Bladder cancer (BC) is the tenth most common cause of cancer worldwide and is the thirteenth leading cause of cancer mortality. The non-muscle invasive (NMI) variant represents 75% of cases and has a mortality rate of less than 1%; however, it has a high recurrence rate. The gold standard of management is transurethral resection in the case of new lesions. However, this is associated with significant morbidity and costs, so the reduction of these procedures would contribute to reducing complications, morbidity, and the burden to the health system associated with therapy. In this clinical scenario, strategies such as active surveillance have emerged that propose to manage low-risk BC with follow-up; however, due to the low evidence available, this is a strategy that is underutilized by clinicians. On the other hand, in the era of biomarkers, it is increasingly known how to use them as a tool in BC. Therefore, the aim of this review is to provide to clinical practitioners the evidence available to date on AS and the potential role of biomarkers in this therapeutic strategy in patients with low-grade/risk NMIBC. This is the first review linking use of biomarkers and active surveillance, including 29 articles.

Bladder cancer (BC) is the most common malignancy of the urinary tract. Approximately 75% of cases are non-muscle-invasive BC (NMIBC), while muscle-invasive BC (MIBC) and advanced tumors account for most cancer-specific mortality. Accurate assessment of tumor invasion is essential, as staging variability may lead to inappropriate treatments. Tumor invasion involves several mechanisms including extracellular matrix (ECM) remodeling mediated by metalloproteinases, angiogenesis, and cell adhesions. Masson’s trichrome staining (MTS) provides relevant information on ECM composition. This study evaluated the application of machine learning to MTS-stained bladder biopsies to predict muscle invasion. A retrospective analysis of bladder biopsy images obtained from transurethral resections and cystectomies (2022–2024). A total of 702 histological images were analyzed. A convolutional neural network (CNN) was trained to classify tumors as MIBC or NMIBC and model outputs were correlated with clinical variables. The CNN achieved an accuracy of 95.2% in the training set and 90.1% in validation. Model-derived probabilities were significantly associated with tumor grade, lesion size, and muscle invasion. Logistic regression demonstrated a strong association with invasive disease (OR = 0.07, p = 0.017). CNN-based analysis of MTS-stained bladder biopsy images enable accurate prediction of muscle invasion, with potential to improve diagnostic precision.

Background: Bladder cancer (BC) is the most common malignancy of the urinary tract. Approximately 75% of cases are non–muscle-invasive BC (NMIBC), while muscle-invasive BC (MIBC) and advanced tumors account for most cancer-specific mortality. Accurate assessment of tumor invasion is essential, as staging variability may lead to inappropriate treatments. Tumor invasion involves several mechanisms including extracellular matrix (ECM) remodeling mediated by metalloproteinases, angiogenesis, and cell adhesions. Masson’s trichrome staining (MTS) provides relevant information on ECM composition. This study evaluated the application of machine learning to MTS-stained bladder biopsies to predict muscle invasion. Methods: A retrospective analysis of bladder biopsies images obtained from transurethral resections and cystectomies (2022–2024). A total of 702 histological images were analyzed. A convolutional neural network (CNN) was trained to classify tumors as MIBC or NMIBC and model outputs were correlated with clinical variables. Results: The CNN achieved an accuracy of 95.2% in the training set and 90.1% in validation. Model-derived probabilities were significantly associated with tumor grade, lesion size, and muscle invasion. Logistic regression demonstrated a strong association with invasive disease (OR = 0.07, p = 0.017). Conclusion: CNN-based analysis of MTS-stained bladder biopsies images enables accurate prediction of muscle invasion, with potential to improve diagnostic precision.