BC screening. We identified 5 miRNAs (miR-30c-5p, miR-142-3p, miR-200c-5p, miR-206, and miR-212-3p) with aberrant expression in serum between BC patients and HCs. Subsequently, we developed a three-miRNA-based signature comprising miR-212-3p, miR-30c-5p, and miR-206, which exhibited higher specificity and accuracy for BC screening compared to single miRNAs alone (AUC = 0.838; 95% confidence interval: 0.774 - 0.890). Additionally, we conducted bioinformatics analysis to explore the potential functions of these selected miRNAs.

In our diagnostic panel, miR-30c-5p was identified as overexpressed in the serum of BC patients, with an AUC of 0.724 (95% CI: 0.649-0.790). Recent studies have linked dysregulation of miR-30c-5p to advanced stage and overall survival outcomes in BC patients (21). This miRNA has been shown to be closely associated with the biological characteristics of tumor cells. For example, in glioma cells, miR-30c-5p could induce apoptosis and inhibit growth and invasion by targeting Bcl2, suggesting its potential as a therapeutic target to improve overall patient survival in glioma (22). Given these findings and our study, miR-30c-5p likely plays a significant role in the development and potential therapeutic targeting of BC as well.

Several studies have consistently demonstrated that miR-212-3p is downregulated in various types of tumors. It has been observed that miR-212-3p exerts anti-tumor effects by targeting NFIA, thereby inhibiting cell proliferation directly through NFIA regulation. It suggests that targeting NFIA with miR-212-3p could potentially serve as a therapeutic strategy for BC (23). Moreover, studies have indicated that overexpression of miR-212-3p can also suppress MAP3K3 protein expression, further inhibiting cell invasion, proliferation, and migration in certain cancers, such as high-grade serous ovarian cancer (HGSOC) (24). Combining these insights into its pathways and targets, the importance of miR-212-3p in BC screening was further validated, demonstrating an AUC of 0.785 (95% CI: 0.715-0.845).

Research has revealed that miR-206 plays a crucial role in regulating BC cell proliferation, invasion, and migration by targeting RMRP. The long non-coding RNA (lncRNA) RMRP was expressed at higher levels in BC cells compared to adjacent tissues and strongly associated with tumor size, lymphatic metastasis, and overall survival in BC patients (25). Furthermore, another study has demonstrated that miR-206 targets YRDC to suppress tumor development in BC (26). This dual role of miR-206 in regulating multiple pathways related to BC pathogenesis suggests its suitability as a serum marker for BC diagnosis, with an AUC of 0.741 (95% CI: 0.668–0.806).

After predicting target genes, a total of 8 genes met the criteria of |log2FC| > 1.5 and p < 0.01, including ATF3, GJA1, JPH2, MVB12B, RUNX1T1, SLC8A1, SPATA6, and TPM3 ( Figure 5 ). Therefore, they might be potential target genes of the three miRNA-based panel. Studies have shown that ATF3 inhibits the metastasis of BC cells, and ATF3 up-regulates actin remodeling mediated by GSN. Thus, ATF3 and GSN could serve as prognostic markers for BC metastasis (27). Additionally, numerous studies have indicated that GJA1 could be a promising prognostic marker for poor survival and a therapeutic target in cervical cancer, suggesting a key regulatory role across various cancers (28). The positive feedback loop involving RUNX1T1 contributes to BC progression and may present a potential treatment strategy (29). CircSLC8A1 is implicated in the CircSLC8A1/miR-130b and miR-494/PTEN axes as a tumor suppressor, potentially offering therapeutic targets for BC treatment (30). Moreover, TPM3 is positively associated with Th2 cells and other immune cell infiltration in BC, influencing the development of BC through immune modulation (31). However, there has been limited research on the relationship between JPH2, MVB12B, SPATA6, and BC. In conclusion, these 8 genes play significant roles in the tumorigenesis and progression of BC and various other cancers. Further exploration into their specific roles in bladder carcinogenesis and progression is warranted.

Overall, our results were demonstrated to be credible and significant; however, several limitations should be addressed. Firstly, the sample size of both patient and control groups in this study was relatively small. Secondly, while there have been numerous BC-related miRNAs in serum, we only investigated 10 miRNAs in this study. Additionally, a more comprehensive bioinformatics analysis of the miRNAs in this experiment would enhance understanding about their clinical utility.

In summary, we developed a three miRNA-based panel (miR-212-3p, miR-30c-5p, and miR-206) with potential as novel and noninvasive serum markers for BC diagnosis and screening.