*4.1. Modeling Bladder Cancer*

Bladder cancer (BC) is the most common cancer of the urinary system and ranks globally as the 10th most common type of cancer, a pathology that affects men more than women [164,165]. BC is defined as a carcinoma of the urothelial cells and can be divided by the tumor depth of bladder-wall invasion in non-muscle-invasive BC (NMIBC), muscle-invasive BC (MIBC), and metastatic BC, subtypes that are characterized by different molecular signatures, BC being one of the most frequently mutated human cancers [166,167]. While the approval of immunotherapy for both NMIBC and MIBC, as well as targeted therapy (erdafitinib) and antibody-drug conjugate therapy (enfortumab vedotin) significantly improved the therapeutical management of BC, the alarming statistics regarding the disease recurrence and overall survival of BC patients highlight the emerging need for developing better preclinical study models [168]. At the moment, the most intensively used preclinical models for BC research are in vitro 2D cell cultures (cell lines, conditionally reprogrammed cell cultures) and 3D organoids, and in vivo carcinogen-induced mouse models, genetically engineered mouse models, and patient-derived xenograft. However, each preclinical model presents unique features, together with different disadvantages such as failure to mimic the 3D tumor microarchitecture, microenvironment, and tumor heterogeneity; and lack of immune system, thus failing to ease the transition between preclinical models and clinics [169–172]. For example, a highly neglected aspect of many

preclinical models is the presence of the tumor microenvironment consisting of malignant and nonmalignant cells (cancer-associated fibroblasts, cancer-associated immune cells), ECM, and blood vessels, an entity that orchestrates cancer progression and modulates therapeutic response [20].

In this view, Liu and colleagues [173] aimed to reconstruct the bladder microenvironment by coculturing four types of cells into a two-layer microfluidic device made of a PDMS piece and a glass slide. The fabricated microfluidic device connected to perfusion equipment featured four indirectly connected cell-culture chambers (BC cells, fibroblasts, macrophages, endothelial cells), ECM channel units (Matrigel), and culture-medium channels. Therefore, using this microfluidic device, four types of cells were allowed to simultaneously interact through soluble biological factors and metabolites that proved to diffuse through the ECM channel units between cell-culture chambers, in a dynamic setup provided by continuous medium perfusion. The validated microfluidic system proved to be a good platform for cell-motility patterns and phenotypic alteration of stromal cells, as well by generation of reticular structures based on BC cells, and opened the perspective of OOC implementation for precision medicine, as BC cells treated with different clinical neo-adjuvant chemotherapy showed different treatment responses, revealing the drug sensitivity of tumor cells in this experimental setup. In another study, a microfluidic chip was designed to coculture BC cells and fibroblasts and further analyze changes in mitochondrial-related protein expression of these cells and their characteristics of energy metabolism [174].

Therefore, to investigate tumor metabolism, the microfluidic chip fabricated was composed of four cell-culture pools, two for BC cells and two for fibroblasts, interconnected through two ECM microchannels (Matrigel) and two microchannels with the outside, which were continuously fed with cell-culture medium injected through the peripheral perfusion channel. By comparing the coculture system with individual cultures of BC cells and fibroblasts, significant differences in lactic-acid concentration and mitochondrial-related protein expression were observed; results that revealed that cells conduct glycolysis more efficiently under coculture conditions, and show enhanced overall mitochondrial activity and protein expression, highlighting the importance of using OOC in bladder tumor energymetabolism studies. Finally, Lee et al. [175] designed a simple 3D microfluidic device based on PDMS and Matrigel to culture MIBC cells and NMIBC cells, for modeling metastasis. The study showed an increased expression of CD44 and RT4 after 2 weeks of culture in MIBC cells as compared with NMIBC cells, associated with a significant increase of MMP-9 gelatin degradation, showing that the OOC system could be further employed for migration and metastasis studies.

#### *4.2. Modeling Urinary Tract Infections*

On the other hand, urinary tract infections (UTI) are among the most common bacterial infections, divided into uncomplicated and complicated infections, isolated infections of the bladder being referred to as cystitis and treated subsequently with antibiotics [176]. However, despite completing the antibiotic-based treatment regimen, cystitis is characterized by a high recurrence frequency, which involves readministration of antibiotics [177]. UTI is most frequently caused by uropathogenic *E. coli* (UPEC) bacteria, which underlies more than 80% of the diagnosed cases [178]. Once entering the urinary bladder, the *E. coli* can float freely in the urine and be eliminated through micturition or form intracellular bacterial communities (IBC) that the bladder struggles to clear [179]. Therefore, to reveal the insights of UPEC infection and IBC formation, Sharma and colleagues [180] used a commercially available OOC purchased from Emulate to model UPEC infection and mimic the interface between the blood vessels and the tissue layers of the human bladder. To reconstruct the bladder native architecture, human bladder microvascular endothelial cells were cocultured with human bladder epithelial cells and exposed dynamically to urine and nutritive cell-culture media, while using the application and release of linear strain to recreate micturition. By infecting the epithelial layer underflow with UPEC and monitoring OOC by microscopy, the bacteria motility, interaction with cells, and IBC formation could

be monitored. Moreover, the addition of human-blood isolated neutrophils into the OOC system revealed that their diapedesis to sites of infection on the epithelial side can lead to the formation of neutrophil swarms and neutrophil extracellular traps (NETs), and that IBCs offer substantial protection to bacteria from antibiotic clearance. Administration of antibiotics in the developed bladder-on-a-chip model through urine revealed their potential to kill bacteria floating freely in the urine much faster than bacteria residing in bladder cells, as well as the increased resistance of IBCs to treatment, aspect that highlight the importance of completely eradicating IBCs to avoid infection recurrence. No doubt, this complex study shows the potential of the bladder-on-chip as relevant platforms for modeling infections, as well as their use for drug screening of multiple antibiotics or novel drugs in a physiologically relevant manner.

### **5. Prostate-on-Chip**

The prostate is an exocrine gland that secretes sperm-nourishing and protective fluid. It is located beneath the urinary bladder and part of the male reproductive system [181]. The prostate-tissue architecture consists of ducts and acini lined with glandular or secretory cells on top of basal cells and surrounded by a fibromuscular stroma [182,183]. Consequently, the prostate displays two major components: one is the epithelium, containing neuroendocrine cells and basal cells that express integrins and hold the differentiation potential toward luminal (secretory) cells that will express androgen receptor protein [184,185]; the other component of the prostate tissue is the stroma, which is separated from the epithelium by a basement membrane. Prostate cancer is the most common cancer diagnosed worldwide in men and is characterized by molecular changes caused by genetic and epigenetic modifications leading to malignant transformation of the cells [186,187]. Androgens regulate prostate development and cell differentiation from embryonic development to adults, and thus, a better understanding of early interactions between prostate cells leading to development would be essential in unraveling the mechanisms underlying prostate cancer [185]. Many studies in this field have been developed, starting with 2D conventional cultures and 3D models, and ending with versatile proposals of prostate-on-chip systems. Likewise, for other previously discussed pathologies, in prostate cancer, 2D cell cultures fail to provide the proper tissue-level complexity, not only in their limitation to one cell type but also due to their missing crucial TME aspects, as well as FSS simulation [188]. The schematic representation of the prostate-on-chip model is presented in Figure 3.

**Figure 3.** Prostate-on-chip.

On the other hand, animal models that do display the in vivo tissue-level complexity fail to mimic human anatomy or physiology [189]. To address this issue, various engineered microfluidic devices have been proposed for concomitant multiple cell-culture types. For example, Picollet-D'hahan, N. et al. used polyelectrolyte nanofilms to create a model of prostate epithelial cancer cells (PC3) and normal prostate epithelial cells (PNT-2) [190]. Furthermore, Jiang et al. [191] reported the development of a human prostate gland model for a better understanding of the epithelium/stroma interaction, both in terms of unrevealing the R1881-mediated epithelial cell (PrEC)-differentiation mechanism towards functional secretory cells and coculturing of these cells with prostate stromal cells (BHPrS1). For this, they designed a microfluidic device fabricated by soft-lithography techniques from Polydimethylsiloxane (PDMS), which incorporates a commercial polyester membrane of 0.8 μm-diameter pores, 1% porosity, and a thickness of 23 μm. The major advantage of this device is its ability to overcome the impossibility of establishing long-term 2D static cocultures of these cell types due to their different culture-media requirements [191]. Moreover, this device was able to sustain an in vivo-like behavior of the cells that displayed groups forming gland-like buds.

#### **6. Benefits and Challenges of OOCs/TOCs in Urology**

Experimental research demonstrated that microfluidic OOC technology might function to screen recently developed anticancer compounds, cellular and nanotechnology-based treatments, improve therapy conditions, and establish the effects (or side effects) of combined therapies in in vivo-like TME (Figure 4).

**Figure 4.** Schematic representation of the available in vitro and in vivo models for the urologyassociated pathology study, highlighting their advantages and disadvantages and revealing the potential key role of the organ/tumor-on-chip devices to bridge the gap between conventional 2D/3D culture systems and animal models in preclinical studies.

Imitating organ-level pathophysiology found in vivo is the key factor in developing successful OOC models and requires clear efficacy validation. One of the main challenges in OOC fabrication and usage remains the straightforward optimization to functionally connect multiple organ systems in order to collect media from the output of the microdevice and feed it into another to accurately mimic the sequential adsorption, distribution, metabolism, and excretion of the compounds. The microfluidic technology allows perfusion of cell-culture medium through or across tissue structures and maintains biomechanical stimuli in a controlled manner (for instance, continuous, intermittent, or cyclic). Moreover, during an experiment, the easy access to cells or media for dosing, sampling, and analysis with small-molecule compounds (antibodies, hormones, drugs, etc.) offers an unquestionable advantage over the animal and other previous static models, including spheroids and organoid cultures. Previous studies already showed that TOCs mimic the native environment of cells, including 3D topology or physical stretch and strain, and they are able to maintain precise control over inter-and intra-organ flow rates, forming a miniaturized version of the human body. Additionally, bladder-on-chip models allow modeling of urinary infections by simulating *E. coli* bacterial infection of the urinary tract cells. These insights are crucial in the framework of studies that focus on in vivo processes such as angiogenesis, tumorigenesis, invasion, metastasis, and infection, enabling in vitro analysis of how local microenvironmental signals and chemical gradients influence these processes [192]. The existence of a perfused endothelium-lined capillarity endorses the superiority of OOCs over static models by providing precise control over cell-culture conditions and over platform pharmacokinetics and drug toxicity, respectively. The ability to adjust key parameters such as concentration gradients, cell patterning, tissue–organ interactions, and to replicate an organ-like mechanical microenvironment that exists in human living TME placed OOCs in the "Top 10 Emerging Technologies" at the World Economic Forum [193].

The multiple benefits of OOCs have made them indisputable alternatives for animal studies or 3D culture assays. However, clinical trials require time, and OOCs are still competing with animal testing to predict clinical responses. Moreover, OOCs are not considered as easy to be handled as conventional cell models. Although they allow long-term experiment and minimal user input, the technical robustness remains a challenge, as the compact and complex microfluidic devices are engaging multiple parameters that simultaneously run to achieve optimal functionality. Air-bubble formation or unintended infection risks at connection points are the most common factors that may lead to experiment failure. Shourabi et al. [194] recently designed an optimized integrated microfluidic gradient generator for mechanical stimulation and drug delivery by dynamic culturing of human lung cancer cells (A549 cell line) made from two PDMS-layer microfluidic chips with a porous membrane interfacing in between. The key feature of this system consists in two bubble trappers designed to remove the unwanted bubbles that could enhance the shear stress on cells. In this way, the concentration gradient generator's performance is guaranteed, and precise control of fluid shear stress on cells is obtained. The results showed that the chip exhibits a high cell-viability rate (95%) and will be employed to study the toxicity effect of different concentrations of cisplatin on renal cells.

Microfluidic OOC microdevices can be also employed for studying the role of cancer stem cells (CSCs) in tumor initiation and cancer relapse [195]. As the chemotherapeutic regiments kill tumor cells but are ineffective on CSCs, the microfluidic OOCs have the potential to give insights on tumor cell–CSC interaction, as well as accelerating drug development to also target this subpopulation of cells [196]. By employing a biomimetic in vitro model such as OOCs for CSC research, various environmental parameters such as oxygen gradient, glucose, and fluid shear stress can be mimicked, offering valuable insights on CSCs' multicellular interaction and drug-resistance mechanisms to conventional therapies [197].

However, there is still a place for developing new materials that facilitates the noninvasive assessment of drug effects and meets the industrial-scale production demands. Novel materials based on biopolymers have been developed in the last few years to overcome the main drawbacks of PDMS-based OOC microdevices, and significant attempts to produce scalable systems have been reported [198]. Therefore, as the technology has evolved in the last years and the introduction of commercial prefabricated tumor chips that are user-friendly and custom-made for culture and fluidic control is expanding [199], it will undoubtedly facilitate the possibilities of medical research for oncology treatments.

### **7. Conclusions and Future Considerations**

Human OOCs/TOCs offer a promising alternative to conventional 2D/3D static cultures, and even to in vivo models that lack similarity with human anatomy and/or physiology. These microfluidic devices hold the potential to sustain relevant preclinical studies for developing new therapies, including anticancer drugs, and bring new insights for a better understanding of pathogenesis mechanisms. Replicating the physiological TME, TOC models are seen as promising and more realistic alternatives for investigating the metastasis, distribution, and mechanism of tumor propagation. Involving microfluidic technologies, OOC models can mimic the complexity of in vivo tumors and can be more accurately used to predict therapeutic efficacy and drug toxicity or side effects. Although OOCs closely mimic the native organ functions, one must consider that most of the cell cultures interact with polymeric substrates and porous polymeric membranes to replicate the physiological microenvironments. Thus, the polymer material properties such as surface roughness, wettability, and mechanical properties significantly affect cell adhesion and proliferation.

Moreover, size porosity and biodegradability influence cell migration and viability. While most of the studies present the use of common synthetic polymers beneficial for OOCs fabrication, other studies employ less common polymeric structures with improved cytocompatibility, wettability, or mechanical stability. Although there is a place for deeper investigations to optimize the materials' properties and fabrication methods, the OOC/TOC models are adding necessary steps to personalized medicine for creating high-precision remedies with the possibility to use biopsy from the patient that could be expanded in vitro and evaluated for cost-effective screening of treatments that are particularly efficient for that patient.

**Author Contributions:** B.G., A.H., M.C. and E.S. reviewed the literature with respect to kidney-onchip models, bladder-on-chip models, and prostate-on-chip models; wrote Sections 3–5, contributed to Sections 1, 6 and 7; and illustrated Figures 2–4. E.I.B., H.I., R.-C.P. and C.Z. reviewed the literature regarding OOC/TOC fabrication; wrote Sections 1, 2, 6 and 7; and illustrated Figure 1. V.J. coordinated and edited the entire work. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work was supported by a grant of the Romanian Ministry of Education and Research, CNCS-UEFISCDI, project number PN-III-P1-1.1-PD-2019-0955, within PNCDI III.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** Data are contained within the article.

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**

