The Development of Non-Invasive Diagnostic Tools in Bladder Cancer.

Bladder cancer is a common urinary tract cancer with a difficult clinical course. With frequent recurrence, patients with a history of bladder cancer often undergo surveillance that involves invasive cystoscopies and biopsies. Not only is this financially burdensome for patients but it is also mentally and physically intensive. Given this predicament, the field has shifted towards the use of non-invasive urinary tests to detect bladder cancer earlier in the disease course and to avoid unnecessary procedures. The first non-invasive test developed was urine cytology; however, that was found to have a low sensitivity, especially for low-grade lesions. There are many tests that are available that utilize common protein biomarkers to enhance the sensitivity of detection. However, many of these tests lack the specificity seen with cytology. With recent technological and research advancements, there are newer detection systems such as RNA sequencing and microfluidics along with novel bladder cancer biomarkers including mRNAs, methylation patterns and exosomes, which have potential to be used in clinical practice. The aim of this review is to highlight established non-invasive bladder cancer diagnostic tests as well as innovative methodologies that are on the horizon for use in bladder cancer detection.

Introduction

Bladder cancer is the 10th most common cancer in the world with an increasing incidence in developed nations.1,2 In the US, it is the fourth most common cancer among men and 10th among women.3 It is estimated that there will be about 83,730 new cases and about 17,200 deaths from bladder cancer among men and women in 2021.4 Similarly, in Europe, the mortality rates per 100,000 person-years are 3.2 for men and 0.9 for women.5 Bladder cancer is the most common malignancy of the urinary tract with an incidence rate four times higher in men compared to women.6 As the world population ages, incidence rates in both sexes are expected to increase in the US and European countries.7 Well-established risk factors include tobacco smoking and occupational exposure to aromatic amines and hydrocarbons.8 Other risk factors are dietary (artificial sweeteners, coffee, and meat consumption) and genetic factors.9 Bladder cancer represents a debilitating clinical entity with a high financial burden, and a substantial impact on one’s quality of life, with studies showing a significant detrimental effect on physical, mental, and social well-being.3,10

At first presentation, about 75% of all bladder cancers are non-muscle invasive bladder cancer (NMIBC).8 There are various stages of NMIBC including carcinoma in situ (CIS), confinement to the mucosa (Ta) and confinement to the submucosa (T1).8 About 70% of NMIBC are Ta, 20% T1, and 10% CIS.11 The gold standard treatment of NMIBC often depends on the stage but commonly involves transurethral resection (TUR) of the tumor and adjuvant intravesical chemotherapy or immunotherapy such as Bacillus Calmette–Guérin (BCG) to reduce recurrence rate and risk of progression; specifically, these therapeutic measures such as BCG are hypothesized to correct the immune system disequilibrium occurring during carcinogenesis through immune-stimulation with a detrimental effect on tumor cells.8,12 The gold standard for diagnosis and surveillance for bladder cancer is white light cystoscopy (WLC) and urinary cytology. WLC has a sensitivity of 85–90% for papillary tumors and 67% for CIS.13 Unfortunately, it has limited detection of subtle changes such as small, low-grade tumors, with up to 20% commonly missed, which may result in early recurrence.13 WLC is also limited in distinguishing benign versus malignant lesions with individuals who had a prior TUR or chemotherapy.13 Furthermore, not only does WLC have poor detection of low-grade and CIS tumors but can also be uncomfortable for patients and increases risk of urinary tract infections and urethral stenosis.10 Due to these limitations, NMIBC has a high recurrence rate (50–70%) with up to 20% progressing to muscle invasive bladder cancer (MIBC).13 Multiple international guidelines and panels have recommended a surveillance schedule based on recurrence risk and degree of progression.11,14,15 Therefore, NMIBC requires constant monitoring and surveillance based on the initial stage at diagnosis.

With recent technological advancements, efforts were made to improve this distinction with methods such as fluorescence cystoscopy, narrow band imaging and confocal laser endomicroscopy. However, these are still invasive and expensive techniques.6,13 Therefore, urine cytology has become increasingly relied upon for non-invasive detection and surveillance due to its low cost and ease of use. Although urine cytology has high specificity (~86%), it is limited due to its low sensitivity (48%), especially with low-grade NMIBC.16 Frequent screenings with WLC can be mentally, physically, and financially burdensome for patients. Due to the invasiveness of the procedure, there is a need for a non-invasive and reliable alternative. Many novel urinary biomarkers are under investigation and six assays are commercially available for use with WLC; however, none are currently adopted into routine clinical practice due to high cost or poor sensitivity. Combinations of these biomarkers are underway to increase specificity and/or sensitivity to avoid unnecessary WLC for low-grade NMIBC tumors. Table 1 provides a general overview of the various assays utilized in the diagnosis of bladder cancer.

Table 1

Bladder Cancer Assays

Test	Mechanism	Sensitivity (%)	Specificity (%)	Reference
WLC	Examine bladder with cystoscope	67–90	83	[13]
WLC with biopsy	Examine bladder with cystoscope	67–90	83	[13]
Cytology	Examine urothelial cells under microscope	48 (16–84)	86	[17]
NMP22 ELISA	Detect NMP22 protein	69	88	[19]
NMP22 BladderChek	Detect NMP22 protein	58	88	[19]
BTA-Stat	Detect human complement factor H-related protein	56–83	72–86	[20,21]
BTA-Trak	Detect human complement factor H-related protein	66–77	69	[6,23]
Urovysion	FISH chromosome abnormalities	72	83	[25]
ImmunoCyt	Immunofluorescence antigen detection	73	66	[32]
Xpert BC Assay	5 mRNAs detection	78	84	[35]
CxBladder	5 mRNAs detection	90	91	[39,40]
ADXBLADDER	MCM5 proteins detection	45–73	70–88	[43–45]
Bladder EpiCheck	DNA methylation patterns	90	83	[48]
Uromonitor	Detect genetic mutations	93	85	[54]
Assure MDx	Detect genetic mutations and methylation patterns	97	93	[55]
UBC test	Cytokeratin detection	30–87	91	[59]

This review aims to highlight the development and implementation of novel non-diagnostic tools in the evaluation of bladder cancer and to delineate future methods that may play a role in the evolving landscape of bladder cancer diagnosis.

Non-Invasive Bladder Cancer Diagnostic Tests

Cytology

Since 1945, urine cytology has served as an easy and reliable non-invasive measure of bladder cancer, with a specificity found to be as high as 95%.17 However, it is limited by its low sensitivity to low-grade tumors. Yafi et al found a sensitivity of 84% for high-grade tumors, but only 14% for low-grade tumors.17 Interpretation is highly user dependent and can be difficult to distinguish based on the specimen quality such as if the tissue environment is heavily inflamed.18

Due to these limitations, many have advocated the use of complementary urinary biomarkers using immunofluorescence and protein expression.18 Yafi et al found improvement in cytology sensitivity through combinations with urinary biomarkers including the Hemoglobin Dipstick, BTA Stat, NMP22 BladderChek, and ImmunoCyt. Of these combinations, the optimal combination was cytology with the NMP22 BladderChek, yielding a sensitivity of 94% and specificity of 84% for high-grade tumors and 31% sensitivity for low-grade tumors.17

NMP22 BladderChek

Nuclear matrix proteins (NMPs) provide support for the cell nucleus. One protein, NMP22, has been found to be overexpressed in urothelial tumors and released into the urine following apoptosis of tumor cells.16 Consequently, it has been found to be up to 25 times greater in bladder cancer cell lines than normal urothelium.17 Thus, it can presumably be used for the diagnosis and subsequent surveillance of bladder cancer.

There are two FDA-approved tests, including the NMP22 Bladder Cancer ELISA and NMP22 BladderChek tests, which utilize this marker. The ELISA, or quantitative test, is performed in a lab, while the BladderChek, or qualitative test, can be done as a point-of-care (POC) assay. Both analyze voided urine for NMP22 markers. In 2015, Chou et al performed a meta-analysis and found an overall specificity of 88% and sensitivity of 69% for the ELISA test and overall specificity of 88% and sensitivity of 58% for the BladderChek test.19 Based on the tumor type, Yafi et al found BladderChek to have a sensitivity of 25% for low grade and 92% for high-grade tumors.17 Bladderchek could be useful clinically to distinguish which patients will need cystoscopy.6 Unfortunately, it has been found to have a fairly high false-positive rate for urinary tract infections (UTIs), calculi, foreign bodies, and other genitourinary cancers.13

Bladder Tumor Antigen Assays

Bladder tumor antigen (BTA) assays involve the use of immunoassays that detect human complement factor H-related protein in urine.16 This protein is released by cancerous cells to interfere with the complement cascade, providing survival advantage for the tumors. Currently, there are two FDA-approved assays, which analyze BTA quantitatively (BTA-Stat) and qualitatively (BTA-Trak).

BTA-Stat is a POC assay, which provides results in 5 minutes, while BTA-Trak is more specialized and consequently takes longer to result. Both tests are relatively easy to perform.9 BTA-stat’s overall specificity is 72–85.7% and sensitivity is 56–83%.20,21 For high-grade tumors, BTA-stat’s sensitivity was found to be around 64–69%.22 Similarly, BTA-Trak’s overall specificity is 69% and sensitivity is 66% with a higher sensitivity of approximately 77% for high-grade tumors.6,23 In a meta-analysis of 13 studies conducted by Guo et al, BTA-stat was found to be superior to cytology only in sensitivity (67% vs 43%), whereas specificity was inferior to cytology.22 Few clinical studies have used BTA-Trak as it requires more equipment and is not as quick. For that reason, BTA-Stat is preferred over BTA-Trak. Overall, BTA assays have been shown to be more sensitive than cytology but are limited by their lack of specificity and are only used as an adjuvant in clinical practice.

Urovysion

The Urovysion test is a multitarget, multicolor fluorescence in situ hybridization (FISH) assay. Its detection is based on specific chromosomal abnormalities on chromosomes 3, 7, and 17 found on exfoliated urothelial cells.6 In urothelial cancers, these exfoliated urothelial cells are seen in higher frequency.24 The assay incorporates multiple probes to specific chromosomal abnormalities to increase its sensitivity.24 It is the most expensive FDA-approved test for the diagnosis and surveillance of bladder cancer recurrence.

Through a pooled meta-analysis of 2477 FISH tests, Urovysion was found to have an overall sensitivity of 72% and an overall specificity of 83%. However, a sensitivity for low-grade cancer was found to be as low as 41%.25 Although hematuria and inflammation do not typically alter the test’s reliability, false positives are detected by the presence of umbrella cells, chromosome tetraploidy, or heteroploidy, which may be due to human polyomavirus infection.24

Although Urovysion appears to be superior to cytology, there have been mixed results. In a study conducted by Lavery et al, voided urine sample sensitivities of Urovysion were compared to urine cytology; urine cytology outperformed Urovysion in both low-grade and high-grade tumor cells.26 The authors asserted that a possible explanation for these results was that all urine samples were taken after bladder washout.26 This resulted in a higher yield of exfoliated cells and allowed optimal evaluation of abnormal cells in urine cytology, which in other studies was limited. Conversely, a study by Dimashkieh et al found Urovysion sensitivity was superior (61.9%) to urine cytology (29.1%), with the caveat of Urovysion generating more false positives.27 Other studies have suggested that the combination of FISH technology with cytology has improved its overall sensitivity.28 With the addition of FISH, the number of unwanted biopsies could be reduced. A positive FISH was found to predict recurrence and progression in patients who had a negative cystoscopy with abnormal cytology.29 More studies are warranted to further establish guidelines and validate FISH as a tool for bladder cancer screening.

ImmunoCyt

ImmunoCyt utilizes immunofluorescence with three monoclonal antibodies to detect antigens M344, LDQ10, and 19A11.30 These antigens are highly expressed on cancerous urothelial cells detected in voided urine.31 ImmunoCyt is FDA-approved for bladder cancer surveillance. Multiple studies have shown it to improve the sensitivity of urine cytology; however, the test lacks the specificity. Specifically, sensitivities and specificities range from 68% to 85% and 72% to 82%, respectively.19

In a meta-analysis of seven articles, ImmunoCyt had a higher pooled sensitivity of 72.5% compared to urine cytology at 56.6%.32 However, ImmunoCyt lacked the specificity with 65.7% compared to 90.6% for cytology.32 It is less affected by hematuria and inflammation compared to other assays, but can be influenced by the presence of UTIs, urolithiasis, BPH.31 Additionally, the technology has high interobserver variability and requires cytopathologists for appropriate and accurate implementation.33 Thus, its use in clinical practice has been limited.

Xpert Bladder Cancer Detection Assay

Xpert Bladder Cancer Detection Assay (Xpert BC) works by quantitating mRNA targets expressed on bladder cancer cells, more specifically ABL1, ANXA10, UPK1B, CRH and IGF2.34 Urine samples are identified as either positive or negative based on a linear regression model that considers the concentration of the markers in a urine sample of patients. Largely expressed quantities in the urine sample are considered a positive test, whereas less expressed quantities would be negative.34

In a large study comparing Xpert BC Monitor to cytology and Urovysion using pre-cystoscopy voided urine samples, Xpert BC was found to have the highest overall sensitivity at 78% compared to 44% (urine cytology) and 59% (Urovysion).35 Although its sensitivity is superior, it lacks specificity. It was found to have an overall specificity of 84% compared to 97% for cytology and 88% for Urovysion.35 When compared to low-grade tumors, Xpert BC was found to outperform cytology in sensitivity at 77% compared to 13%.36 In a recent study conducted by Cancel-Tassin et al, 500 patient-voided urine samples with a previous NMIBC diagnosis and a positive cystoscopy or CT urogram were analyzed by Xpert BC and cytology.37 Although specificity was found to be higher for cytology (73% vs 98%), the Xpert BC Monitor had a NPV of 99.7% for exclusion of aggressive tumors.37 In a recent meta-analysis, Laukhtina et al evaluated studies utilizing Xpert BC for diagnosis of recurrence during NMIBC follow-up. Overall, ten studies had a pooled sensitivity, specificity, and NPV of 72%, 76%, and 92% respectively, although there was significant heterogeneity among the studies. In a subgroup analysis, Xpert BC demonstrated similar diagnostic detection for high-grade recurrence to those in the overall population.38

CxBladder

Similar to Xpert BC, CxBladder is another novel test that measures the expression of five mRNAs, CDK1, CXCR2, HOXA12, IGFBP4, and MDK in voided urine.29 There are multiple assays available for bladder detection and surveillance. The CxBladder Detect is performed alongside cystoscopy and utilizes genomes to determine the detection of bladder cancer. CxBladder Triage is similar but incorporates age, gender, and smoking history to exclude bladder cancer for low-risk patients presenting with hematuria. CxBladder Monitor can be used for surveillance to prevent recurrence of NMIBC.9

O’Sullivan et al analyzed voided urine samples of patients with hematuria after cystoscopy.39 The samples were analyzed by multiple non-invasive assays including cytology, NMP22 Bladderchek and CxBladder. CxBladder was found to have 90% sensitivity and 91% specificity, while cytology had a sensitivity of 56% and specificity of 94%.39 A larger study conducted by Lotan et al found similar results; urine samples were collected from patients undergoing clinical surveillance for bladder cancer. CxBladder was found to outperform both NMP22 Bladderchek and cytology.40 The CxBladder had sensitivity of 91% compared to 22% by cytology. NPV was much higher (96%) compared to cytology (87%).40 A surveillance study with 309 patients conducted by Koya et al found that the addition of CxBladder Monitor reduced the number of annual cystoscopies by 39%.41 This greatly reduced patient anxiety and discomfort but kept the same detection rates.41 Thus, studies have shown the potential of CxBladder’s use in clinical practice for bladder cancer diagnosis and screening.

Adxbladder

The ADXBLADDER is an ELISA test, which uses antibodies to detect minichromosome maintenance protein 5 (MMP5) found in voided urine.42 MCM5 proteins are important in initiating DNA replication and have been found to be highly expressed in proliferating and cancerous cells.42 In the urothelium, it was found that normal epithelium has a low or absent expression of MMP5. The amount of MMP5 found in voided urine has been found to be correlated to the grade of the cancer, with more MMP5 indicating higher grade.42

The ADXBLADDER test is inexpensive, quick, and convenient, proving itself to be potentially useful in a clinical setting. However, it is relatively new and little research has been performed evaluating its efficacy. There has been broad variability among multiple studies, with its sensitivity ranging from 45% to 73% and specificity from 62% to 88%.38,43–45 The sensitivity and specificity for detection of high-grade tumor was found to be 71% and 76%, respectively, in a meta-analysis of 3 studies.38 ADXBLADDER may be a good addition to the non-invasive tests used with cytology in order to increase the sensitivity, however since the specificity is still lower than that of cytology, it would not be recommended to be used alone.

Epicheck

DNA methylation plays a role in regulating gene expression without changing the DNA code. Several studies have highlighted the presence of methylated loci in the context of bladder cancer, indicating its potential application as a diagnostic and prognostic biomarker.46,47 The Bladder EpiCheck test was recently developed for the surveillance of bladder cancer recurrence. The analysis is based on the detection of the DNA methylation status of 15 genomic loci, which are strongly associated with bladder cancer in specimens of voided urine.48 The test provides a value between 0 and 100; this value, also known as the EpiScore, is based on the methylation patterns present, where a positive score (>60) indicates that methylation patterns match NMIBC.48

Laukhtina et al performed a meta-analysis on five studies using Epicheck for recurrence and found a pooled sensitivity, specificity and NPV of 74%, 84%, and 94%, respectively. For high-grade tumors, the NPV was the same; however, the sensitivity was 80% and specificity was 78%.38 In a validation study conducted by Wasserstrom et al, Epicheck was found to have a higher sensitivity of 90%.49 However, the sensitivity varied on staging and grading of tumors, with a higher sensitivity found in higher stages and higher grades. The overall specificity was 83% and the NPV was 97%.49 Compared to urine cytology, Epicheck showed a greater sensitivity, 90% vs 38%, in both low- and high-grade tumors, but a lower specificity, 83% vs 96%.49

In a blinded prospective multicenter study conducted by Witjes et al, Bladder Epicheck and urine cytology were performed on urine samples collected from patients undergoing standard cystoscopy in an outpatient urology clinic. The study concluded that when excluding low-grade Ta tumors, the Epicheck test had a sensitivity of 91.7%, NPV of 99.3%, and a high specificity of 88.0%.50 In a recent blinded clinical prospective trial, Cochetti et al assessed the diagnostic performance of Epicheck for surveillance of high-risk bladder cancer compared to photodynamic diagnosis (PDD)-guided cystoscopy.51 Compared to urine cytology, Bladder Epicheck had a higher sensitivity (100% vs 88.9%) but a lower specificity (90.9% vs 100%). PDD-guided cystoscopy had the lowest sensitivity and specificity (61% vs 41% respectively). Bladder Epicheck had the highest area under the curve (AUC) compared to both PDD-guided cystoscopy and cytology (0.95 vs 0.51 vs 0.94). The authors suggested a potential algorithm to combine Epicheck with urine cytology and found that it would have predicted the correct diagnosis 90% of the time and would subsequently reduce the number of cystoscopies.51 Thus, Bladder Epicheck has the potential to be used in clinical practice as a tool to rule-out recurrence and therefore potentially avoid unnecessary cystoscopies.

Uromonitor and AssureMDx

Alterations in specific genes are known to drive cancerous growth. Uromonitor is a urine-based assay that detects hotspot gene mutations in telomerase reverse transcriptase (TERT) and FGFR3. TERT maintains telomere integrity; mutations have been upregulated in many forms of bladder cancer and may play an important role in carcinogenesis. Specifically, there is molecular evidence suggesting the involvement of hTR, HTERT, and CSK2 gene expression in bladder cancer carcinogenesis.52 Many FGFR3 mutations have been found in bladder cancers, with a higher number in NMIBC.53 Sieverink et al compared Uromonitor-V2 to cytology and found a superior sensitivity (93% vs 26%) but slightly lower specificity (85% to 91%). PPV (79% vs 63%) and NPV (95% vs 68%) were both higher in Uromonitor-V2 than cytology.54 The high NPV suggests that the Uromonitor may be beneficial in detecting recurrence in NMIBC patients. In a network meta-analysis, Uromonitor was found to be significantly higher than other urinary biomarker tests for detection of NMIBC recurrence. It was superior in sensitivity, PPV, and NPV; however, cytology remained superior in specificity.38 Another commercially available test similar to Uromonitor, AssureMDx, detects the TERT, FGFR3, HRAS with methylation analysis of OTX1, ONECUT2, and TWIST1. When combined with age, AssureMDx has a sensitivity of 97% and specificity of 93% in a multicenter cohort study.55 Further evaluation of the Uromonitor and AssureMDx is warranted to evaluate its use in clinical practice.

Future Directions

Single-Cell Sequencing Enabled Hexokinase 2 Assay

Single-cell RNA sequencing (scRNA-seq) is a novel tool used to analyze targeted cells and to study the tumor microenvironment. Conventional RNA sequencing works by analyzing the transcriptomic information collected in bulk from a mixture of different cells from the same cancer tissue.56 However, the information extracted from bulk RNA sequencing collects from other regularly expressed cells such as fibroblasts, endothelial and immune cells. scRNA-seq allows the analysis of a select population of cancerous cells, thereby providing more tumor-specific data.56 A promising use of this innovative technology is its incorporation into bladder cancer screening as it allows for the identification of specific transcription-related information to cancerous cells.

The Hexokinase 2 (HK2) assay is an example of the potential of this technology. A recent study by Wang et al analyzed the potential of scRNA-seq on exfoliated urothelial cells from a voided urine sample.57 HK2 was chosen as a biomarker for high-throughput screening of cells in urine and detecting exfoliated tumor cells showing elevated glycolysis, a tumor-specific property. The sensitivity, specificity, PPV, and NPV of the assay was 90%, 88%, 83% and 93%, respectively.57

UBC Assay

There are other biomarkers under investigation for the detection of bladder cancer. Cytokeratins, proteins that make up the cytoskeleton of epithelial cells, are released into urine after cell death. Of these proteins, cytokeratins 8, 18, 19 and 20 have been found to have a positive association with bladder cancer.58 One test, the Urinary Bladder Cancer (UBC) ELISA assay, was developed to detect the presence of the cytokeratin fragments 8 and 17 in voided urine.59 This test was developed as a POC assay to rapidly identify cancerous cells. Sensitivity for CIS was found to be 87%, 30% for low-grade tumors, and 72% for high-grade tumors with the specificity 91%.59

miRNA

MicroRNAs (miRNAs) are small non-coding RNAs that regulate gene expression post-transcriptionally through interactions with complementary target sites on mRNAs.60 They can either induce mRNA degradation or impair its translation. Therefore, altered miRNA expression can lead to carcinogenesis, progression, and metastases.60–62 Various studies have discovered miRNAs that are unique to bladder cancer. Puerta-Gil et al analyzed bladder cancer urinary samples for expression of miR-143, miR-222, and miR-452 to discover how their expression correlated with pathogenesis. miR-222 and miR-452 were found to significantly increase with tumor grade, tumor size, and presence of CIS, indicating their potential role in bladder cancer diagnostics. miR-222 along with miR-143 expression correlated with clinical recurrence and progression. In addition, miR-452 expression was highly correlated with lymph node metastasis.60 Urquidi et al investigated a miRNA biomarker panel using voided urine samples from 85 subjects with known bladder cancer; in their study, they identified a 25-target diagnostic signature that could predict bladder cancer with an estimated sensitivity of 87% and specificity of 100%.63 Despite its potential utility in bladder cancer diagnosis, further investigation is warranted to validate miRNA’s role as a non-invasive bladder cancer biomarker and its use in the clinical setting.

Extracellular Vesicles

Novel agents investigated for cancer detection include exosomes, microvesicles and extracellular vesicles, which are secreted by cells and may play a role in tumor progression. They are composed of a small lipid bilayer, which allows tumor components to avoid degradation by the circulation and provides protection from the extracellular space.64

Bladder tumors secrete specific exosomes that may be elevated in NMIBC, allowing them to potentially serve as non-invasive biomarkers for NMIBC through ELISA detection. The level of circulating exosomes correlates with tumor progression. The exosomes allow the cancer cells to interact with the surrounding tumor microenvironment and thus affect angiogenesis, invasion, immune response, and metastasis.64 Several exosomal miRNAs detected in urine of bladder cancer patients were found to be correlated with disease. They are hypothesized to act as cancer messengers and have shown potential promise as bladder cancer biomarkers.65 Elsharkawi et al evaluated ELISA detection of CD9 on exosomes in urine and serum and found that specificity was 100% in serum and 83.3% in urine, whereas sensitivity was 82.4% in serum and 92.6% in urine.64 Further investigation of tumor exosomes and microvesicles is needed to determine their effectiveness in NMIBC detection.

Bladder Cancer Detection Device

Various bladder cancer detection devices have been explored to improve diagnostics. One such device, the Microfluidic Urinary Photo-Specific Diagnostic of bladder cancer (MicroUPSD), relies on an immunoaffinity microfluidic platform to detect bladder epithelial cells. MacGregor et al covalently immobilized bioactive anti-epithelial cell adhesion molecule (EpCAM) antibodies to microchannels to selectively capture bladder cancer cells that have a greater expression of EpCAM in excreted urine.66 They evaluated centrifuged and settled urine samples with the device and compared it to standard cytology and cystoscopy. The device was found to have a 100% sensitivity in both centrifuged and settled urine samples, compared to 20% in cytology. The specificity for the settled urine was 80% compared to 100% in cytology.66 The group found a specificity higher for settled urine compared to centrifuged urine, indicating the device could potentially be used as a POC assay outside of pathology laboratories. A potential drawback to this device is that EpCAM is not bladder cancer specific and thus can lead to false positives from other cancers such as prostate cancer.66

Artificial Intelligence

With recent technological advancements, methods utilizing artificial intelligence (AI) have been proposed to aid in bladder cancer detection. Algorithms have been created to detect subtle changes in the bladder based on cystoscopic images and videos. Six studies created their own specific AI algorithms and were found to have a pooled sensitivity of 89.7% and specificity of 96.1% for detection of bladder cancer.67 Seemingly, the ideal scenario includes incorporating this technology along with real-time cystoscopy in order to optimize the detection of bladder cancer and to potentially reduce the need for subsequent cystoscopies.67 In addition, incorporation of AI may assist physicians-in-training in detection of bladder cancer. Wu et al found that their AI Algorithm was superior to expert urologists in the detection of complex lesions such as CIS and very small lesions.68 Further prospective studies are needed to determine incorporation of AI in clinical practice and its use in different patient populations.

Expert Opinion

With the increasing incidence of bladder cancer and recent technological advancements in its detection, urinary biomarkers hold great promise for future utilization in clinical practice. Compared to the standard screening method of cystoscopy, urinary biomarkers may greatly reduce the burden and stress placed on patients. Laukhtina et al calculated the number of cystoscopies avoided (true negatives and false negatives) and risk of missing recurrences by avoiding cystoscopy (true negative) for detection of NMIBC recurrence using urinary biomarkers; they found that urinary biomarkers could avoid 500–740 cystoscopies and would miss recurrence in 10–78 patients per 1000 patients.38 Furthermore, in addition to reducing mental stress for the patient, urinary biomarker utilization may decrease the financial burden imposed on patients during bladder cancer diagnosis. WLC alone costs around US$210, WLC with biopsy costs approximately US$370, and cytology costs $100 USD.69–71 Common urinary biomarkers are currently available ranging from US$25 to US$80.71,72 However, novel tests are less cost-effective, with Bladder Epicheck ranging from US$168 to US$476 and Urovysion estimated to cost approximately $800 USD.71,73 Many tests are not yet mass produced and may have a lower cost once used more robustly. However, it may be difficult to determine the true financial burden as great variability exists among countries with respect to product cost.

Most urinary biomarkers show superiority to urine cytology and have the potential to be used in clinical practice for the diagnosis and surveillance of NMIBC. Based on the specific patient presentation, a unique biomarker kit or panel combination may be indicated, allowing for physicians to provide more patient-centered care. Due to limited studies, it remains to be seen whether urinary biomarkers could be incorporated into screening for initial diagnosis of bladder cancer. Urinary biomarkers are on the horizon for detection of high-grade tumors, but further research is warranted to assess their detection of low-grade and CIS tumors. Unfortunately, there are few prospective studies investigating urinary biomarkers in low-grade NMIBC tumors and the available cross-sectional studies seem to overestimate the sensitivity of biomarkers.74

Cell-based biomarkers such as ImmunoCyt and Urovysion appear to have higher sensitivities and specificities in low-grade NMIBC compared to urine cytology and markers analyzing soluble tumor-associated antigens.74 Due to these findings, a prospective multi-center randomized study, UroFollow, was initiated to investigate whether non-invasive biomarkers could be used for follow-up for patients with low and intermediate grade NMIBC over a 3-year period.74 Currently, though urinary biomarkers may not yet be ready to replace cystoscopy, they show promise in conjugation with current practices for monitoring bladder cancer recurrence.

Conclusion

Bladder cancer remains a complex clinical entity to diagnose, monitor and treat. Specifically, the ability to diagnose bladder cancer in a reliable, reproducible way represents a particularly challenging subject. Recently, there has been interest in incorporating non-invasive means of bladder cancer detection and surveillance, as well as in improving surveillance of low-grade lesions with modalities that enhance tumor detection. This review highlights novel means of detecting bladder cancer. With the adoption of non-invasive methodologies, patients may avoid the potential physical, mental, and financial burden of repeated cystoscopy procedures. Therefore, non-invasive methodologies may not only benefit patients but also health care systems and physicians.