Urothelial carcinoma (UC) is the second most frequent urogenital cancer after prostate cancer, which among all cancers, was ranked eleventh globally1. Almost 30% of UC patients develop recurrence within two years of treatment2. This high recurrence rate makes UC one of the most expensive cancers to manage. The treatment costs reach up to US$162,574 per patient annually in the US and Canada, surpassing breast (US$85,772) and lung cancers (US$135,567)3, 4, 5.

Understanding UC recurrence is hindered by inadequate in vivo models and the inability to effectively study the tumor microenvironment. Traditional 2D monolayer culture fails to accurately model cancer etiology, as they lack cell-to-cell and cell-to-matrix interactions, tissue architecture, metabolic networks, and mechanical properties. A 3D model has proven critical not only for UC but also for advancing understanding in various cancers such as prostate and breast cancer6. Moreover, it provides enhanced accuracy in mimicking the TME, understanding metastasis, and improving research outcomes in drug testing. Cells can sense dimensionality and other physical and biological features of the ECM via integrins. Cells react differently to 2D and 3D substrates, triggering various signaling pathways in each7. Protease activation which leads to extracellular matrix (ECM) degradation is hindered in 2D models8. Conversely, cells in 3D cultures can sense and respond to ECM features, triggering distinct signaling pathways compared to 2D cultures7. For instance, 3D cultures show higher expression of integrins and MMP9, which are crucial for cell spheroid formation9. Although animal models offer advantages over 2D cultures, issues of reproducibility, scalability, and cost limit their use10.

3D culture models provide a more accurate representation for studying UC pathogenesis by maintaining the tumor microenvironment (TME) and multicellularity. They allow genetic and phenotypic analysis of sequential events during pathogenesis, thus clarifying the disease cascade. Moreover, 3D cultures could better facilitate research on cancer cell proliferation, migration, invasion, and response to chemo- and radio-therapy. Co-culturing spheroids with components of the TME, such as stromal and immune cells, helps in understanding how tumors adapt to their microenvironment. For example, Martinelli (2022) showed that SDHB and SDHD silenced pheochromocytoma spheroids exhibit different behaviors when interacting with the TME, with impaired stroma metabolism inhibiting tumor aggressiveness11. Spheroid models allow researchers to explore tumor-TME interactions and discover potential therapeutic targets, which are crucial for developing effective cancer treatments.

In addition to radiotherapy, other drugs, such as metformin, can significantly affect spheroid patterns and interactions within the TME. Martinelli (2022) demonstrated that metformin induces distinct responses in pheochromocytoma/paraganglioma tumor cells and primary fibroblasts. The study showed that metformin differentially impacted tumor spheroids and fibroblasts within the TME, altering their metabolic profiles and reducing tumor aggressiveness. Specifically, metformin suppressed proliferation in tumor cells while inducing changes in fibroblast metabolism, demonstrating how drugs can modulate both tumor cells and TME components12. This demonstrated that drugs not only directly target cancer cells but also alter TME dynamics, which can influence overall tumor behavior. This highlights the importance of considering drug effects on multiple cell types within the TME when evaluating treatment responses.

3D cell cultures are categorized into spheroid and organoid models. Spheroids arise from either single-cell proliferation by a monotype of cell or through multi-cell aggregation by multiple types of cells, the latter resulting from natural cell tendencies to agglomerate in suspension cultures13, 14, 15. Self-renewal in spheroids can be demonstrated by culturing cells at very low densities in ultra-low attachment microwells, allowing the identification of cells with self-renewal capabilities16. Organoid culture is a cell culture method to recapitulate 3D architecture resembling an organ. Unlike organotypic culture which involves re-culturing a whole organ derived from an organism, organoid culture is grown from multicellular and conducive biochemical factors to develop an organ-like structure17, 18. An organoid may exhibit three characteristics: self-organization, multicellularity, and functionality19. Firstly, stem cells of the organ of interest are cultured in culture media with optimal supplements to promote growth and differentiation. Within 14-30 days, the culture will grow and arrange itself in a 3D structure resembling an organ which demonstrates partial or complete organ function17.

Various methods have been developed to grow organoids for research and therapeutic purposes. Lung organoids from induced pluripotent stem cells (iPSCs) are used to study COPD and lung transplants20, 21. Neurological disorders are studied using brain organoids that are developed through the differentiation of iPSCs into organoids. The organoid demonstrates various brain cell types and specific brain regions capable of replicating physiological and pathological conditions of the blood-brain barrier22. Small intestine organoids are used for disease modeling, regenerative medicine, drug screening, and gene repair studies23.

The human urinary bladder, lined by a specialized stratified epithelium, comprises three layers: the luminal layer with umbrella cells, the intermediate layer with polygonal cells, and the basal layer with cuboidal cells and bladder stem cells. This is followed by connective tissues, muscle, and fat layers. This structure reorganizes from 5-7 layers when relaxed to 2-3 layers when the bladder is full without structural damage, illustrating its transitional epithelium nature24.

Although organoid development of various human organs has been recognized since early 2015, the organoid culture of the urinary bladder is still lacking. Bladder organoid cultures must replicate this tissue organization, forming 4-6 closely packed cell layers with distinct basal, intermediate, and luminal layers capable of retaining urine and preventing its permeabilization into the underlying structure25. Self-renewal and differentiation are the primary characteristics of human adult stem cells that underpin organoid culture26, 27. By seeding stem cells on epithelial stroma in ideal conditions, researchers can expand and differentiate these cells using genetic tools, producing tissue-like structures for research and therapeutic applications.

Spheroids are commonly derived from differentiated tumor cell lines and do not require stem cells for their formation. In contrast, organoids rely on stem cells for their long-term maintenance in vitro, as they retain high replicative potential. Stem cells and cancer stem cells (CSCs) are subpopulations found in tissue biopsies. Due to limited biopsy samples, researchers often use cell lines for 3D culture development. Non-cancerous cell lines lack or have limited stem cell populations compared to cancerous cell lines, which possess CSC subpopulations capable of developing into spheroids and organoids28.

An accurate and cost-effective model to study UC biology and chemotherapy resistance mechanisms is urgently needed, as these are major causes of recurrence29. The UC organoid culture model holds potential for evaluating existing and novel therapeutic targets, thus personalizing UC management and enhancing convenience for patients and healthcare providers. This review reports current advances in the 3D culture of UC cell lines, highlighting its applications, limitations, challenges, and future perspectives.

Article Search and Selection Strategies: The electronic databases PubMed, Scopus, and Web of Science were searched using the keywords “3D culture,” “urothelial carcinoma cell line,” and “spheroid.” A complete list of the search strings used is presented in the supplementary file. The search was restricted to articles published from January 2000 to April 2022 and in English. The articles were screened for inclusion based on the following exclusion criteria: focus on 3D brachytherapy, 3D ultrasound, 3D radiotherapy, 3D PET and CT images, immunohistochemical 3D reconstruction, 3D modeling from anatomical images, non-UC cell lines, non-original research or review articles, non-English articles, and those without available full texts. Figure 1 depicts an outline of the literature search in a PRISMA flow diagram.

Figure 1

Figure 3

Figure 4

The search produced 525 hits, of which 473 articles were deemed eligible based on the abstract, and 195 were included after reference check and full-text screening. The publication year ranged from January 2000 to April 2022. Based on the analysis of 195 studies, several key trends were observed regarding the approaches used to develop 3D UC culture models (supplementary data). Early research mostly used simpler approaches such as hanging drop (HD), methylcellulose (MC), and collagen-based scaffolds. However, there has been a recent trend towards the use of more advanced materials and techniques. Complex platforms such as microfluidic chips and ultra-low attachment dishes (ULA) are more favored because they provide a more accurate simulation of the TME.

Besides, there has been a growing trend towards multimodal approaches. Researchers are increasingly combining various methods to enhance the accuracy of 3D models. For instance, combining ULA with Matrigel (ULA+M) or collagen with fibronectin (ULA+coll+fibronectin) demonstrates an effort to mimic both the physical and biochemical properties of the TME more comprehensively. Additionally, there has been a clear shift towards scaffold-free culturing techniques, particularly the use of ULA and poly 2-hydroxyethyl methacrylate (Poly HEMA), which minimize the potential interference with the cellular behavior caused by scaffolds.

There is also advancement in integrating biomimetic designs and microfluidic technology into 3D UC models. This innovation improves the physiological relevance of the models by demonstrating dynamic conditions, such as nutrient flow and waste elimination, better mimicking the in vivo TME. Another significant development has been the increasing use of coculture systems, where UC cells are co-cultured with other cell types, such as fibroblasts and human umbilical vein endothelial cells (HUVECs). This approach helps replicate cellular interactions within the TME and provides valuable insights into cancer progression and metastasis.

To date, there are hundreds of approaches and methods to culture UC cell lines under 3D conditions for various applications (Supplementary Tables 1 and 2). The earliest literature described the forced floating method, where cells are cultured in suspensions, independent of surface attachments30. There are three ways to suspend cells: by culturing in ULA wells, poly-hydroxyethyl methacrylate (poly-HEMA) pre-coated dishes and the hanging drop (HD) method. In the HD method, 10-20µL cell suspensions were placed on a plate. The plate was then flipped upside down, which pulled down cells to spontaneously form aggregates at the bottom of the drop13, 31. This method is also used to generate embryoid bodies from embryonic stem cells. Alternatively, the rotary culturing system enables spontaneous cell aggregation32. Another approach, static liquid overlay, induces cells to aggregate instead of sticking to a surface when suspended cells are cultivated on a nonadherent substrate33. Spheroids can also be formed by the air-liquid interface (ALI) system, where the bottom and upper parts are separately exposed to liquid and air, respectively34. Some researchers combined forced floating and liquid overlay techniques to obtain the desired outcome based on their research objectives. More recently, spheroids and organoids have been grown on nonadherent microfluidic chips35.

Although organoid culture of various organs was introduced in 2000, there was limited research focused on developing a complete structural organization and functional urinary bladder organoid using cell lines. Among the 195 selected articles, a majority discussed the spheroid culture method, which we considered a very well-established method. However, less than 10 papers discussed the organoid culture method. They defined urinary bladder organoid as 4-6 layers of stratified urothelium on the surface of 2-3 mm3 of either collagen, de-epithelialized stroma, Matrigel™ or reconstituted basement membrane, with the presence of tight and gap junctions on the surface of the cells31, 36, 37, 38. The most sophisticated bladder organoid was a 3D human vesical equivalent of the bladder which was created from primary cells. Histological investigation revealed morphological similarities to natural human mucosa with the presence of ECM, a capillary-like network, basement membrane and tight junctions in the mature stratified urothelium. The vesical equivalent was then transplanted with T24 and RT4 cell lines to study their invasive potential31. This review reported that there are limited studies performed to demonstrate organoid differentiation into a stratified urothelium layer, which is 9/195 (4.6%)28, 34, 36, 39, 40, 41, 42, 43, 44 with no reports describing methods to develop functional urinary bladder vesical structure. Novel analysis and validation methods for assessing urine storage cavity during contraction will facilitate its translational potential.

Development of an ideal organoid model will aid in closing the gap between simplistic 2D monolayer culture and complicated transgenic animal host. Biomimetic design inspires researchers to keep innovating and improving the design of current 3D culture models. Once the system is well-established, patient-derived UC organoids will constitute a sophisticated and accurate model system for investigating tumor evolution and treatment response in the context of personalized cancer treatments. Besides, the platform has the potential to be developed for the regeneration of bladder organs, which will greatly benefit patients with radical cystectomy.

Future research should focus on integrating interdisciplinary approaches, such as tissue engineering with computational tools. Molecular docking can be applied to study interactions of drugs and therapeutic targets, while machine learning algorithms can predict patient-specific outcomes. The incorporation of microfluidic chips will enable high-throughput screening, offering a high degree of external control and self-automation, which are keys for clinical translation. Furthermore, future research could also explore the interaction between the urothelium and endogenous microbiota. Given the potential oncogenic role of microbiota, organoid models offer a promising platform to study these interactions. Incorporating such studies may provide additional insights into tumor biology and open new avenues for therapeutic interventions.

These advancements in 3D UC models could reduce dependency on animal models, improve drug testing accuracy, and lead to more personalized and effective treatments. Robust UC 3D models promise to revolutionize therapeutic approaches and provide tangible clinical benefits for patients. The combination of emerging technologies, such as machine learning and microfluidic platforms, further enhances the potential for these models to be implemented in healthcare, pushing the boundaries of precision medicine.

2D: Two-dimensional, 3D: Three-dimensional, 5-FU: Fluorouracil, ALI: Air-liquid interface, CAR-T: Chimeric antigen receptor T cells, COPD: Chronic obstructive pulmonary disease, CSCs: Cancer stem cells, ECM: Extracellular matrix, HD: Hanging drop, iPSCs: Induced pluripotent stem cells, MC: Methylcellulose, MMP9: Matrix metallopeptidase 9, MUC1: Mucin 1, PGS: Polyglycerol sebacate, Poly HEMA: Poly 2-hydroxyethyl methacrylate, PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses, SDHB: Succinate dehydrogenase complex iron sulfur subunit B, SDHD: Succinate dehydrogenase complex subunit D, TF5: Thymosin fraction 5, TME: Tumour microenvironment, TRL: Technology Readiness Levels, UC: Urothelial carcinoma, ULA: Ultra-low attachment

None.

Siti Farizan Mansor: Design & Conceptualization, Investigation, Writing - Original draft, table & figure. Tengku Muhamad Faris Syafiq: Writing - Original Draft, Writing - Review & Editing. Badrul Hisham Yahaya: Writing - Review & Editing. All authors read and approved the final manuscript.

This study was funded by Ministry of Higher Education Malaysia through Fundamental Research Grant Scheme (DP KPT FRGS/1/2019/SKK15/UITM/03/1), Project ID 15531, RMC file number 600-IRMI/FRGS 5/3 (286/2019).

Data and materials used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Not applicable.

Not applicable.

The authors declare that they have no competing interests.