1.1. Bladder Cancer

Bladder cancer is a significant global health concern due to its high incidence, substantial mortality, and the challenges it presents in diagnosis and treatment. It is a heterogeneous disease with varying clinical presentations and outcomes, requiring individualized management strategies. Men are affected about three to four times more often than women, likely due to differences in exposure to risk factors such as smoking and occupational carcinogens [1] (Figure 1). The median age at diagnosis is 73 years, making it a disease predominantly of the elderly. Bladder cancer accounts for an estimated 213,000 deaths annually worldwide. While non-muscle-invasive bladder cancer (NMIBC) has a lower mortality rate, muscle-invasive bladder cancer (MIBC) and metastatic cases contribute significantly to mortality [2,3].

1.2. The Critical Role of Early Detection in Bladder Cancer Management

Early detection of bladder cancer is pivotal for improving outcomes, as it enables timely interventions that significantly enhance survival rates, prevent disease progression, and reduce the burden of advanced treatments. Early-stage bladder cancer, particularly non-muscle-invasive bladder cancer (NMIBC), has a far better prognosis compared to advanced stages, with a 5-year survival rate of ~77% for localized disease, compared to ~30% for regional spread and <10% for distant metastases. Early identification of NMIBC enables effective management with intravesical therapies, such as Bacillus Calmette-Guérin (BCG), preventing progression to muscle-invasive bladder cancer (MIBC), which is more aggressive and requires radical treatments [4,5,6]. Early diagnosis broadens treatment options, favoring minimally invasive approaches like transurethral resection of bladder tumors (TURBT) and bladder-preserving therapies, while reducing the need for radical cystectomy and systemic chemotherapy, thereby preserving bladder function and quality of life. It also facilitates better risk stratification, allowing high-risk patients to be closely monitored for recurrence or progression using advanced non-invasive diagnostic tools, such as urine-based DNA methylation assays and NMP22 tests, which improve patient compliance and screening feasibility. Furthermore, early detection reduces the costs associated with lifelong surveillance and advanced disease management, helps prevent metastasis through timely systemic and targeted therapies, and ultimately improves survival outcomes while lessening the physical, emotional, and financial burdens on patients [7,8,9,10].

1.3. The Role of DNA Methylation in Epigenetic Regulation and Development

DNA methylation is a fundamental epigenetic mechanism involving the addition of a methyl group to the fifth carbon of the cytosine ring, primarily within CpG dinucleotides (Figure 2). This reaction is catalyzed by DNA methyltransferases (DNMTs), including DNMT1 (which maintains methylation during DNA replication) and DNMT3A/3B (which establish new methylation patterns). Methylation in promoter regions typically represses gene transcription by recruiting methyl-CpG-binding domain proteins (MBDs), which compact chromatin and restrict transcriptional access. This process is vital for regulating gene expression, silencing repetitive sequences, and maintaining genomic stability. However, when disrupted, it can contribute to the development of different diseases, including cancer [11,12,13,14].

DNA methylation plays a vital role in gene regulation by influencing transcriptional activity, genomic stability, and cellular processes. Methylation in gene promoter regions is closely linked to transcriptional repression, as methylated promoters recruit methyl-CpG-binding domain proteins (MBDs) that form complexes with histone-modifying enzymes to compact chromatin, restricting access to the transcriptional machinery. This mechanism is critical during development, such as silencing pluripotency genes in differentiated cells. Additionally, DNA methylation helps to maintain genomic stability by silencing repetitive sequences and transposable elements, preventing inappropriate recombination and genomic instability [15]. It contributes to X-chromosome inactivation in females, equalizing gene dosage with males, and plays a role in genomic imprinting by regulating parent-specific expression of certain genes essential for normal growth and development. However, aberrant DNA methylation is implicated in various diseases, particularly cancer. Hypermethylation of tumor suppressor gene promoters, such as p16 or MLH1, leads to gene silencing and facilitates uncontrolled cell proliferation. Conversely, global hypomethylation, a hallmark of many cancers, activates oncogenes and transposable elements, increasing genomic instability. Thus, DNA methylation is a dynamic process that maintains cellular homeostasis but, when disrupted, contributes to disease pathogenesis [16,17,18].

1.4. DNA Methylation and Bladder Cancer: A Dual Role in Genomic Stability and Tumorigenesis

In bladder cancer, DNA methylation patterns become dysregulated, contributing to tumorigenesis and disease progression. Two major types of alterations are commonly observed: global hypomethylation and locus-specific methylation changes [19].

1.5. Unlocking the Diagnostic Potential of Urine in Bladder Cancer

Urine serves as an excellent diagnostic medium for bladder cancer due to its accessibility and the non-invasive nature of its collection, making it ideal for repeated sampling and monitoring. As a direct contact medium with the bladder, urine contains a wealth of molecular information shed by tumor cells, allowing for the detection of a variety of biomarkers. These include DNA (e.g., mutations, methylation patterns), RNA (e.g., microRNAs, mRNAs), proteins (e.g., NMP22, cytokeratins), and metabolites, all of which can provide valuable insights into tumor biology and progression. The ease of urine collection reduces patient discomfort compared to invasive procedures like cystoscopy, while its composition reflects dynamic changes in the tumor microenvironment, offering opportunities for early detection, risk stratification, and monitoring of treatment responses. These attributes position urine-based diagnostics as a powerful tool for improving the non-invasive management of bladder cancer [23].

Urine DNA methylation assays have emerged as a promising tool for the non-invasive diagnosis and monitoring of bladder cancer. These assays detect abnormal methylation patterns in cell-free DNA or exfoliated tumor cells present in the urine, offering insights into tumor biology and epigenetic alterations associated with cancer. Aberrant DNA methylation, such as hypermethylation of tumor suppressor gene promoters (e.g., CDKN2A, RASSF1A, DAPK), is a hallmark of bladder cancer and plays a critical role in tumor progression and recurrence. The high sensitivity and specificity of methylation assays make them particularly valuable for detecting bladder cancer at early stages, even before visible tumors are detected by imaging or cystoscopy. Additionally, these assays are ideal for monitoring patients during follow-up to detect residual disease or recurrence, reducing the need for invasive procedures. Advances in technologies such as methylation-specific PCR (MSP), droplet digital PCR (ddPCR), and next-generation sequencing (NGS) have enhanced the reliability and scalability of these tests. Urine DNA methylation assays hold immense potential not only for improving early detection but also for risk stratification and guiding personalized treatment strategies, making them a key innovation in bladder cancer management [24,25].

Clinical trials have demonstrated the utility of urine-based DNA methylation assays for various clinical objectives in bladder cancer management, including diagnosis, monitoring, and biomarker discovery. For early diagnosis, techniques with high sensitivity, such as quantitative methylation-specific PCR (qMSP) and droplet digital PCR (ddPCR), have been evaluated for their ability to detect low-abundance methylated DNA associated with bladder tumors. For instance, a trial investigating the methylation of CDKN2A and RASSF1A in urine samples reported high sensitivity and specificity for distinguishing bladder cancer patients from healthy controls, even in early-stage disease. These findings highlight the potential of qMSP and ddPCR to facilitate early detection, reducing reliance on invasive cystoscopy and improving patient compliance [25,26].

In the context of disease monitoring, precision techniques like ddPCR and targeted next-generation sequencing (NGS) have shown promise in detecting residual disease and predicting recurrence. A multicenter trial utilizing ddPCR to monitor RASSF10 and APC methylation in urine demonstrated that methylation levels correlate strongly with recurrence risk, providing a non-invasive method for long-term surveillance. Such approaches enable timely intervention, improving outcomes while reducing the need for frequent invasive procedures. Additionally, studies combining ddPCR with other urine biomarkers, such as protein or RNA signatures, have enhanced the predictive power of these assays for recurrence detection. For biomarker discovery, whole-genome or targeted NGS platforms have been pivotal in identifying novel methylation markers for bladder cancer. Trials utilizing (NGS) to profile urine samples have identified several methylation candidates, such as ZIC4 and GATA4, which can distinguish between different subtypes and stages of bladder cancer. These biomarkers are being validated in ongoing clinical trials for their diagnostic, prognostic, and predictive utility. Furthermore, NGS-based studies have revealed methylation patterns specific to high-risk bladder cancer, aiding in patient stratification and the development of personalized treatment strategies [23,24].

1.6. Comparative Analysis of Urine DNA Methylation Assays and Other Bladder Cancer Biomarkers

Urine-based DNA methylation assays have emerged as highly promising tools in bladder cancer diagnostics, specifically when evaluated alongside established and emerging urinary biomarkers such as cytology, NMP22, UroVysion FISH, and multiplex platforms like Cxbladder. While urine cytology is highly specific (>90%), its low sensitivity, particularly for low-grade tumors, limits its use as a standalone diagnostic [27]. NMP22 provides moderate sensitivity (50–70%) but lower specificity and can yield false positives in benign conditions, whereas UroVysion FISH offers improved accuracy for high-grade tumors with 60–80% sensitivity and 70–85% specificity. Cxbladder, which measures multi-gene mRNA signatures, shows high negative predictive value, making it useful for ruling out malignancy [28]. By comparison, DNA methylation assays targeting genes like CDKN2A, RASSF1A, and TWIST1 demonstrate consistently high sensitivity (70–90%) and specificity (85–95%) even for early-stage disease [29]. For initial diagnosis in patients with hematuria, combining methylation markers with traditional tests like cytology or NMP22 enhances diagnostic sensitivity without compromising specificity. During follow-up care, these assays also reduce reliance on invasive cystoscopy, with studies confirming their utility in surveillance settings [29,30]. There is also growing interest in combining DNA methylation assays with other biomarker types such as proteins (e.g., NMP22) and RNA signatures (e.g., Cxbladder) to boost accuracy and reduce diagnostic uncertainty. These approaches are particularly beneficial for high-risk patients or when test results are inconclusive, helping to improve clinical decision-making and avoid missed diagnoses or unnecessary procedures [31,32]. While molecular tools like UroVysion and Cxbladder offer enhanced performance, they come with higher costs and require specialized infrastructure, limiting their accessibility. In contrast, methylation assays using techniques such as qMSP and ddPCR offer strong diagnostic accuracy with better scalability and cost-efficiency. These assays can be centralized, streamlining laboratory processes and enabling integration with other molecular diagnostics, an advantage for clinics seeking accurate, practical, and affordable bladder cancer testing options [24,33,34].

1.7. CDKN2A, RASSF1A, and Beyond: A Clinical Perspective on Urinary DNA Methylation Testing in Bladder Cancer

Several clinical studies have investigated the potential of urinary DNA methylation biomarkers such as CDKN2A and RASSF1A in the diagnosis of bladder cancer, each contributing valuable insights into assay performance and clinical applicability. A large multicenter study evaluated a methylation-based urine test in over 1000 patients with hematuria, demonstrating high sensitivity (89.2%) and specificity (87.8%), notably outperforming traditional methods like urine cytology and NMP22 [35]. Similarly, a study validated a qMSP-based panel including GHSR and MAL, achieving 80% sensitivity and 93% specificity in urine samples [36]. Earlier research reviewed the broader landscape of DNA methylation assays, comparing detection technologies such as ddPCR and NGS, and highlighting challenges related to assay standardization and variability in DNA sources [37]. Supporting these findings, another study reported a correlation between CDKN2A and RASSF1A promoter hypermethylation and tumor grade in urothelial carcinoma, reinforcing their prognostic relevance [38]. More recently, a novel marker, mDMRTA2, was introduced and validated through a urine-based assay with strong diagnostic accuracy [39]. Collectively, these studies underscore the growing evidence for urinary methylation biomarkers while emphasizing the need for standardized methodologies, larger cohorts, and real-world validation to support clinical adoption.

1.8. Multi-Omics and Machine Learning: Unlocking Precision Oncology in Bladder Cancer

Recent progress in cancer systems biology has underscored the growing impact of multi-omics integration in refining prognostic models for bladder cancer. By merging DNA methylation data with other molecular layers such as transcriptomics and genomic alterations, researchers are gaining a more complete understanding of tumor dynamics. For example, one study used a machine learning–based multi-omics approach to investigate mitotic catastrophe heterogeneity in bladder cancer, identifying distinct molecular subtypes and the gene ANLN as a significant prognostic marker, demonstrating how integrative strategies can uncover meaningful biological patterns [40]. Another study combined RNA-seq, miRNA-seq, methylation profiles, and number variations using a transfer learning-based Cox model, resulting in improved prognostic accuracy across various datasets [41]. Integrating multi-omics data with clinical treatment information has revealed gene combinations associated with both treatment response and patient outcomes, thereby highlighting novel therapeutic opportunities [42]. This approach isn’t limited to bladder cancer; for instance, a study in endometrial carcinoma identified a novel methylation-based prognostic signature by linking methylation with gene expression data [43]. Within bladder cancer specifically, large-scale methylation analyses have identified key prognostic biomarkers [44], and early-stage genomic studies have provided insights into molecular subtypes and potential treatment targets [45].

In parallel, the incorporation of artificial intelligence (AI) and machine learning (ML) into multi-omics analysis has accelerated advances in cancer diagnostics and risk prediction. Deep learning methods have been particularly successful at integrating complex datasets, with a benchmark study revealing that models like moGAT are strong performers in classification, while efmmdVAE and efVAE show strength in clustering tasks [46]. These findings suggest that selecting the right model depends heavily on the research objective and data structure. Graph-based approaches are also gaining ground—models like LASSO-MOGAT have achieved high accuracy by integrating mRNA, miRNA, and methylation data through Graph Attention Networks and LASSO-based feature selection [47]. Together, these developments highlight the promise of AI-driven, multi-omics strategies in advancing precision medicine and improving patient outcomes in oncology.