|MP 2.04.07||Urinary Tumor Markers for Bladder Cancer|
|Original Policy Date
|Last Review Status/Date
Reviewed with literature search/7:2013
|Return to Medical Policy Index|
Our medical policies are designed for informational purposes only and are not an authorization, or an explanation of benefits, or a contract. Receipt of benefits is subject to satisfaction of all terms and conditions of the coverage. Medical technology is constantly changing, and we reserve the right to review and update our policies periodically.
The diagnosis of bladder cancer is generally made by cystoscopy and biopsy. Moreover, bladder cancer has a very high frequency of recurrence and therefore follow-up cystoscopy, along with urine cytology, is done periodically to identify recurrence early. Urine biomarkers that might be used to either supplement or supplant these tests have been actively investigated.
Urinary bladder cancer, a relatively common form of cancer in the United States, results in significant morbidity and mortality. Bladder cancer (urothelial carcinoma), typically presents as a tumor confined to the superficial mucosa of the bladder. The most frequent symptom of early bladder cancer is hematuria; however, urinary tract symptoms (i.e., urinary frequency, urgency, and dysuria) may also occur.
For patients with hematuria, American Urological Association (AUA) guidelines recommend cystoscopic evaluation of all adults older than age 40 years with microscopic hematuria and for those younger than age 40 years with risk factors for developing bladder cancer. Confirmatory diagnosis of bladder cancer is made by cystoscopic examination, considered to be the gold standard, and biopsy. At initial diagnosis, approximately 70% of patients have cancers confined to the epithelium or subepithelial connective tissue. Non-muscle invasive disease is usually treated with transurethral resection, with or without intravesical therapy, depending on depth of invasion and tumor grade. However, a 50-75% incidence of recurrence has been noted in these patients, with 10% to 15% progressing to muscle invasion over a 5-year period. Current follow-up protocols include flexible cystoscopy and urine cytology every 3 months for 1 to 3 years, every 6 months for an additional 2 to 3 years, and then annually thereafter, assuming no recurrence.
While urine cytology is a specific test (from 90–100%), its sensitivity is lower, ranging from 50–60% overall and is considered even lower for low-grade tumors. Therefore, interest has been reported in identifying tumor markers in voided urine that would provide a more sensitive and objective test for tumor recurrence.
Tests cleared by the U.S. Food and Drug Administration (FDA):
The BTA (bladder tumor antigen) stat® test, (Polymedco Inc., Cortlandt Manor, NY) is a qualitative, point-of-care test with an immediate result that identifies a human complement factor H-related protein that was shown to be produced by several human bladder cell lines but not by other epithelial cell lines.
The BTA stat® test is an in vitro immunoassay intended for the qualitative detection of bladder tumor-associated antigen in the urine of persons diagnosed with bladder cancer. The BTA TRAK® test (Polymedco Inc., Cortlandt Manor, NY) provides a quantitative determination of the same protein. This test requires trained personnel and a reference laboratory. Both tests have sensitivities comparable to that of cytology for high-grade tumors and better than cytology for low-grade tumors.
Nuclear matrix protein 22 (NMP-22) is a protein associated with the nuclear mitotic apparatus. It is thought that this protein is released from the nuclei of tumor cells during apoptosis. Normally, only very low levels of NMP-22 can be detected in the urine, and elevated levels may be associated with bladder cancer. NMP-22 may be detected in the urine using an immunoassay.
Fluorescence in situ hybridization (FISH) DNA probe technology has also been used to detect chromosomal abnormalities in voided urine to assist not only in bladder cancer surveillance but also in the initial identification of bladder cancer. FISH DNA probe technology is a technique to visualize nucleic acid sequences within cells by creating short sequences of fluorescently labeled, single-strand DNA, called probes, which match target sequences. The probes bind to complementary strands of DNA, allowing for identification of the location of the chromosomes targeted. UroVysion® (Vysis Inc., Downers Grove, IL) is a commercially available FISH test.
The ImmunoCyt™ test (DiagnoCure Inc., Quebec) uses fluorescence immunohistochemistry with antibodies to a mucin glycoprotein and a carcinoembryonic antigen (CEA). These antigens are found on bladder tumor cells. The test is used for monitoring bladder cancer in conjunction with cytology and cystoscopy.
In addition to the FDA-cleared tests, Predictive Biosciences (Lexington, MA) is marketing a urine-based test, called CertNDx™, to assess Fibroblast Growth Factor Receptor 3 (FGFR3) mutations. The test is intended to be used in combination with cytology for identifying patients with hematuria at risk of bladder cancer. It is being offered through Predictive Bioscience’s network of Clinical Laboratory Improvement Amendment (CLIA) laboratories. FGFR3 mutations may be associated with lower-grade bladder tumors that have a good prognosis.
Other urinary markers
A number of other urinary tumor markers, not currently commercially available in the United States, are under investigation. These include:
- BLCA-1 and BCLA-4;
- Hyaluronic acid and hyaluronidase;
- Lewis X antigen;
- Microsatellite markers;
- Soluble Fas;
- Survivin (can be isolated from urine and also from tumor samples);
- Cytokeratin 8, 18, 19, 20
Urinary tumor marker tests cleared by the FDA and in clinical use include:
- The quantitative BTA TRAK® and the qualitative point-of-care BTA (bladder tumor antigen) stat® test, both by Polymedco Inc., Cortlandt Manor, NY.
- The quantitative immunoassay NMP22® and the qualitative, point-of-care test NMP22® BladderChek®, both by Matritech Inc., Newton, MA.
- The UroVysion® Bladder Cancer Kit (Vysis Inc., Downers Grove, IL), a FISH test.
- The ImmunoCyt™ test, also marketed as UCyt+™ (DiagnoCure Inc., Quebec).
With the exception of the ImmunoCyt test, which is only cleared for monitoring bladder cancer recurrence, all tests are FDA-cleared as adjunctive tests for use in the initial diagnosis of bladder cancer and surveillance of bladder cancer patients, in conjunction with standard procedures.
The use of urinary tumor markers is considered investigational in the diagnosis of, monitoring, and/or screening for bladder cancer.
* FDA Approved indications
For the purpose of this policy, standard diagnostic procedures for bladder cancer consist of urine cytology and cystoscopy, with or without biopsy.
The BTA (bladder tumor antigen) stat® and nuclear matrix protein 22 (NMP22) are immunoassay tests.
When performed qualitatively in the physician’s office, CPT code 86294 Immunoassay for tumor antigen, qualitative and semi-quantitative (e.g., bladder tumor antigen), may be used to describe the BTA stat test and CPT code 86386 Nuclear Matrix Protein 22 (NMP22), qualitative may be used to describe the NMP22 test. Prior to 2012, CPT code 86294 was used for both BTA and NMP22 testing.
For clinical laboratories performing a quantitative version of these tests, CPT code 86316 Immunoassay for tumor antigen; other antigen, quantitative (e.g., CA50, 72-4, 549) each, may be used to describe the test. Other tumor assays that may be coded for using 86316 are addressed in a separate policy, No. 2.03.02.
Effective in 2011, there are specific CPT codes for urinary FISH testing:
88120: Cytopathology, in situ hybridization (e.g., FISH), urinary tract specimen with morphometric analysis, 3-5 molecular probes, each specimen; manual
88121: using computer-assisted technology.
Prior to 2011, examples of coding that laboratory companies used for this FISH test included:
- CPT codes 88271(x4) with 88274
- CPT code 88367 (x4); and
- CPT code 88368 (x4)
BlueCard/National Account Issues
Some state or federal mandates (e.g., FEP) prohibit plans from denying FDA-approved technologies as investigational. In these instances plans may have to consider the coverage eligibility of FDA-approved technologies on the basis of medical necessity alone.
This policy was originally created in 1997 and was updated regularly with searches of the MEDLINE database. The most recent literature search was performed for the period April 2012 through January 31, 2013. Following is a summary of the key literature to date:
All of the U.S. Food and Drug Administration (FDA)-approved tests for urinary tumor markers involve the use of standard laboratory procedures.
Urinary bladder tumor markers [i.e. BTA (bladder tumor antigen) STAT, NMP22 (nuclear matrix protein 22), UroVysion and Immunocyt]
Studies have evaluated the diagnostic performance of individual markers compared to urine cytology, the standard urine-based test for bladder tumor diagnosis and surveillance. Cystoscopy and biopsy are generally used as the gold standard comparison. Of particular interest are the relative performance of individual markers and the performance of individual markers compared to combinations of markers.
There are a number of diagnostic accuracy studies evaluating urinary tumor markers, as well as systematic reviews of these studies. A 2011 article by Parker and Spiess reviewed the published literature and summarized the sensitivity and specificity of several urine tumor markers in bladder cancer for diagnosis and/or monitoring of recurrence. (1) Selected information from the article is reported in the table below. (Diagnostic accuracy was not reported separately for initial diagnosis versus cancer monitoring).
Test Sensitivity range (%) Specificity range (%) Cytology 12 to 79 78 to 99 BTA STAT 50 to 70 67 to 78 NMP22 50 to 92 66 to 87 ImmunoCyt 67 to 85 62 to 85 FISH (UroVysion) 69 to 92 89 to 95
Sensitivity range (%)
Specificity range (%)
12 to 79
78 to 99
50 to 70
67 to 78
50 to 92
66 to 87
67 to 85
62 to 85
69 to 92
89 to 95
FISH, fluorescence in situ hybridization
In 2010, The U.K. Health Technology Assessment Program published a systematic review of studies on the diagnostic performance of several urine biomarkers. (2) The review included 71 studies on the test performance of cytology and urine biomarkers. A majority of the studies included patients both with and without a history of bladder cancer, or included only patients with a history of bladder cancer. Few studies were identified that focused on the evaluation of urinary markers for the initial diagnosis of bladder cancer. Pooled analyses of study findings combined results of tests used for initial diagnosis of bladder cancer and tests used to identify bladder cancer recurrence. Studies used cystoscopy with biopsy as the reference standard. Results of pooled patient-level analyses are:
Sensitivity % (95% confidence interval [CI])
Specificity % (95% CI)
CI, confidence interval
Representative studies focusing on the diagnostic accuracy of urinary tumor markers for the initial detection of bladder cancer are described below:
In 2013, Todenhofer and colleagues published a prospective study including 2,113 individuals suspected of having bladder cancer. (3) All patients underwent cystoscopy and upper urinary tract imaging, and urine samples were analyzed. A total of 502 (24%) of patients were found to have bladder cancer. The sensitivities of cytology, FISH and UCyt+ (also known as ImmunoCyt) for detecting bladder cancer were 80%, 69% and 72%, respectively. Specificities were 83%, 83% and 79%, respectively.
In 2008, Schmitz-Drager and colleagues in Germany prospectively evaluated the UCyt+ test in urine samples obtained from 301 consecutive patients with a first episode of painless hematuria and no history of bladder cancer. (4) Sixty-five patients (22%) patients presented with gross hematuria and 228 (78%) with microhematuria. The sensitivity of the UCyt+ was at least as high as that of the other diagnostic tests, but specificity was lower. Among patients with gross hematuria, sensitivity and specificity were 87% and 96% for cystoscopy, 88% and 80% for UCyt+ and 47% and 91% for cytology. Among patients with microhemoturia, sensitivity and specificity were 80% and 99% with cystoscopy, 80% and 89% with UCyt+ and 40% and 97% with cytology.
Representative studies focusing on the diagnostic accuracy of urinary tumor markers for detecting bladder cancer recurrence are described below:
A cross-sectional study from Germany, published by Horstmann and colleagues in 2009, compared the performance of UroVysion, ImmunoCyt and NMP22 used to detect bladder cancer recurrence in a sample of 221 patients diagnosed with non-muscle-invasive transitional cell carcinoma. (5) Patients subsequently underwent cystoscopy as part of regular follow-up (n=49) or transurethral resection of the bladder (TURB) for suspicion of recurrent disease (n=172). Findings from cystoscopy or TURB were considered the gold standard diagnosis. The investigators evaluated the diagnostic performance of individual markers, urinary cytology, and all possible combinations of markers. When combinations of markers were used, the test was considered positive if at least one marker was positive.
Cytology was the most sensitive single marker (84%) but was less specific than ImmunoCyt (62% and 72%, respectively). The authors commented that the performance of cytology was better than in previous similar studies, and the performance of other single markers was similar to previous studies. All combinations of 2 tests increased the sensitivity. Sensitivities varied from 94%, with a combination of cytology and NMP22, to 87% for the combination of cytology and UroVysion. Combining 2 tests generally lowered the specificity. In monitoring patients for bladder cancer recurrence; sensitivity is the more important test characteristic. Still, the combination with the best tradeoff of sensitivity and specificity was cytology and ImmunoCyt, which had a sensitivity of 93% and a specificity of 56%. Combining 3 tests increased the sensitivity even further. Two combinations attained a sensitivity of 98%, NMP22 and ImmunoCyt combined with either cytology or UroVysion. Specificity of these combinations was low, 31-32%. The best tradeoff with 3 markers was the combination of cytology, ImmunoCyt, and UroVysion, which had a sensitivity of 93% and a specificity of 49%. Combining all 4 tests did not substantially improve the diagnostic performance.
In 2009, Sullivan and colleagues published a cross-sectional study that compared the urinary tumor markers ImmunoCyt and UroVysion in patients with a history of bladder cancer. (6) A single voided sample was obtained from 100 patients. Immediately after urine collection, patients underwent cystoscopy to identify cancer recurrence. Cystoscopy with biopsy was the gold standard; only biopsy-proven cases were considered positive. The urine sample was divided and used to evaluate cytology, ImmunoCyt and UroVysion; each type of analysis was conducted blindly in a different laboratory. Of the 100 samples, 2 were considered inadequate for cytology, 2 were inadequate for ImmunoCyt analysis, and 12 had cell counts too low for UroVysion analysis. Thus, sample size was 98 for cytology and ImmunoCyt and 88 for UroVysion. Sensitivities were 21% for cytology, 76% for ImmunoCyt, and 13% for UroVysion. Specificities were 97% for cytology, 63% for ImmunoCyt, and 90% for UroVysion. Diagnostic performance of the combination of cytology and ImmunoCyt, but not cytology and UroVysion, was reported. In the analysis of 2 tests, sensitivity was calculated with either test positive and specificity with both tests negative. For the combination of cytology and ImmunoCyt, the sensitivity was 75% and specificity was 63%. The specificity of this combination of tests was similar to that found by Horstmann and colleagues, described above, (3) which was 56%. The combined sensitivity was lower than in the Horstmann study (93%), likely due to the higher sensitivity of urinary cytology found by Horstmann et al. The Sullivan study was limited by a small sample size. Moreover, the study was supported by DiagnoCure, the manufacturer of ImmunoCyt; the Horstmann study did not receive industry funding.
A prospective study by Kamat and colleagues evaluated the accuracy of 5 bladder cancer surveillance protocols for identifying recurrence in patients with a history of bladder cancer. (7) Four patient management strategies were compared: cystoscopy alone; cystoscopy and NMP22; cystoscopy and Urovysion; and cystoscopy and cytology. In addition, a fifth hypothetical protocol was evaluated; cystoscopy and contingent strategy in which a Urovysion test was only performed if the NMP22 test was positive. After an initial evaluation, patients were followed with routine cystoscopy every 3-6 months. For patients with a negative cystoscopy at baseline and in whom a tumor was detected at the first follow-up (i.e., within 6 months), it was assumed that this was a true result reflecting a missed diagnosis at the initial examination. Cancer was detected in 13 of 200 (6.5%) patients at the baseline evaluation and in 12 of 187 (6.4%) initially negative patients at first follow-up. Each of the patient management strategies described above correctly identified all 13 patients diagnosed with cancer at study entry. The proportion of false-positives at baseline was 2 of 15 (13%) patients testing positive using cystoscopy alone, 19 of 32 (59%) positives with cystoscopy and NMP22, 30 of 43 (70%) positives with cystoscopy and Urovysion, 14 of 27 (52%) positives with cystoscopy and cytology, and 6 of 19 (32%) positives with cystoscopy and NMP22, followed by Urovysion if the NMP22 test was positive. The number of initial false-positives that were confirmed positive at the first follow-up for each strategy was 0, 1, 5, 2, and 1, respectively. The 2 invasive tumors (out of 12 total tumors) identified at first follow-up were missed by all 5 surveillance strategies; urinary tumor markers only detected non-invasive tumors.
Conclusions: Numerous studies have evaluated the accuracy of the urinary tumor markers BTA STAT, NMP22, UroVysion and ImmunoCyt for diagnosing and/or monitoring bladder cancer. Several systematic reviews of these studies have been published. In studies on the initial diagnosis of bladder cancer and/or detection of recurrent bladder cancer, urinary tumor markers tend to have higher sensitivity but not higher specificity than cytology. Combining tumor markers with cytology can improve overall diagnostic accuracy.
Urinary bladder tumor markers for detecting upper urinary tract (UUT) tract disease in patients with a history of bladder cancer and a negative cystoscopy
No studies were identified that specifically addressed the diagnostic accuracy of urinary tumor markers for diagnosing upper tract cancers in patients with a history of bladder cancer. Several studies have addressed the accuracy of urinary tumor markers for diagnosing upper urinary tract diseases. However, the populations included in this study were either patients with suspected disease or a mixed group of patients with suspected disease and a history of bladder cancer or upper urinary tract cancer. For example, Lodde and colleagues in Austria evaluated the accuracy of Immunocyt for detecting upper urinary tract transitional cell carcinoma (UT-TTC). (8) The study included 37 patients with signs or symptoms suggestive of UT-TCC; 14 patients (38%) had a history of bladder cancer. Sixteen of 37 patients (43%) were found to have UT-TCC. All patients also underwent cystoscopy, renal ultrasonography and intravenous excretory urography. Using voided urine samples, Immunocyt had 75% sensitivity and 95% specificity for identifying UT-TCC. This compares to a sensitivity of 50% and specificity of 100% for cytology. Using ureteral urine samples, Immunocyt had a sensitivity of 91% and cytology had a sensitivity of 82%. Both tests had 100% specificity using ureteral urine. The combination of Immunocyt and cytology had a sensitivity of 88% in voided urine samples and a sensitivity of 100% in ureteral urine. In 2011, Xu and colleagues in China reported on the diagnostic accuracy of UroVysion FISH for detecting upper tract urothelial carcinoma. (9) The study included urine specimens from 85 patients suspected of having upper urinary tract disease. Patients underwent cystoscopy after urine collection. Seventeen patients (20%) had a history of UT urothelial carcinoma and 8 (9%) had a history of bladder cancer. The remaining patients had signs or symptoms of disease such as hematuria. The sensitivity of FISH for diagnosing UT carcinoma was 79% and the sensitivity of cytology was 45%. Specificity was 98% for FISH and 100% for cytology. When findings from cytology and FISH were combined, the sensitivity was 86% and the specificity was 98%. Neither study separately reported findings for detection of recurrence in patients with a history of urinary tract cancer, or for patients with a negative cystoscopy.
In 2012, Picozzi and colleagues published a systematic review of studies that reported data related to upper urinary tract recurrence following radical cystectomy for bladder cancer. (10) Upper tract recurrence was defined as any documented recurrence in the renal collecting system or ureter. The authors identified 27 studies with a total of 13,185 participants. The overall prevalence of UT in the studies ranged from 0.75% to 6.4% and, among the cancers detected, 64.6% were advanced and 35.6% were metastatic. The Picozzi review also reported on the diagnostic yield of protocols used to follow patients after treatment for bladder cancer. As reported in the review, in 14 studies, 63 of 166 patients (38%) with upper urinary tract recurrence were identified by follow-up investigations and in the remaining 103 (62%) of patients, diagnosis was based on symptoms. In 9 studies that used urine cytology, 10 of 112 (9%) patients with recurrence were identified by positive cytology. In 13 studies that used upper tract imaging, 40 of 161 (25%) patients with recurrence were identified by imaging. Put another way, approximately 2,000 urine cytology examinations or 800 radiological examinations were performed to identify 1 patient with UT recurrence. The authors stated that they were not able to determine whether there was a survival advantage in patients whose tumors were identified by cytology or UT imaging compared to symptoms because the data on this subject were poor. The Picozzi review did not discuss the use of urinary tumor markers for diagnosis of UUT recurrence.
Conclusions: No studies were identified that focused specifically on the use of urinary tumor markers for detecting upper urinary tract recurrences in patients with a history of bladder cancer. Several studies have evaluated urinary tumor markers for detecting upper urinary tract disease in samples of patients both with and without a history of urinary carcinoma. Available studies generally found that urinary tumor markers had higher sensitivity but not higher specificity than cytology, and combining urinary markers and cytology improved diagnostic accuracy.
FGFR3 (Fibroblast Growth Factor Receptor 3) mutations
Several studies have evaluated urine-based assays for identifying FGFR3 mutations. A 2012 study was published by Fernandez and colleagues; several authors were employees of Predictive Biosciences, the manufacturer of the CertNDx test. (11) The study included 323 individuals who had been treated for bladder cancer; 48 of these had a recurrence of bladder cancer and the remaining 275 had no current evidence of disease. Seven patients without disease did not have sufficient DNA for FGFR3 mutation testing and were excluded from further analysis. FGFR3 mutations were detected in 15 samples, 5 from patients with cancer recurrence and 10 from individuals without evidence of disease. This resulted in a sensitivity of 5/48 (10%) and a specificity of 258/268 (96%). When results of FGFR3 mutation analysis were combined with the findings of other tests (matrix metalloproteinase 2 (MMP2), Twist 1 and Nid2 methylation), the markers had a 92% sensitivity (44/48) and 51% specificity (136/268) for detecting cancer recurrence.
In a retrospective study, Rieger-Christ and colleagues compared the accuracy of FGFR3 mutation analysis, cytology and the combination of the two in identifying bladder tumors. (12) The study included 192 patients with bladder cancer, 72 who underwent TURB (Group A) and 120 who underwent cystectomy (Group B). Urine samples were collected prior to surgery. DNA preparations were screened for FGFR3 mutations using single-strand conformation polymorphism (SSCP) and DNA sequencing. (The study did not appear to use the CertNDx test). Cytology results were available for 62/72 (86%) in the TURB group and 62/120 (52%) in the cystectomy group. Sensitivity of the FGFR3 test alone was 68% for Group A and 24% for Group B. The sensitivity of cytology alone was 32% for Group A and 90% for Group B. For the combination of FGFR3 and cytology, the sensitivity was 78% for Group A and 93.5% for Group B.
In addition, Zuiverloon and colleagues have applied FGFR3 mutation analysis to the detection and prediction of bladder cancer recurrence. The research team, based in the Netherlands, developed an assay to identify common FGFR3 mutations in urine samples. An study published in 2010 identified the FGFR3 mutation status of tumors in 200 patients with low-grade non-muscle invasive bladder cancer. FGFR3 mutations were identified in 134 (67%) patients. (13) The 134 patients with an FGFR3-mutant tumor provided 463 urine samples, and 45 concomitant histologically proven recurrences of bladder cancer were found. The sensitivity of the assay to detect concomitant recurrences was 26/45 (58%). After at least 12 months of follow-up from the time of the last urine sample, an additional 34 recurrences were identified. Overall, 85 of 105 (81%) FGFR3-positive urine samples were associated with a bladder cancer recurrence compared to 41 of 358 (11%) FGFR3-negative urine samples. In a Cox time-to-event analysis, an FGFR3-positive urine was associated with a 3.8-fold higher risk of having a recurrence (p<0.0001). Another study by this research team was published in 2012. (14) A total of 716 urine samples were collected from 136 patients with non-muscle invasive bladder cancer (at least 3 samples per patient were required for study entry. During a median of 3 years of follow-up, there were 552 histologically proven bladder cancer recurrences. The sensitivity of FGFR3 for detecting a recurrence was 201/408 (49%) and 124/187 (66%), respectively. In comparison, the sensitivity of cytology was 211/377 (56%) and the specificity was 106/185 (57%). Combining FGFR3 and cytology increased sensitivity to 76% but lowered specificity to 42%.
Other urinary bladder tumor markers
A 2009 review article on potential new tumor markers comments that bladder cancer tumor markers is a rapidly evolving field in which new markers are constantly identified. (15) The review concluded, “1) there exists a dizzying number of markers identified using newer expertise, and 2) much more work will need to be done to delineate which markers may be clinically applicable and which will be discarded.”
Regarding the variety of other potential tumor markers in bladder cancer, most of the published studies evaluating them have included small numbers of patients and were preliminary investigations. Literature searches identified 1 meta-analysis and several larger prospective studies on additional markers. The meta-analysis, published in 2012 by Ku and colleagues, addressed using urine survivin as a marker for diagnosing bladder cancer and used cystoscopy and/or histopathology as a reference standard. (16) They identified 14 studies, 3 of which were conducted in the United States and 3 of which identified recruitment as prospective. A meta-analysis of data from the studies found a pooled sensitivity for the urine survivin test of 0.77 (95% confidence interval [CI]: 0.74-0.80) and a pooled specificity of 0.92 (95% CI: 0.90-0.93). In a pre-planned subgroup analysis comparing the diagnostic accuracy of survivin and cytology, a pooled analysis of data from 6 studies found that survivin had a significantly better sensitivity than cytology, but a significantly lower specificity; the sensitivity and specificity of cytology for diagnosing bladder cancer was 0.43 and 0.98, respectively.
Among the prospective studies evaluating potential tumor markers, Eissa and colleagues in Egypt aimed to determine the ability of HYAL1 and survivin to identify malignant bladder tumors. (17) A total of 278 patients underwent urine analysis and cystoscopy, and 100 healthy volunteers who did not undergo cystoscopy served as controls. Among patients, 166 were found to have bladder cancer, and 112 had benign bladder lesions. Using qualitative real-time polymerase chain reaction (RT-PCR) analysis, HYAL1 was identified in 153 (92%) malignant samples and 12 (11%) of benign samples, and survivin in 126 (76%) of malignant samples and 12 (11%) of benign samples. HYAL1 and survivin were not identified in any of the control samples. Using the best cutoffs for discriminating the malignant and non-malignant groups, the sensitivity of HYAL1 was 92.2% at 94.3% specificity. This was higher than a comparable analysis of survivin, which had a 75.9% sensitivity and 94.3% specificity. Using semi-quantitative RT-PCR analysis, the sensitivity of HYAL1 was 91% and of survivin was 95.9%; specificity in both cases was 100%.
Similarly, Passerotti and colleagues in Brazil evaluated urinary hyaluronate for diagnosing transitional cell carcinoma (TCC). (18) Urine samples were taken from 350 patients prior to surgery (cystoscopy, cystectomy or transuretheral resection for bladder cancer). Postoperatively, a total of 160 patients (46%) were found to have TCC. Using area under the curve analysis, the investigators identified the optimal urinary hyaluronan cutoff to be 13.0 ug; with this cutoff, the test had an sensitivity of 82.3% and a specificity of 81.2% for identifying the presence of TCC.
Moreover, a 2010 study by Li and colleagues in China prospectively evaluated the cytokeratin 20 (CK20) test for detecting urothelial carcinoma. (19) Diagnostic accuracy of CK20 was compared to cytology and the Immunocyt test, using cystoscopy with histological diagnosis as the reference standard. The study included 169 patients who were hospitalized for a urological condition; 22 healthy individuals were included as controls. Thirty-four of 169 (20%) patients were excluded from the analysis due to missing data. Of the remaining 135 patients, 93 had urothelial carcinoma (primary tumors in 68 and recurrent tumors in 25), 26 had other urogenital malignancies, and 16 had benign lesions. A total of 132 patients had findings available on all 3 tests. The sensitivity of liquid-based cytology alone was 49.4% and the specificity was 91.1%. The combination of cytology and CK20 yielded a sensitivity of 81.6% and a specificity of 88.9%. When all 3 tests were used together (any positive test scored as positive), the sensitivity was 90.8% and the specificity was 84.4%. It is worth noting that the Immunocyt test is FDA-approved for use in patients already diagnosed with bladder cancer, not for initial diagnosis. The authors did not specify whether or not all study participants, who were inpatients in a department of urology, had previously received a diagnosis for their condition.
Conclusions: Studies have evaluated various other potential urinary tumor markers but there is insufficient evidence on the diagnostic accuracy of any particular marker.
Impact of urinary tumor marker tests on patient care
Because of the potential consequences of missing a diagnosis of recurrent bladder cancer, it is unlikely that the schedule of cystoscopies will be altered unless the sensitivity of urinary marker/markers approaches 100%. However, some authors have suggested that consideration be given to lengthening the intervals of cystoscopy in patients with low levels of an accurate marker and low-grade bladder cancer. In addition, while urinary tumor markers might not alter the schedule of cystoscopies, if their results suggest a high likelihood of tumor recurrence, the resulting cystoscopy might be performed more thoroughly, or investigation of the upper urinary tract might be instigated. (20)
No studies were identified that prospectively evaluated health outcomes in patients who were managed with and without the use of urinary tumor marker tests. In addition, there were no published studies to date comparing different cystoscopy protocols, used in conjunction with urinary markers, to monitor recurrence.
A 2011 study by Shariat and colleagues used a decision-curve analysis to assess the impact of urinary marker testing using the NMP22 test on the decision to refer for cystoscopy and concluded that the marker did not aid clinical decision making in most cases. (21) The study included 2,222 patients with nonmuscle-invasive bladder cancer and negative cytology, at various stages of surveillance. (Patients with positive urinary cytology were excluded, since standard practice is to refer these patients for cystoscopy). According to the study protocol, all patients underwent cystoscopy, and 581 (26%) were found to have disease recurrence; of these, 234 (40%) had disease progression. NMP22 level was found to be significantly associated with both disease recurrence and progression (p<0.001 for both).
In the analysis, the clinical net benefit of the NMP22 test was evaluated by summing the benefits (true-positives), subtracting the harms (false-positives), and weighing these values by the “threshold probability,” defined as the minimum probability of bladder cancer or recurrence at which a patient or clinician would opt for cystoscopy. The investigators found only a small clinical net benefit of the NMP22 test over the strategy of “cystoscopy for all patients,” and this benefit occurred only at threshold probabilities over 8%. For example, for patients with at least a 15% risk of recurrence, using a model containing age, sex, and NMP22, 229 (23%) cystoscopies could be avoided, 236 (90%) recurrences would be identified and 25 (15%) recurrences would be missed. Thus, for clinicians or patients who would opt for a cystoscopy even if patients had a low risk of recurrence e.g. 5%, NMP22 would not add clinical benefit and the optimal strategy would be to offer cystoscopy to all at-risk patients. The authors attributed the low clinical net benefit to the high risk of bladder cancer recurrence in patients with negative cytology
Conclusions: There is a lack of evidence that health outcomes are improved in patients managed with urinary tumor marker tests compared to those managed without tumor marker tests and a lack of direct evidence that cystoscopy protocols can be changed when urinary tumor marker tests are used. A 2011 modeling study found low potential clinical benefit of urinary tumor marker testing for patients with nonmuscle invasive bladder cancer in terms of avoiding cystoscopy.
Urinary markers for screening asymptomatic individuals for bladder cancer
In 2010, the U.S. Preventive Services Task Force (USPSTF) published an updated evidence review on screening adults for bladder cancer. (22) The quality of direct evidence that screening for bladder cancer reduces morbidity or mortality was poor. There were no randomized controlled trials (RCTs), and only one prospective study, which was rated as being poor quality. The systematic review did not identify any studies evaluating the sensitivity or specificity of diagnostic tests for bladder patients in asymptomatic average-risk patients. Moreover, the review did not identify any suitable studies on whether treatment of screen-detected bladder cancer reduces disease-specific morbidity and mortality, or on potential harms of screening for bladder cancer. The authors concluded that “major gaps in evidence make it impossible to reach any reliable conclusions about screening.”
A modeling study published in 2006 reported that screening the general population for bladder cancer using tumor markers would not be beneficial but that screening an asymptomatic high-risk population would yield a benefit similar to other cancer screening programs (e.g., prostate, colon, and breast cancer). (23)
In 2009, Lotan and colleagues published a prospective study in which 1,502 individuals at high-risk of bladder cancer due to age plus smoking and/or occupational exposure were screened. (24) Approximately 60% of the sample was recruited from a Veterans Administration hospital and 1,175 (78%) of the study population was male. Approximately 73% of participants had undergone urinalysis within 3 years of screening. Individuals with a history of urological malignancy or gross malignancy and those with current urinary problems that might increase the false-positive rate were excluded. The study used the NMP22 BladderChek test and was supported by the test manufacturer. Individuals with positive BladderChek tests underwent additional testing, beginning with urinalysis. Those found to have infection on urinalysis were treated and their urine was re-tested; others who tested positive received cystoscopy and cytology. Individuals with a negative BladderChek test did not have to undergo additional testing.
Eighty-five (5.7%) of the 1,502 participants had a positive BladderChek test. Of these, 69 (81%) underwent cystoscopy; 14 refused, and 2 patients with urethral strictures were unable to be examined. Two of the 85 patients were found to have bladder cancer (non-invasive), yielding a positive predictive value of 2.4%. There was also one case of atypia. Follow-up at a mean of 12 months was obtained for 1,309 of 1,502 (87%) screened individuals. No additional cancers were diagnosed in the group that had had positive BladderChek tests. Two participants with negative BladderChek screen had been diagnosed with bladder cancer; both tumors were less than 1 cm. Since no follow-up tests were done on participants who initially tested negative, it cannot be known whether these were false-negative findings or new cancers. The authors report that the cancer prevalence in this population was lower than expected, which could be due in part to the large proportion that had previously undergone urinalysis. Study limitations include lack of follow-up testing on approximately 20% of participants who tested positive andlack of early cystoscopy and incomplete 1-yeartelephone follow-up in those who tested negative.Because of these limitations, accurate test operating characteristics (e.g., sensitivity)cannot be calculated.
Conclusions: There are no RCTs evaluating the impact of screening for bladder cancer on health outcomes in asymptomatic individuals. There is also insufficient observational evidence on the diagnostic accuracy of urinary tumor markers used to screen asymptomatic individuals for bladder cancer.
Clinical Input Received through Specialty Medical Societies and Academic Medical Centers
In response to requests, input was received through 2 Physician Specialty Societies and 5 Academic Medical Centers in 2012 prior to the annual policy update. While the various Physician Specialty Societies and Academic Medical Centers may collaborate with and make recommendations during this process, through the provision of appropriate reviewers, input received does not represent an endorsement or position statement by the Physician Specialty Societies or Academic Medical Centers, unless otherwise noted. There was unanimous agreement that FDA-approved urinary tumor markers may be considered medically necessary as an adjunct test in the diagnosis and monitoring of bladder cancer in conjunction with standard diagnostic procedures. In contrast, there was mixed support but no uniform consensus on the incremental value of urinary tumor markers compared to urinary cytology alone and for whether urinary tumor markers lead to changes in patient management. There was unanimous agreement that use of urinary tumor markers are investigational to screen for bladder cancer in asymptomatic individuals
Numerous studies have evaluated the accuracy of the urinary tumor markers BTA STAT, NMP22, UroVysion and ImmunoCyt for diagnosing and/or monitoring bladder cancer. These urinary tumor markers tend to have higher sensitivity but not higher specificity than cytology and combining tumor markers with cytology can improve overall diagnostic accuracy. There is insufficient evidence that urinary tumor markers improve the accuracy of initial diagnosis of upper urinary tract disease or monitoring patients with a history of urinary tract disease for upper tract recurrence. There is little evidence on the impact of urinary bladder tumor marker tests on patient management, e.g., the frequency of cystoscopy, or on the impact of tests on health outcomes. Moreover, there is also a lack of evidence on the impact of screening asymptomatic individuals for bladder cancer using urinary tumor markers.
In 2012-2013, the evidence on the use of urinary tumor markers was reevaluated with focus on whether clinical utility has been established. It was concluded that, due to continued lack of published evidence of clinical utility, mixed input from clinical input on whether results of urinary tumor marker tests leads to change in patient management and a lack of support for urinary marker use from clinical practice guidelines, the use of urinary tumor markers individuals is considered investigational.
Practice Guidelines and Position Statements
National Comprehensive Cancer Network (NCCN) 2013 bladder cancer guideline included the following statement regarding monitoring patients with high-grade bladder tumors:: “…Urine molecular tests for urothelial tumor markers are now available. Most of these tests have a better sensitivity for detecting bladder cancer than urine cytology, but specificity is lower. However, it remains unclear whether these tests offer additional information which is useful for detection and management of non-muscle invasive bladder tumors. Therefore, The NCCN Bladder Cancer panel members consider this a category 2B recommendation.” (25)
In 2011, the U.S. Preventive Services Task Force concluded that there was insufficient evidence to assess the benefits and harms of screening for bladder cancer in asymptomatic adults. (20) The recommendation was graded as an “I” recommendation, indicating insufficient evidence. This statement replaced a 2004 recommendation against routine screening for bladder cancer.
The National Academy of Clinical Biochemistry Laboratory Medicine Practice Guidelines, published in 2010, do not recommend use of any of the FDA-approved urinary tumor marker tests for diagnosis of bladder tumors or for monitoring bladder cancer patients. (26) The guideline stated: “At this time, no tumor markers tests can be recommended for use in the diagnosis and clinical management of bladder cancer. This includes tests for making a differential diagnosis, assessing prognosis, staging of the disease or monitoring patients for the early detection of recurrent disease. There are no prospective clinical trial data that establish the utility of any of the FDA cleared markers or the proposed markers for increasing survival time, decreasing the cost of treatment or improving the quality of life of bladder cancer patients.”
The American Urological Association’s 2007 guideline on management of bladder cancer included the following statement regarding urine-based markers for bladder cancer: “Despite their present and future potential, the critical evaluation and comparison of urine-based markers is beyond the scope of the current guideline involving the management of nonmuscle invasive bladder cancer.” (27)
Ongoing Clinical Trials
Screening for Bladder Cancer (NCT00848627): (28) This trial is evaluating screening for bladder cancer in men age 60 years and older with a history of smoking; estimated enrollment is 1,500 individuals. It is sponsored by the Baylor College of Medicine, and the estimated study completion date is February 2015.
Medicare National Coverage
No Medicare national coverage determination.
- Parker J, Spiess PE. Current and emerging bladder cancer urinary biomarkers. ScientificWorldJournal 2011; 11:1103-12.
- Mowatt G, Zhu S, Kilonzo M et al. Systematic review of the clinical effectiveness and cost-effectiveness of photodynamic diagnosis and urine biomarkers (FISH, ImmunoCyt, NMP22) and cytology for the detection and follow-up of bladder cancer. Health Technol Assess 2010; 14(4):1-331.
- Todenhofer T, Hennenlotter J, Aufderklamm S et al. Individual risk assessment in bladder cancer patients based on a multi-marker panel. J Cancer Res Clin Oncol 2013; 139(1):49-56.
- Schmitz-Drager BJ, Tirsar LA, Schmitz-Drager C et al. Immunocytology in the assessment of patients with asymptomatic hematuria. World J Urol 2008; 26(1):31-7.
- Horstmann M, Patschan O, Hennenlotter J et al. Combinations of urine-based tumor markers in bladder cancer surveillance. Scand J Urol Nephrol 2009; 43(6):461-6.
- Sullivan PS, Nooraie F, Sanchez H et al. Comparison of ImmunoCyt, UroVysion and urine cytology in detection of recurrent urothelial carcinoma. Cancer Cytopathol 2009; 117(3):167-73.
- Kamat AM, Karam JA, Grossman HB et al. Prospective trial to identify optimal bladder cancer surveillance protocol: reducing costs while maximizing sensitivity. BJU Int 2011; 108(7):1119-24.
- Lodde M, Mian C, Wiener H et al. Detection of upper urinary tract transitional cell carcinoma with ImmunoCyt: a preliminary report. Urology 2001; 58(3):362-6.
- Xu C, Zeng Q, Hou J et al. Utility of a modality combining FISH and cytology in upper tract urothelial carcinoma detection in voided urine samples of Chinese patients. Urology 2011; 77(3):636-41.
- Picozzi S, Ricci C, Gaeta M et al. Upper urinary tract recurrence following radical cystectomy for bladder cancer: a meta-analysis on 13,185 patients. J Urol 2012; 188(6):2046-54.
- Fernandez CA, Milholland JM, Zwarthoff EC et al. A noninvasive multi-analyte diagnostic assay: combining protein and DNA markers to stratify bladder cancer patients. Research and Reports in Urology 2012; 4(17-26).
- Rieger-Christ KM, Mourtzinos A, Lee PJ et al. Identification of fibroblast growth factor receptor 3 mutations in urine sediment DNA samples complements cytology in bladder tumor detection. Cancer 2003; 98(4):737-44.
- Zuiverloon TC, van dAMN, van dKTH et al. Fibroblast growth factor receptor 3 mutation analysis on voided urine for surveillance of patients with low-grade non-muscle-invasive bladder cancer. Clin Cancer Res 2010; 16(11-Jan):3011-8.
- Zuiverloon TC, Beukers W, van der Keur KA et al. Combinations of Urinary Biomarkers for Surveillance of Patients with Incident Nonmuscle Invasive Bladder Cancer: The European FP7 UROMOL Project. J Urol 2012.
- Shirodkar SP, Lokeshwar VB. Potential new markers in the early detection of bladder cancer. Curr Opin Urol 2009; 19(5):488-93.
- Ku JH, Godoy G, Amiel GE et al. Urine survivin as a diagnostic biomarker for bladder cancer: a systematic review. BJU Int 2012; 110(5):630-6.
- Eissa S, Swellam M, Shehata H et al. Expression of HYAL1 and survivin RNA as diagnostic molecular markers for bladder cancer. J Urol 2010; 183(2):493-8.
- Passerotti CC, Srougi M, Bomfim AC et al. Testing for urinary hyaluronate improves detection and grading of transitional cell carcinoma. Urol Oncol 2011; 29(6):710-5.
- Li HX, Li M, Li CL et al. Immunocyt and cytokeratin 20 immunocytochemistry as adjunct markers for urine cytologic detection of bladder cancer. Ann Quant Cytol Histol 2010; 32(1):45-52.
- Grocela JA, McDougal WS. Utility of nuclear matrix protein (NMP-22) in the detection of recurrent bladder cancer. Urol Clin North Am 2000; 27(1):47-51.
- Shariat SF, Savage C, Chromecki TF et al. Assessing the clinical benefit of nuclear matrix protein 22 in the surveillance of patients with nonmuscle-invasive bladder cancer and negative cytology. Cancer 2011; 117(13):2893-7.
- Chou R, Dana TSafbcaroteftUS. Preventive Services Task Force. Ann Intern Med 2010; 153(7):461-8.
- Svatek RS, Sagalowsky AI, Lotan Y. Economic impact of screening for bladder cancer using bladder tumor markers: a decision analysis. Urol Oncol 2006; 24(4):338-43.
- Lotan Y, Elias K, Svatek RS et al. Bladder cancer screening in a high risk asymptomatic population using a point of care urine based protein tumor marker. J Urol 2009; 182(1):52-8.
- Network NCC. NCCN Practice Guidelines in Oncology. Bladder Cancer V1. 2013. Available online at: http://www.nccn.org/professionals/physician_gls/pdf/bladder.pdf. Last accessed February, 2013.
- Fritsche HA, Grossman HB, Lerner S et al. National Academy of Clinical Biochemistry Guidelines for use of tumor markers in bladder cancer. Use of tumor markers in liver, bladder, cervical, and gastric cancers, Chapter 3. 2010. Available online at: http://www.aacc.org/SiteCollectionDocuments/NACB/LMPG/tumor/chp3h_bladder.pdf. Last accessed February, 2013.
- American Urological Association. Bladder Cancer Clinical Guideline. Chapter 1: The Management of Bladder Cancer: Diagnosis and Treatment Recommendations. Available online at: http://www.auanet.org/content/guidelines-and-quality-care/clinical-guidelines/main-reports/bladcan07/chapter1.pdf. Last accessed February, 2013.
- Sponsored by Baylor College of Medicine. Screening for Bladder Cancer (NCT00848627). Available online at: www.clinicaltrials.gov. Last accessed February, 2013.
|CPT||86294||Immunoassay for tumor antigen; qualitative or semiquantitative (e.g., bladder tumor antigen)|
|86316||Immunoassay for tumor antigen; other antigen, quantitative, each|
|88120||Cytopathology, in situ hybridization (eg, FISH), urinary tract specimen with morphometric analysis, 3-5 molecular probes, each specimen; manual (new code 1/1/11)|
|88121||Cytopathology, in situ hybridization (eg, FISH), urinary tract specimen with morphometric analysis, 3-5 molecular probes, each specimen; using computer-assisted technology (new code 1/1/11)|
|ICD-9 Diagnosis||188 – 188.9||Malignant neoplasm of bladder|
|233.7||Carcinoma in situ of bladder|
|239.4||Neoplasm of unspecified nature, bladder|
|V10.51||Personal history of malignant neoplasm of bladder
|ICD-10-CM (effective 10/1/1)||C67.0-C67.9||Malignant neoplasm of bladder code range|
|D09.0||Carcinoma in situ of bladder|
|D49.4||Neoplasm of unspecified behavior of bladder|
|Z85.51||Personal history of malignant neoplasm of bladder|
|ICD-10-PCS (effective 10/1/14)||Not applicable. ICD-10-PCS codes are only used for inpatient services. There are no ICD procedure codes forlaboratory tests.|
|Type of Service||Pathology|
|Place of Service||Laboratory/Physician’s Office|
Bladder Tumor Antigen
FISH, Bladder Cancer Testing
Tumor Marker, Bladder Cancer
|01/30/98||Add to Medicine section||New policy|
|07/10/98||Replace policy||S-code added|
|10/15/00||Replace policy||BCBS “S” code deleted, new coding information|
|08/15/01||Replace policy||New information on NMP-22 added|
|04/29/03||Replace policy||Policy updated with literature search for the period of July 2001 through December 2002; policy statement unchanged, references added|
|06/27/05||Replace policy||Literature review update for the period of December 2002 through May 2005; reference numbers 9 through11 and 13 and14 added. Policy statement revised to indicate the use of bladder cancer tumor markers involving immunoassay tests NMP-22 or BTA statâmay be considered medically necessary as an adjunct “in the diagnosis” of bladder cancer “only” in conjunction with cystoscopy. Information on fluorescence in situ hybridization (FISH) added; additional policy statement indicating FISH may be considered medically necessary added. Finally, added investigational policy statements for use of these test for screening in asymptomatic persons|
|9/27/05||Replace policy||Diagnosis code for hematuria added to coding section.|
|10/10/06||Replace policy||Literature review update completed for May 2005 through July 2006. Information about ImmunoCyt added to policy; policy statement added that this is medically necessary in the monitoring of patients with bladder cancer but investigational for other uses. Comments added about other markers being studied and that these are investigational. Reference numbers 15–18 added. Discontinued HCPCS code S3701 removed from policy.|
|04/09/08||Replace policy||Policy updated with literature review; no change in policy statements. Reference numbers 19 and 20 added.|
|05/14/09||Replace policy||Policy reviewed with literature search from March 2008 through March 2009. The policy statement has been re-worded for clarification but the intent of policy statements is unchanged. The policy statement has been modified to include FDA-approved uses; new reference numbers 21-27 added.|
|05/13/10||Replace policy||Policy reviewed with literature search from April 2009 through March 2010. Policy statements unchanged. Rationale rewritten; new references 1, 4-5, 7-12, 14-15 added, other references re-numbered or removed.|
|5/12/11||Replace policy||Policy updated with literature search from March 2010 through March 2011. References 7, 13, 14, 17 and 22 added; other references renumbered and/or updated. Policy statements unchanged.|
|7/12/12||Replace policy||Policy updated with literature search through March 2012. Policy changed to investigational for diagnosing, monitoring and/or screening for bladder cancer. References 1, 6-8, 10, 12, 15 and 20 added; other references renumbered or removed. Clinical input added.|
|03/14/13||Replace policy||Policy updated with literature search through January 31, 2013. Policy statement unchanged. References 3, 4, 8-10 and 14 added; other references renumbered or removed.|