This contractor evaluated peer-reviewed, evidence-based literature reporting the clinical utility of UBBs in MH using the GRADE (Grading of Recommendations Assessment, Development and Evaluation) methodology.2 The PICO parameters documented below were prespecified to ensure transparency and reproducibility:
Population: Patients with hematuria undergoing urologic evaluations
Intervention: Use of urine biomarkers to assess risk of underlying urinary tract malignancy
Comparator(s): Guideline-directed cystoscopy
Outcome(s): Ability of the biomarker to detect urothelial cancer with high accuracy to reduce cystoscopy
Search Protocol
Studies were identified by searching PubMed for peer-reviewed literature using search terms including:
- (bladder cancer AND biomarkers)
- (urinary biomarker OR urine biomarker)
- (hematuria AND biomarker AND protein AND bladder)
- (fluorescence in situ hybridization OR FISH OR UroVysion) AND (microhematuria OR hematuria OR gross hematuria OR gross haematuria OR microhaematuria OR haematuria)
- (gene expression profile OR Cxbladder OR Xpert Bladder Cancer Detection) AND (microhematuria OR hematuria OR gross hematuria OR gross haematuria OR microhaematuria OR haematuria)
- (antibody detection test OR NMP22 BladderChek) AND (microhematuria OR hematuria OR gross hematuria OR gross haematuria OR microhaematuria OR haematuria)
- (nomograms AND hematuria)
Google and Google Scholar were also used to search for society guidelines and recommendations. Initial searches were limited to studies published within ten years of the search date, but if those searches didn’t identify literature the ten-year limit was removed. Additional literature was identified within the bibliographies of included studies.
Summation of Discovered Evidence and Analysis Incorporation
Sixty-one papers were identified and included for review in this LCD: 36 observational, one randomized controlled trial, three systematic reviews without meta-analysis, six meta-analyses, seven guideline documents, and eight other publication types such as assay development studies and non-systematic reviews.
Background
Hematuria, defined as the presence of blood in the urine, is a common reason for urologic evaluation and accounts for a substantial proportion of outpatient urologic visits.3 It is categorized as either gross (GH), in which blood is visible in the urine, or microscopic (MH), defined as ≥3 red blood cells per high‑power field (RBC/HPF) on a single urinalysis.4,5 MH may be identified incidentally or in symptomatic settings and can occur with or without accompanying proteinuria. The differential diagnosis for hematuria is broad and includes benign prostatic hyperplasia, urinary tract infection, nephrolithiasis, urethral stricture disease, trauma, urologic procedures, medication effects, and malignancies of the urinary tract. Accurate identification of the underlying cause is essential to ensure appropriate clinical management.
Bladder cancer (BC) is a clinically significant cause of hematuria. Approximately 13% of patients with GH and 3% of those with MH are ultimately diagnosed with BC.1 Urothelial carcinoma (UC) (formally transitional cell carcinoma, TCC) is the predominant histologic subtype and includes both non-muscle invasive (NMIBC) and muscle-invasive (MIBC) disease. Diagnosis is confirmed by histologic evaluation of biopsy tissue obtained via cystoscopy. Evidence-based guidelines support risk-stratified evaluation of hematuria to identify patients who warrant further diagnostic assessment for BC. UBBs have been developed to aid in detection and risk-stratification.
Society Guidelines and Systematic Reviews
NCCN (National Comprehensive Cancer Network)
Although the NCCN does not have guidelines specifically related to hematuria, the guidelines do address the use of biomarkers stating that they have a limited role in the initial evaluation and management of BC.6 Patients suspected of having BC are recommended to undergo cystoscopy, TURBT (transurethral resection of bladder tumor), and EUA (examination under anesthesia) for diagnosis and disease risk-stratification to guide treatment options.
EAU (European Association of Urology)
In 2024, the EAU revised its guidelines on NMIBC; while these updates do not directly address MH, they offer relevant information on the use of UBBs in BC management.7 In the evaluation of patients presenting with hematuria or other symptoms suggestive of BC, it is widely recognized that no currently available test can replace cystoscopy for primary detection. Nonetheless, urinary cytology or biomarkers may serve as useful adjuncts, particularly for identifying tumors that cystoscopy may miss, such as carcinoma in situ.
AUA (American Urological Association)
The American Urological Association (AUA) has shifted from a universal evaluation strategy for MH to a risk-stratified model based on accumulating evidence that many patients have a very low‑likelihood of malignancy. Earlier 2012 guidance recommended cystoscopy and CT urography for most adults with MH, but subsequent studies demonstrated that this approach exposed large numbers of low-risk individuals to unnecessary radiation and invasive procedures despite minimal clinical yield.4,8,9 In 2020 the AUA guidelines were updated to introduce patient-provider shared decision making and risk-stratification.5 Patients with MH related to possible malignant genitourinary conditions were stratified into low-, intermediate-, or high-risk for genitourinary malignancy based on several factors including, but not limited to, age, sex, smoking use, degree and persistence of MH, past medical history, and clinical examination. The guidelines briefly discussed biomarkers and stated that while there was a potential role in the future for UBBs in the detection of BC in hematuria patients, there was insufficient evidence to recommend the use of these markers routinely. A 2025 update further refined this framework by limiting low or negligible‑risk management to repeat urinalysis within six months, reclassifying persistent MH into higher‑risk categories, and allowing intermediate‑risk patients who wish to avoid cystoscopy to use urine cytology or validated urinary biomarkers, although renal and bladder ultrasound remains required. High risk patients continue to warrant cystoscopy and axial imaging.1
Analytic and Clinical Validity
Urine cytology
Urine cytology is the earliest and most widely established noninvasive test for urothelial tumor detection in patients with hematuria.10,11 It relies on microscopic evaluation of exfoliated urothelial cells obtained from voided or instrumented urine, typically interpreted by a cytopathologist, with malignant changes identified on the basis of cellular morphology. Across observational studies and multiple systematic and nonsystematic reviews assessing noninvasive UBBs for BC detection and surveillance, urine cytology consistently demonstrates high specificity (82.8%–99.2%) but variable and often limited sensitivity particularly for low-grade UC (15.8%–78.7%).10-13 However, despite its limitations, urine cytology continues to serve as a benchmark against which newer biomarker tests are compared.
Cell-based and Single Analyte UBBs
Several single analyte urinary biomarkers have been developed in an effort to improve upon the limitations of urine cytology, particularly its low sensitivity for low‑grade UC. Although many of these assays received FDA clearance and remain commercially available, none are guideline-endorsed as standalone diagnostic tests for hematuria evaluation. Across observational studies and systematic reviews, their performance generally reflects a consistent pattern: improved sensitivity relative to cytology but reduced specificity.
Protein based assays such as BTA and NMP22 illustrate this pattern most clearly. In hematuria cohorts, BTA sensitivity is commonly ~60-70%, but specificity can drop to ~55-70%, largely due to false positives related to benign urinary factors.14 NMP22 demonstrates a similar profile, with sensitivity around 60-75% and specificity ~58-86%, but often markedly lower in the presence of benign abnormalities.15,16
Cell-based tests achieve somewhat more consistent specificity but still follow the broader sensitivity-specificity compromise. ImmunoCyt/uCyt+ generally demonstrates sensitivity of ~80-88% and specificity of ~75-83%, while UroVysion FISH shows sensitivity of ~64-75% with specificity ~85-95%. Systematic reviews consistently conclude that neither assay reliably outperforms cytology nor provides sufficient accuracy to support routine use in the initial evaluation of hematuria.15-17
Newer assays such as ADXBLADDER report sensitivity in the ~70-75% range and specificity ~68-72%, with systematic reviews emphasizing significant heterogeneity, high false positive rates, and insufficient evidence to support its use as a primary diagnostic tool.18,19
Across all single analyte biomarkers, evidence consistently shows improvements in sensitivity, but reduced specificity compared to cytology. Performance is frequently compromised by benign urinary abnormalities, limiting positive predictive value (PPV) and decreasing reliability in the initial evaluation of hematuria.
Multi-analyte UBBs
Gene Expression Profiles
Gene expression profiles (GEPs) analyze the activity of genes by measuring the relative amount of RNA expressed under various conditions and have been leveraged in the development of urine-based assays to identify the presence of biomarkers indicative of BC in urine. Early work by Holyoake et al identified a gene panel capable of distinguishing UC from controls, with sensitivity ranging from 48% for Ta tumors to 100% for ≥T1 disease at approximately 85% specificity.20 Candidates were chosen from a previously reported microarray expression dataset of pTa and pT1 tumors from a cohort of 27 Japanese patients with intravesical recurrence and were further evaluated in Ta-T4 tumors and normal urothelium using qRT-PCR.21 A four-gene panel of CDC2, MDK, IGFBP5, and HOXA13 (defined as uRNA-D) was advanced for further study based on the ability to reproducibly quantify urinary expression by qRT-PCR and to mitigate confounding signal from blood-derived or inflammatory cells.20 Subsequent iterations incorporated CXCR2 to mitigate inflammation‑associated false positives (Cxbladder Detect; CxbD) and later combined molecular signals with clinical factors such as age, sex, smoking history, and hematuria characteristics to create Cxbladder Triage (CxbT).22,23 These additions improved sensitivity and negative predictive value (NPV) but reduced specificity because of the broader variable set included in the predictive model. In a hematuria cohort of 485 patients with 66 UC cases, CxbD demonstrated 73% sensitivity (68% for Ta) at 85% specificity, with a 10% sample‑failure rate.22 Using combined genetic and clinical variables, CxbT achieved sensitivity of 95% and NPV of 98%, though specificity was limited (45%).23 Analytical validation studies reported sensitivities of 77% and specificities of 94% for CxbD, while CxbT demonstrated sensitivity of 95% and NPV of 98% with more modest specificity.24
Multiple studies have evaluated the clinical application of Cxbladder tests in the assessment of hematuria. 25-30 Davidson et al evaluated CxbT within a clinical workflow that included renal ultrasound or CT imaging in 884 patients with hematuria (566 GH; 318 MH).25 When used alongside imaging, one low‑grade pTa tumor was initially missed but detected three months later on follow‑up cystoscopy after persistent hematuria. Lotan et al prospectively evaluated 390 MH patients, classifying them as lower-risk (n=135) or not lower-risk (n=255) based on criteria of: ≤10 pack‑years of smoking and 3–29 RBC/HPF.27 Lower‑risk patients were offered CxbT instead of cystoscopy, while 82% of not‑lower‑risk patients underwent cystoscopy, with tumors detected in ~10%. CxbT results were available in 81 lower‑risk patients; 87.7% tested negative, and 80.3% of these deferred cystoscopies. One Ta tumor was identified at 13‑month follow‑up after a repeat positive CxbT. Overall, stratification of patients using CxbT reduced the rate of cystoscopy from 67% to 27% with a sensitivity of 90% and an NPV of 99%.
Filson et al reported on a cohort of over 3,000 patients in an integrated health system, comparing those who received CxbT with a matched untested cohort.29 After excluding patients with recent gross hematuria, prior BC, or age outside 35–88 years, individuals were stratified using the Hematuria Risk Index (HRI).31 Among those tested, 79.6% (2670/3353) were classified as low probability of UC; only 3.8% underwent cystoscopy compared with 46.5% of matched controls. In contrast, 73.4% of CxbT‑high‑risk patients underwent cystoscopy versus 45.7% of high‑risk controls. Despite these differences in procedure use, UC detection rates were similar between tested and untested cohorts (0.33% vs. 0.60%; P=0.105), as were high‑grade detection rates (0.22% vs. 0.34%; P=0.809). A smaller institutional study compared CxbT to the “triple workup” of cystoscopy, cytology, and upper tract imaging in a cohort of 258 patients which demonstrated a sensitivity of 92.9% and an NPV of 92.9% for the detection of UC consistent with previous studies though interpretation was limited by small case numbers.30
Wallace et al developed a five‑gene assay (ABL1, CRH, IGF2, ANXA10, and UPK1B) for BC detection based on differential gene expression identified in an initial cohort of 444 patients, including 125 with histologically confirmed BC.32 In a subsequent validation cohort of 370 patients, 85% with pTa tumors, the assay showed 83% sensitivity for high‑grade and 65% for low‑grade disease, with specificity of 90% in hematuria evaluations and 77% in surveillance. Higher‑than‑expected cancer prevalence (24% and 28%) required adjusted predictive values, and the mixture of diagnostic and surveillance patients limited generalizability.
Valenberg et al evaluated the commercial Xpert Bladder Cancer Detection (XpertBC‑D) in 828 patients, 46% with microhematuria, with overall cancer prevalence of 7% (11% GH; 3% MH).33 XpertBC‑D demonstrated higher sensitivity (78%) than UroVysion FISH (59%) and cytology (44%), with similar performance across hematuria types, but sensitivity for low‑grade tumors remained limited (55%). Specificity was 84%, lower than FISH (97%) and cytology (88%), and decreased to 72% in patients with nephrolithiasis, prompting concerns that positive results were insufficient for confirmation and could miss tumors in a rule‑out strategy.
A further real‑world study of 324 hematuria patients (213 MH; 111 GH) without prior UC identified 8 low‑grade and 20 high‑grade cancers.34 XpertBC‑D again showed higher sensitivity than cytology (96.4% vs. 60.7%) but lower specificity (80% vs. 86%). Gross hematuria, male sex, abnormal cytology, and a positive XpertBC‑D result were associated with increased UC risk, while smoking history and occupational exposure were not.
Methylation profiles
A methylation profile (MP) reflects disease‑specific patterns of DNA methylation.35 Numerous urine‑based methylation markers for BC have been described, but few assays have reached clinical use, and those that have are primarily validated in surveillance populations rather than in initial hematuria evaluations. The first commercially available, FDA cleared methylation-based UBB, Bladder EpiCheck (BE), measures methylation at 15 loci and reports an “EpiScore” with values ≥60 considered positive; the specific loci are not disclosed. Although FDA clearance is limited to surveillance of previously diagnosed NMIBC, the test is marketed internationally for both surveillance and initial evaluation.
In a multicenter prospective study of 353 patients undergoing 3‑month surveillance after incident or recurrent UC resection, BE demonstrated 68.2% sensitivity and 88% specificity, with an NPV of 95.1% and PPV of 44.8%.36 Sensitivity for low‑grade tumors was substantially lower; when low‑grade lesions were excluded, sensitivity rose to 91.7% with an NPV of 99.3%. The authors noted that missed low‑grade recurrences would likely be detected at the next scheduled cystoscopy. Other studies similarly reported high‑grade sensitivities of 78.9%–100% and NPVs >98%, with performance unaffected by age, sex, or interval from last recurrence.37
Trenti et al compared BE to Xpert BC in NMIBC surveillance.38 Sensitivities were similar (66.3% for Xpert, 64.13% for BE) and improved to 78.95% when limited to high‑grade tumors. When both assays were combined, sensitivity rose to 92.11%, though the combined NPV (92.24%) remained below that reported in earlier BE studies.
Piatti et al evaluated Bladder CARE, a urine-based test that quantifies DNA methylation at three loci (DXL6-AS1, WDR63, and GHSR) in a case-control cohort of patients with and without BC.39 Across all samples, the authors reported a 93.5% sensitivity, 92.6% specificity, 87.8% PPV, and 96.2% NPV. However, the stage and grade of tumor was not reported nor was the test assessed in patients presenting with hematuria.
Genomic profiles
BC molecular profiling has led to the identification of distinct cellular subtypes with distinct profiles and pathways. Several genes have been identified as having a high rate of recurrent hotspot mutations.40 For example, studies have shown that up to 75% of low-grade NMIBCs carry mutations in FGFR3, yet this frequency drops to less than 20% in MIBC.41 This is largely attributed to changes in the cellular subtype during the progression of NMIBC to MIBC.42 Mutations in the TERT promoter have also been reported to be highly prevalent in NMIBC, and multiple targeted assays that test both DNA extracted from exfoliated cells as well as cell free DNA (cfDNA) have been developed, albeit primary outside of the United States.43-45 Targeted hot spot panels that include additional genes commonly mutated in BC have also been developed.46-49 However, none of these assays have been rigorously validated for use in a routine clinical setting.
Protein profiles
Multiplex protein‑based signatures quantify coordinated changes in tumor‑associated proteins in urine. Oncuria is a UBB test that measures ten proteins. The assay was initially developed by combining proteins identified from gene expression signatures of UC with proteins known to be elevated in patients with BC. Early development work narrowed a 14-protein candidate pool to an 8 biomarker ELISA panel; in a cohort of 127 patients (64 with BC; stage/grade not reported), Goodison et al reported 92% sensitivity and 97% specificity for BC detection.50 Subsequent refinement and validation across multiple studies produced a finalized panel of ten proteins, which was converted into a multiplex bead-based immunoassay.51,52 Hirasawa et al evaluated this Oncuria assay in 362 patients (46 with BC).53 Among BC cases, 61.4% were NMIBC and 38.6% MIBC, with 80% high-grade. A predictive model incorporating age, sex, and race achieved 89% sensitivity for low-grade tumors and 94% for high-grade, with an NPV of 98% and PPV of 61%. Pagano et al subsequently evaluated the clinical performance of Oncuria Detect (OD), the commercial version, in a large multi-center real world study of 931 hematuria patients.54 A training set of 617 patients (121 with BC) and a test set of 383 patients (73 with BC) were used to lock and validate the algorithm. Whether patients presented with GH or MH was not reported. In the test set, OD demonstrated 87% sensitivity, 72% specificity, and a 96% NPV.
More recently, Heard et al evaluated OD alone and in combination with the Hematuria Cancer Risk Score (HCRS), a demographic based nomogram, in 365 patients undergoing evaluation for GH (n=82) or MH (n=283).55 Fifty-five patients were diagnosed with BC. HCRS alone demonstrated 90.9% sensitivity, 71.9% specificity, and an AUC of 0.890. OD alone showed 78.2% sensitivity, 76.8% specificity, and an AUC of 0.838. Combined, the tests produced 87.3% sensitivity, 80.6% specificity, and an AUC of 0.909. However, false negative rates were substantially higher in patients with gross hematuria, yielding sensitivities of 57.9% (HCRS), 73.7% (OD), and 68.4% (combined). As a result, the combined approach achieved an NPV of 98.7% in MH, but only 75% in GH. The authors noted that although the rate of BC in patients with MH was low (2.5%), 46% of patients with GH had BC, and that part of the study cohort overlapped with the original validation population, introducing potential study bias.
Combination profiles
Recent studies have demonstrated that combining biomarker tests results in an increase in performance. For example, the incorporation of six DNA SNPs in FGFR3 and TERT into CxbD and CxbT has demonstrated substantial improvements in diagnostic performance over earlier mRNA-only versions of the test.24,56,57 Lotan et al showed that the addition resulted in an increase in diagnostic performance in a mixed cohort of patients with both GH and MH, including improvements in sensitivity (74% to 97%) and specificity (82% to 90%) for CxbD, and increased specificity (63% to 78%) for CxbT, with negative predictive values approaching 99.7%.56 The enhanced tests missed three and two cancers in the pooled cohort respectively, four low-grade Ta tumors and one papillary urothelial neoplasm of low malignant potential (PUNLMP). Further evaluation by Savage et al in a multi-center cohort of 615 patients who received cystoscopy showed that CxbT+ ruled out 70.9% (416/587) of patients from further evaluation.57 Of those, 99.3% had normal cystoscopy missing two low-grade and one high-grade tumor(s).
UroDiag integrates the qualitative detection of four recurrent FGFR3 mutations with the quantitation of three methylation biomarkers (HS3ST2, SEPTIN9, SLIT2) into a single multiplex PCR based platform.58,59 In a validation cohort of 272 patients (167 NMIBC; 105 cystoscopy‑negative), the assay demonstrated 97.6% sensitivity, 84.8% specificity, AUC 0.96, and an estimated NPV of 99.6%.58 However, it has not been clinically evaluated in patients without a prior diagnosis.
A foundational 2016 study of 154 patients (74 newly diagnosed BC; 80 benign hematuria) evaluated a combined methylation panel (TWIST1, ONECUT2, OTX1) with a mutation panel (FGFR3, TERT, HRAS), yielding 97% sensitivity, 83% specificity, and >99% NPV.60 Limitations included small sample size, minimal comparator data, and no GH vs. MH stratification. A subsequent multi-center 2020 prospective validation study addressed these limitations in 1,005 consecutive hematuria patients, 838 of whom had complete biomarker results.61 In this cohort, the optimized diagnostic model incorporating hematuria type achieved 96% sensitivity, 73% specificity, and 99% NPV, with an AUC of 0.95, and detected all upper‑tract urothelial cancers. Performance differed substantially by hematuria type: among the 381 patients with microscopic hematuria, only 14 cancers were present and the assay produced 338 true‑negative results, enabling an 89% deferred cystoscopy rate; in contrast, among 457 patients with gross hematuria, where cancer prevalence was higher (98 cancers), the assay identified 191 true negatives, corresponding to a 42% deferred cystoscopy rate. When applied across the full cohort, use of the assay as a triage tool would have reduced cystoscopy utilization by 53% (529 of 1,005 patients), while maintaining high sensitivity and extremely high NPV, demonstrating substantial potential clinical utility, particularly in the low‑prevalence microscopic hematuria population.
Contractor Advisory Committee (CAC) Meeting Summary
To further understand provider perspectives on current practices for patients presenting with microhematuria, Novitas Solutions and First Coast Service Options, hosted a Contractor Advisory Committee (CAC) Meeting via webinar on Feb 19, 2026. The meeting was open to the public, and questions, bibliography and transcripts are publicly available. The purpose of the meeting was to obtain advice from multi-jurisdictional CAC members and Subject Matter Experts (SMEs) regarding the best practices, and strength of published evidence on the use of UBBs in patients with microhematuria. The SMEs were asked open-ended questions about the appropriate intended-use patient population, the various types of tests, the risks associated with avoiding cystoscopy and frequency of testing. The panelists were allowed to answer without interruption. CAC members and SMEs who were not selected for the panel were provided with an opportunity to submit comments in writing.
A majority of panelists and respondents aligned with the available literature and professional society guidelines, emphasizing that UBBs are most useful in intermediate‑risk patients and in those who refuse or cannot undergo cystoscopy due to anxiety, access limitations, or medical contraindications. Several panelists also highlighted that UBBs may help improve patient engagement, noting that patients who receive a positive biomarker result are often more willing to proceed with cystoscopy because the result provides a clearer rationale for invasive evaluation. Participants further emphasized that the use of UBBs may help prioritize limited urologic resources, reduce unnecessary imaging or endoscopy in low‑yield situations, and support more efficient triage within overburdened clinical systems. Some SMEs also underscored the need for improved care coordination, particularly in primary‑care‑driven evaluations, where follow through with specialist referral or cystoscopy may be inconsistent