The human renal system is an exceptional example of biological structure and organization, efficiently filtering waste from the bloodstream while selectively retaining nutrients. However, the urothelium of the bladder is prone to transformation. Each year, over 425,000 people around the world develop urothelial bladder cancer, and more than 165,000 people die of this condition. The age-standardized rate of bladder cancer can differ by nearly ten-fold based on geography and is approximately three to five times higher in men than in women. Among the most common risk factors for future development and recurrence of bladder cancer are smoking [1
], body mass [2
], lack of physical activity [3
], and age [4
]. Furthermore, the mortality rate is highly dependent on stage at diagnosis. Patients diagnosed while cancerous cells are confined to the urothelium have a five-year relative survival rate of 95.7%, while patients who are diagnosed after metastasis have a five-year relative survival rate of just 5.0% [5
Patients are typically diagnosed with urothelial bladder cancer following symptoms like haematuria [6
], increased the urinary frequency or urgency, irritation during urination [7
], and bone and/or flank pain. Patients with bladder cancer that has not spread beyond the urothelium typically undergo transurethral resection of the bladder tumor, while tumors that have spread into and beyond the muscle lining of the bladder usually require a combination of cystectomy, chemotherapy, and/or radiation therapy [8
]. However, non-invasive tumors recur in roughly 70% of patients [9
], necessitating long-term monitoring. A typical course of follow-up monitoring entails cystoscopy and cytology quarterly for the first two years following treatment, bi-annually for the next two years, and annually thereafter [10
]. Such extensive monitoring contributes to bladder cancer’s status as the most expensive type of cancer on a per-patient basis [11
]. The current standard of periodic cystoscopies increases the risk of urinary tract infections [13
] and causes significant patient discomfort both during and after the procedure [14
]. Likewise, the turnaround time for clinical cytology is on the order of 2–3 days, a lengthy wait for results [15
Despite significant benefits that will arise from improving the standard of care for bladder cancer, no device for detecting bladder cancer based on biomarkers has yet achieved a recommendation for widespread clinical implementation [16
]. The performance of existing tests varies with a number of factors, including hematuria [18
], the specific clinician performing the assay [20
], and the presence of other bladder conditions [21
]. Furthermore, these tests tend to have lower sensitivity (proportion of positive test results in which a patient indeed has bladder cancer) with earlier stages of bladder cancer [22
], limiting their ability to detect bladder cancer when treatment is most likely to have a positive outcome. As a result, currently available non-invasive methods are not necessarily more cost-effective than the prevailing standard of cystoscopy and cytology [25
]. At present, no biomarker-based assay is recommended for standard clinical implementation, due to a lack of prognostic value and monitoring efficacy [16
Given the shortcomings of current screening and diagnostic methods for bladder cancer, microfluidic devices hold promise for improved patient care and outcomes. In comparison to the current combination of invasive cystoscopy and slow cytology, robust devices for screening, diagnosis, and monitoring of bladder cancer will expand physicians’ ability to detect cases in early stages when survival rates are highest. The devices will also provide frequent updates on patients’ progress during treatment and beyond. In other areas of research, the development of microfluidic devices has vastly expanded the ability to probe microscale biological environments. By manipulating the small-scale flow and mixing of fluids, scientists are able to precisely perturb and monitor systems in a way not possible through standard laboratory bench techniques [27
]. Areas as diverse as chemical synthesis [28
], cell culture [29
], and single-cell omics [30
] have all benefited from microfluidics as a platform for parallelized processing, multiplexed assays, and high-throughput measurements. Specifically, in cancer research, microfluidic devices have been used for the isolation of circulating tumor cells from blood, including tumors originating in the lung [31
], breast [33
], and prostate [35
Further strengths of microfluidic devices make them particularly well-suited to the screening and diagnosis of bladder cancer. Rather than probing individual or a select few bladder cancer biomarkers at a time, multiplexed microfluidic assays could increase the number of analytes being tested in a single protocol [37
], contributing to a more sensitive and nuanced understanding of an individual patient’s case of bladder cancer. Similarly, the ability to process multiple patient samples in parallel would reduce the amount of time spent waiting on assay results [37
] and support a faster feedback loop between patient symptoms and clinical treatment. Microfluidics’ ease of fabrication lends itself well to rapid prototyping and translation of design ideas into functional devices. Thus, the development of effective microfluidic screening and diagnostic methods for bladder cancer will likely lead to improved patient health outcomes, significant cost savings, and efficient use of clinical resources.
Here, we review recent developments in microfluidic devices that noninvasively detect clinically relevant markers of bladder cancer. We describe various approaches to the problem of early bladder cancer detection and evaluate their merits in the screening and diagnosis of bladder cancer. By compiling current methods and considering the elements necessary for the clinical implementation of a screening or diagnostic tool, we hope to provide insights into the factors most likely to lead to significant improvements in bladder cancer detection and treatment standards.
5. Future Perspectives
Today, bladder cancer is one of the most common types of cancer, and demographic trends in common risk factors, such as age and smoking, point toward further increases in the number of cases worldwide. With the ageing of populations in developed nations [110
], the proportion of individuals at increased risk of bladder cancer will also rise. Similarly, global trends in smoking hint at populations at risk of bladder cancer: the prevalence of smoking in Eastern Europe is high and stable, while in sub-Saharan Africa, the prevalence is projected to increase [112
]. Given that bladder cancer treatment and long-term monitoring incur the highest costs of any cancer type on a per-patient basis, it is likely that bladder cancer will place larger and larger burdens on health care systems around the globe. To mitigate this scenario, effective detection methods for bladder cancer bear value, both for patient lives and for healthcare resources.
Future efforts should focus on formulating a set of criteria that an effective, non-invasive bladder cancer detection method should meet. As with any medical device, an ideal tool would be both low-cost and robust, such that the burden of bladder cancer is reduced without sacrificing medical insights. As promising devices progress from the lab bench to clinical trials, the metrics chosen to evaluate accuracy should cover (1) high specificity to exclude patients without bladder cancer and (2) high sensitivity to provide early detection. The aim of promoting early-stage detection would provide clinicians with the flexibility to select for cheaper and less damaging treatment options. Currently, existing monitoring methods are optimized for patients with advanced bladder cancer [22
]. Given that the five-year survival rate of patients whose bladder cancer is confined to the urothelium at diagnosis is over 19 times higher than the rate for patients whose cancer has metastasized [5
], screening and diagnostic devices with consistent performance across cancer stages would bring significant benefits to bladder cancer patient care.
Early detection of bladder cancer is crucial to improving patient outcomes [114
]. Device performance in early-stage cases of bladder cancer could be feasibly improved through two complementary approaches: (1) development of assays that detect vanishingly small concentrations of materials linked to the presence of bladder cancer (e.g., biomarkers, exfoliated tumor cells, cell-free genetic material, etc.); and (2) development of assays targeting materials with high biological specificity to only individuals with bladder cancer (as compared to materials present in healthy individuals’ urine and exhibiting elevated concentrations in bladder cancer patients’ urine). The first approach may consist of iterative improvements upon currently-existing devices or the realization of novel techniques to assay known bladder cancer markers, while the second will likely require fundamental scientific inquiries to identify materials whose mere presence is specific and precise enough to indicate that a patient has bladder cancer.
The development of devices targeting materials with high biological specificity would provide significant benefits over currently available devices dependent on the appropriate selection of biomarker concentration cut-off values, as uniform application of cut-off values can lead to erroneous results in patients just below or above the cut-off [115
]. For biomarkers that are typically present in the urine of healthy individuals and have elevated levels in bladder cancer patients, somewhat arbitrary choices of cut-off concentrations can lead to stark differences in medical decision-making. But on the other hand, allowing flexibility in the selection of cut-off values to account for different patient populations and medical practitioners inhibits the ability to accurately compare research results and evaluate device efficacy [117
]. Variations in cut-off concentrations may drive potentially conflicting conclusions on the ability of a given device to accurately identify individuals with bladder cancer. Given these issues with determining appropriate concentration cutoffs, which affect all devices targeting biomarkers that are present in normal urine but have elevated levels in that of bladder cancer patients, the identification of biomarkers produced only by cancerous cells and not by other healthy cells of the bladder could strengthen medical decision-making through a more nearly binary classification of patients into healthy or cancerous. Through research into assays targeting biomarkers with high biological specificity to bladder cancer, researchers can mitigate the effects of technical noise and assay decision criteria for improved clinical insights and diagnoses.
Additionally, an emphasis should be placed on the development of screening and diagnostic methods amenable to clinical implementation. For instance, many currently available ELISA-based assays have temperature-dependent antibody behaviour and require expensive plate readers [119
], factors that contribute to the recommendation against their general use in population-wide screening. In contrast, wearable sensors and phone-based detection exemplify the link between ease of implementation and breadth of use. By enabling doctors to take unobtrusive, straightforward measurements of patient physiology, such technologies have brought medical monitoring into patients’ daily lives for long-term, continuous observation of individual medical cases. Commercially available wearable sensors provide near-instant readouts on parameters such as heart rate, temperature, and oxygen saturation in a user-friendly manner [120
]; even more directly relevant to bladder cancer microfluidic devices are phone-based sensors with functionalities such as calculating quantitative ELISA assay results [121
] and detecting point mutations in tumor samples [122
]. At a deeper level, wearable and phone-based sensors point to further guiding principles in the development of microfluidic devices to assay bladder cancer: the importance of designing devices that are compatible with hospitals’ pre-existing equipment and with patients’ daily regimens. In addition to strong screening and/or detection performance, a microfluidic device’s long-term impact on bladder cancer care will be shaped by its ability to seamlessly integrate into practitioners’ diagnostic and treatment routines.
While detecting exfoliated bladder cancer cells in voided urine provides a pain-free approach, sensitivity of cell-based tests would be limited by the number of exfoliated tumor cells. Since low-grade tumors shed few cells [113
] and the presence of any circulating tumor cells confers poor prognosis [123
], low numbers of exfoliated cells could limit the utility of cell-based tools for early detection. Thus, non-cell based biomarkers such as extracellular vesicles, an agent of cellular communication for budding tumors within the tumor microenvironment [124
], may be a promising alternative for detection at earlier stages. Given the newfound recognition of extracellular vesicles as a source of biomarkers, methods to isolate them are varied, including size filtration, antibody capture, and precipitation [125
]. However, standardization of isolation techniques must be addressed before clinical utilization, ideally achieving a balance between precision, cost, and duration of tests. Robust developments in this area would first require a unified goal of standardized EV isolation techniques. Given this need for robust EV isolation methods, the clinical utility of EV isolation remains an area of particular opportunity for future research.
In the future, screening and diagnostic tools that function without specialized equipment would provide a path toward broad acceptance, while high-throughput tools would enable rapid evaluation of cancer cases and increase the number of patients able to receive care. Furthermore, bladder cancer detection tools with simple, straightforward designs would reduce training requirements and inter-practitioner variability, thereby decreasing healthcare costs and ensuring reliability in bladder cancer screening and diagnosis. Non-invasiveness of the detection method is another area that should be taken into consideration. Invasive methods such as cystoscopy cause discomfort and pain to patients, limiting their application in early screening and routine monitoring of bladder cancer.
For us to have a more comprehensive and patient-friendly standard of care, non-invasive diagnostic methods are in great need. With a bladder cancer detection tool that provides the above traits, medical practitioners could deliver point-of-care, real-time assessments of a patient’s bladder cancer status. Conscious consideration of these features will be key to achieve sweeping improvements to the standard of care for bladder cancer patients through effective microfluidic screening and diagnostic devices.
6. Concluding Remarks
Considering the future, one can envision an even larger role for microfluidic devices in the treatment of bladder cancer. At present, research into novel cancer detection methods is ongoing in hopes of developing an alternative that surpasses invasive cystoscopy. Current FDA-approved diagnostic tests are still used in conjunction with, but never in place of, standard diagnostic tests. Building on the current standards of screening and diagnostics, non-invasive microfluidic devices could provide a platform for uncovering rich information on the diversity of individual patients’ cases. Many biomarkers of bladder cancer are present in the urine of patients, which enables the collection of large volumes of patient samples with minimal patient disturbance, particularly as compared to the volume limitations and pain of drawing blood that occurs for investigations on circulating tumor cells of other types of cancer. It is likely that microfluidic devices will have a largely complementary role to established standards such as cystoscopy and cytology (e.g., confirming the results from established standards or suggesting a need for further tests during long-term monitoring). To transition from promising results to widespread clinical implementation, early adoption of microfluidic devices will depend on partnerships between researchers and medical practitioners committed to bringing the promise of microfluidic devices to fruition. Over time, as such partnerships demonstrate the accuracy, throughput, and ease-of-use of particularly successful microfluidic devices, increasing numbers of practitioners will opt to incorporate applicable devices in their diagnosis process, as part of a self-reinforcing cycle.
Key capabilities of microdevices will provide a strong foundation for personalized care and mechanistic frameworks at the level of both individual cases and patient populations over time. The ease of obtaining patient urine samples facilitates testing and development of novel devices, but also provides ample material to detect and isolate materials indicating the presence of cancerous cells. While some screening and diagnostic devices target biomolecules or proteins, others isolate exfoliated tumor cells [80
] and nucleic acid-containing vesicles [81
]; this genetic material could provide a starting point for downstream genetic analyses to inform treatment decisions and provide insights into bladder cancer development and variants without requiring expensive and invasive biopsy procedures. In studies on lung [126
] and breast cancer [127
], microfluidic devices for mimicking tumor microenvironments provide a path toward tailored treatment approaches, while single-cell ‘omics’ analysis yields invaluable insight into the heterogeneity and molecular processes of cancer [128
]. This is especially important in addressing inter-patient tumor heterogeneity [129
]. For instance, by applying techniques of single-cell transcriptomics to patients’ exfoliated tumor cells, researchers could capture and sequence cellular mRNA to quantify transcription levels of individual cancer cells, enabling analysis of correlated genes, development of networks of gene co-regulation, and identification of distinct cell types with unique genetic signatures [130
]. In the clinic, such fine-level knowledge of cancer cells could point toward the exact factors of dysregulation driving the development of a patient’s tumor and underpin the development of testable hypotheses for druggable mutations and effective treatment methods.
In particular, the ability to computationally cluster data from related cell populations could provide an in silico method to isolate data from rare exfoliated tumor cells (as compared to contaminant urothelial cells) [133
] and could also indicate distinct subpopulations of cancerous cells that would likely benefit from treatment with a coordinated, concerted panel of drugs informed by knowledge of the characteristics of individual cells [136
]. More broadly, application of single-cell transcriptomics to exfoliated bladder cancer cells from a large population of patients would provide a reference transcriptomic dataset of bladder cancer variants, useful for placing patients within a broader context of prior knowledge and for predicting efficacy of potential treatment paths based on historical data [139
]. Likewise, techniques for investigating DNA methylation and changes in chromatin accessibility could provide mechanistic insight into the genomic changes that drive alterations in gene expression, cellular physiology, and progression to a cancerous state [140
]. As a complementary direction to that of single-cell omics, clinicians could identify which drugs are most likely to be effective for a given patient through isolation, ex vivo expansion, and drug screens on a patient’s own circulating tumor cells. Such an approach has yielded prognostic information on the responses of breast cancer patients to medical treatment [142
], and microfluidic devices that isolate exfoliated tumor cells could bring the power of personalized medicine to bladder cancer treatment.
Rather than being limited to monitoring bladder cancer through cell counts or biomarkers, clinicians could understand changes in genetic circuits at the scale of individual patients and identify druggable mutations for improved patient outcomes based on microfluidic devices that isolate genetic material from bladder cancer cells. Thus, the development of such microfluidic devices would provide synergistic effects on patient treatment by detecting the presence of bladder cancer and informing the selection of effective treatment options. A non-invasive method to obtain primary source human bladder cancer cells would enable the evaluation of medical treatments directly on samples from patients themselves, providing a robust test system with direct relevance to today’s bladder cancer patients and accelerating the implementation of research developments in clinical settings. Conversely, widespread clinical implementation of high-throughput assays on cancer cells isolated from patients would also bear impact on research directions, as the growth and proliferation of data on patients’ cancer cell genomics and treatment outcomes would inform our understanding of bladder cancer development and mechanisms within the larger context of multiple measurement modalities across diverse populations of patients. Thus, we anticipate that research on non-invasive devices with capabilities for both detection and an individualized analysis of bladder cancer phenotypes will yield exciting results that benefit patients, clinicians, and researchers alike.