|Subject: Gene Expression Profiling for Bladder Cancer|
|Document #: GENE.00056||Publish Date: 04/12/2023|
|Status: Reviewed||Last Review Date: 02/16/2023|
This document addresses commercially available gene expression tests for the diagnosis of bladder cancer, the selection of treatment and the monitoring of individuals with a history of bladder cancer.
Investigational and Not Medically Necessary:
Gene expression profiling for diagnosing, managing and monitoring bladder cancer is considered investigational and not medically necessary.
Diagnosis of Bladder Cancer
Several studies have evaluated a Cxbladder test for the detection of bladder cancer . In 2012, O’Sullivan and colleagues prospectively studied 485 individuals with no history of urothelial carcinoma (UC) who presented with macroscopic hematuria and underwent cystoscopy. Bladder cancer was diagnosed based on cystoscopic findings and histopathological analysis. In addition to cystoscopy, investigators analyzed urine samples using an early version of the Cxbladder test, as well as using urinary cytology and the biomarker tests NWMP22 and BladderChek. A total of 66 of the 485 individuals (13.6%) were diagnosed with UC. Cxbladder detected 54 of the 66 cases (sensitivity of 82%) at a prespecified specificity of 85%. The sensitivity of Cxbladder was higher than cytology (56.1%) and the NMP22 BladderChek test (37.9%) but the specificity was lower. The specificity of cytology was 94.5% and of NMP222 BladderChek was 96.4%.
Breen and colleagues (2015) evaluated five datasets owned by the manufacturer of Cxbladder. The datasets included a total of 939 individuals, including 476 from the O’Sullivan (2012) study, described above. In the initial dataset, the sensitivity of CxBladder Detect was 79.5%, which was higher than the other included tests, cytology (sensitivity of 45.5%), NMP22 (sensitivity of 44.9%) and FISH testing (sensitivity of 40.0%). The specificity of CxBladder Detect was 82.2%, which was lower than the other tests, which had specificities of 96.3%, 89.0% and 87.3%, respectively.
Another analysis of the O’Sullivan (2012) data was published by Kavalieris and colleagues (2015). This study addressed the development and validation of a model to “triage out” individuals presenting with hematuria who have a low probability of UC (test now known as Cxbladder Triage). The dataset was supplemented with additional cohorts with hematuria. The final dataset included 587 samples from individuals, 72 UC-positive and 515 UC-negative. Logistic regression analysis found that four clinical characteristics: age 60 or higher, male gender, history of smoking, and high frequency of macrohematuria, were significantly associated with an increased risk of UC. The investigators then tested the gene profile plus an index created from the clinical characteristic variables. Their testing generated models with negative predictive values (NPV) of at least 97%. This is a preliminary modeling study and did not prospectively evaluate diagnostic accuracy in a clinical sample or evaluate the clinical utility of Cxbladder Triage.
A systematic review and meta-analysis by independent investigators (Sathianathen, 2018) evaluated the diagnostic accuracy of various urinary biomarkers. The review included studies that compared the accuracy of biomarkers compared with cystoscopy in individuals presenting with primary hematuria. A total of 14 studies met the review’s inclusion criteria. Pooled estimates were calculated for each biomarker that had more than 1 study. Sensitivities were 67% for BTA, 78% for NMP22 Quantitative, 79% for NMP22 Qualitative, 82% for Cxbladder and 82% for UCyt. Specificities were 69% for BTA, 82% for NMP22 Quantitative, 84% for NMP22 Qualitative, 85% for Cxbladder and 87% for UCyt. In comparison, cytology findings from the same studies had specificities ranging from 92% to 98%. The authors concluded that the diagnostic performance of the biomarkers was inadequate to replace cystoscopy for individuals with primary hematuria.
A manufacturer-sponsored study examined the potential impact of Cxbladder test results on clinical decision-making (Darling, 2017; Lough, 2017). They recruited 12 urologists and presented each of them with information on 33 individuals with hematuria. The urologists were first presented with clinical information only and were asked to provide a ‘yes’ or ‘no’ response regarding whether they thought that urological evaluation was needed. The urologists were then provided results on the Cxbladder test and asked whether or not their initial recommendation changed. Cases were selected from a database of individuals enrolled in previous Cxbladder studies and were selected to represent a cross-section of demographic groups. Out of a total of 396 urologist/subject decisions, urologists initially recommended at least one invasive diagnostic test in 259 (65.4%) of cases. They recommended a total of 689 diagnostic procedures, of which 425 were considered invasive. After reviewing Cxbladder test results, the total number of invasive procedures recommended was 397, an 11% reduction. Participating urologists were not provided with results from any other tests, knew the study was sponsored by the manufacturer of Cxbladder, and were provided honoraria. it is not known whether any changes in management recommendations would have improved clinical outcomes.
In 2021, Raman and colleagues published data on the diagnostic performance of Cxbladder Resolve, the Cxbladder test that is focused on identifying individuals with hematuria who are likely to have high-grade tumors. An initial development dataset included 863 individuals with hematuria who were scheduled to undergo cystoscopy for possible UC. Cxbladder Resolve categorized 96 individuals for “high priority” workup. Among these 96 individuals, 47 had high-impact tumors (HIT) and 9 had low-impact tumors (LIT). Among the remaining individuals, 2 had HIT and 31 had LIT. Cxbladder Resolve was found to have a sensitivity to identify UC of 95.5% and a sensitivity to identify HIT of 95.9%, with a negative predictive value (NPV) of 99.2%. An external validation dataset included 548 individuals who were eligible to participate and provided a urine sample. In this sample, 29 individuals were classified for high priority workup, including 9 with HIT. Among the other subjects, 1 had HIT and 4 had LIT. Cxbladder Resolve had a sensitivity for identifying HIT of 90.0%, a specificity of 96.3% and an NPV of 99.8%. In addition, the investigators evaluated the external validation cohort using an algorithm involving Cxbladder Resolve testing in combination with other Cxbladder tests. A total of 285 of 548 (52%) individuals had negative Cxbladder Triage results. The remaining 263 individuals underwent testing with Cxbladder Detect and 164 samples (30%) were identified as low probability of UC. The first 2 rounds of testing yielded 2 false-negative results. Cxbladder Resolve testing was done on the remaining 99 individuals (those who were not ruled out by the first 2 rounds of testing). A total of 23 of these were identified for high-priority workup, 9 of whom had HIT. There was 1 additional case of HIT and 3 cases of LIT among the individuals not categorized as needing a high-priority workup. This study did not evaluate the impact on health outcomes of individuals managed with versus without Cxbladder Resolve testing.
The American Urological Association (AUA) 2020 Microhematuria guideline, developed in conjunction with the Society of Urodynamics, Female Pelvic Medicine & Urogenital Reconstruction (SUFU) stated, “Clinicians should not use urine cytology or urine-based tumor markers in the initial evaluation of patients with microhematuria. (Strong Recommendation; Evidence Level: Grade C).”
Surveillance of Individuals with a History of Bladder Cancer
Kavalieris and colleagues (2017) reported on the development and validation of the Cxbladder Monitor test, proposed for the surveillance of individuals with a history of urothelial cancer. The prospective study evaluated 763 individuals with a history of UC undergoing surveillance, including cystoscopy, for possible recurrent disease. Study participants were followed for up to 6 months, which included three scheduled clinical visits per person. In a training set of data from 339 individuals, the investigators identified three clinical variables that were strongly associated with UC: previous tumor type (primary or recurrent), number of years since the previous tumor, and gene expression data. The latter factor was a statistically modified version of the results of the Cxbladder test. In a validation set of samples from 424 individuals, the test (gene expression profiling plus clinical variables) had a sensitivity of 92% and an NPV of 96% (specificity was not reported). A total of 14 of 156 individuals (9%) with recurrent UC identified by cystoscopy tested negative with the Cxbladder Monitor test.
Koya and colleagues (2020) reported on findings from New Zealand’s healthcare system. The study was a retrospective audit of three public health providers who incorporated the Cxbladder Monitor test into their routine clinical surveillance of individuals with a history of bladder cancer. In the surveillance protocol, Cxbladder Monitor-negative individuals were considered to be at low risk of recurrence and did not undergo cystoscopy at that visit, but at their next scheduled visit 12 months later. Thus, annual surveillance visits alternated between Cxbladder Monitor tests (when negative) and cystoscopy. Individuals with a positive Cxbladder Monitor test and those considered to be high-risk individuals underwent cystoscopy. (Details of definitions of low risk and high risk were not available). During the 35-month study period, 305 individuals participated in the surveillance protocol, 257 (83.2%) low-risk and 52 (16.8%) high-risk. A total of 196 of 253 (77.5%) low-risk individuals had a negative Cxbladder result. Confirmed recurrence occurred in 3 of these 196 individuals over 35 months. In addition, 3 of 57 Cxbladder Monitor-positive low-risk individuals had confirmed recurrences. Among evaluable high-risk individuals, 4 of 49 had confirmed recurrences during the study; all 4 had positive Cxbladder Monitor results. Overall, 57 of 305 (19%) individuals were considered low risk and had negative Cxbladder findings and could thus potentially be managed with fewer cystoscopies. This study was conducted in New Zealand; it is not clear how many bladder cancer survivors or physicians in the United States would chose to follow this protocol. For low-risk individuals with a history of bladder cancer, the National Comprehensive Cancer Network (NCCN) recommends cystoscopy at 3 and 12 months and annually thereafter.
Xpert® Bladder Cancer Monitor
In 2021, Sharma and colleagues published a systematic review of studies published through April 19, 2021 on the diagnostic accuracy of the Xpert Bladder Cancer Monitor test for predicting tumor recurrence in individuals with non-muscle invasive bladder cancer (NMIBC). The authors identified 11 prospective observational studies published between 2017 and 2021. The studies included a total of 2896 participants and the tumor recurrence rate in individual studies ranged from 8% to 39%. In a meta-analysis of data from all 11 studies, the pooled sensitivity of the Xpert test was 0.73 (95% confidence interval [CI], 0.65% to 0.80%) and the pooled specificity was 0.77 (95% CI, 0.69 to 0.84). In a sub-analysis by tumor grade, the pooled sensitivity and specificity for high-grade tumor recurrence was 0.86 (95% CI, 0.77 to 0.92 and 0.78 (95% CI, 0.75 to 0.81), respectively. The pooled sensitivity and specificity for low-grade tumor recurrence was 0.58 (95% CI, 0.47 to 0.68) and 0.79 (0.75 to 0.82), respectively. The authors noted: “The diagnostic accuracy of the Xpert BC [bladder cancer] monitor noted in the present study is similar to other biomarkers and suffers similar limitations, i.e. poor diagnostic accuracy for low-grade tumors and poor specificity.”
A 2022 study by Fasulo and colleagues evaluated the potential clinical utility of the Xpert Bladder Cancer Monitor test for avoiding unnecessary cystoscopies. They conducted an observational study that included 139 individuals who experienced tumor recurrence and had received an Xpert test at the time of study enrollment. For individuals who tested negative, additional tests were performed after 4, 8 and 12 months. Testing continued as long as the individual tested negative. At study enrollment, 57 individuals (41.0%) had a negative test and 36 (63.2%) of these continued receiving active surveillance (AS). After 4 months, 34 of the remaining 46 individuals (73.9%) were still negative and 25 (73.5%) continued to remain on AS. At the 8 month follow-up, 25 of the remaining 29 individuals tested negative and, of these, 21 (84%) remained on AS. At the final follow-up at 12 months, all of the remaining 25 individuals tested negative and 21 (84%) of these remained on AS. The authors stated that, using the Xpert test manufacturer’s guidance that individuals with 2 consecutive negative tests could avoid cystoscopy, 73.9% of individuals could have avoided cystoscopies and 26.4% of individuals who failed AS would have been missed. After 3 negative tests, 86.2% of individuals could have avoided cystoscopies and 16.0% of those who failed AS would have been missed. Limitation of the study include that it did not prospectively manage individuals based on the results of the Xpert test, and it was uncontrolled i.e. it did not compare health outcomes in individuals managed with and without Xpert test results.
The AUA and SUFU do not recommend use of urinary biomarkers for surveillance of individuals with bladder cancer. In their 2016 guideline, amended in 2020, entitled Diagnosis and Treatment of Non-Muscle Invasive Bladder Cancer (NMIBC), they included the following relevant recommendations:
In surveillance of NMIBC, a clinician should not use urinary biomarkers in place of cystoscopic evaluation. (Strong Recommendation; Evidence Strength: Grade B)
In a patient with a history of low-risk cancer and a normal cystoscopy, a clinician should not routinely use a urinary biomarker or cytology during surveillance. (Expert Opinion)
The NCCN’s guideline on bladder cancer (V2.2022) discusses management of individuals with bladder cancer. For intermediate-risk individuals with NMIBC, the NCCN recommends follow-up with cystoscopy, imaging and urine cytology. For individuals with high-risk NMIBC, the NCCN recommends cystoscopy, upper tract imaging, abdominal pelvic imaging and urine cytology. All of the above are category 2A recommendations. In addition the guideline states, “consider urinary urothelial tumor markers” (category 2B recommendation). The document notes, “it remains unclear whether these tests offer additional information that is useful for detection and management of non-muscle-invasive bladder tumors. Therefore, the panel considers this to be a category 2B recommendation.”
Management of Bladder Cancer
A gene expression profiling test, Decipher Bladder, is proposed to aid in treatment decisions in individuals with locally advanced muscle-invasive bladder cancer. Seiler and colleagues (2017) published preliminary research that served as the basis for development of the test. They analyzed bladder tumor samples, prior to neoadjuvant chemotherapy, and determined that classification of tumors into four molecular sub-types would be optimal for predicting clinical outcomes. They then trained a genomic subtyping classifier to predict the four molecular subtypes: claudin-low, basal, luminal-infiltrated, and luminal. In the discovery cohort (n=233), the overall accuracy of the classifier was 76%, and in the validation cohort (n=82), the accuracy was 73%. The authors also conducted an analysis of clinical outcomes by molecular sub-type, using an independent non-neo-adjuvant chemotherapy dataset. Their findings suggested that individuals with basal tumors might benefit the most from neoadjuvant chemotherapies. Individuals with basal tumors, as determined by the classifier, had a 3-year overall survival (OS) rate of 49.2% in the cohort without neo-adjuvant chemotherapy compared with 77.8% in the neo-adjuvant chemotherapy cohort.
Batista da Costa (2019) explored the ability of the gene expression classifier (the Decipher test) to identify neuroendocrine (NE)-like tumors when histologic features of NE carcinoma were missing. The training cohort consisted of Decipher test results in 175 individuals with muscle-invasive bladder cancer and the radical cystectomy cohort included 225 individuals who underwent cystectomy without receiving neoadjuvant therapy. The classifier identified 8 (4.2%) samples in the training cohort as having NE-like tumors and 4 (1.8%) individuals in the radical cystectomy cohort. In the radical cystectomy cohort, individuals with NE-like tumors had a lower 1-year progression-free survival (PFS) rate compared with the rest of the cohort (65% vs 82%, p=0.046). Grivas and colleagues (2020) also focused on NE-like tumors. They used the Decipher test to identify NE-like tumors in a retrospective study of individuals with bladder cancer who underwent cisplatin-based neoadjuvant chemotherapy followed by radical cystectomy. Ten (4.3%) of 234 individuals were classified as having the NE-like subtype. In the primary analysis of 211 individuals with complete information, individuals with NE-like tumors had worse outcomes than those with non-NE-like tumors, including having all 5 cancer-specific deaths.
No prospective studies were available that compared outcomes in individuals with muscle-invasive bladder cancer who did and did not undergo testing with Decipher Bladder to aid in the decision regarding neo-adjuvant chemotherapy.
The NCCN Bladder Cancer guideline (V2.2022) does not recommend different approaches to chemotherapy treatment of muscle invasive bladder cancer according to molecular sub-type.
Bladder cancer is the sixth most common type of cancer in the United States. The American Cancer Society (2022) estimates that, in 2022, there will be approximately 81,180 new cases of bladder cancer and 17,100 deaths. Risk factors for bladder cancer include age, family history, tobacco use and environmental exposure to carcinogens. (NCI, 2020). About 90% of bladder cancer cases are urothelial (transitional cancer), originating in the cells that line the urinary tract; 70-80% of bladder cancer diagnosed in the United States is non-muscle-invasive; the remaining 20-30% are muscle-invasive tumors (DeGeorge, 2017).
Painless hematuria is the most common presenting sign of bladder cancer. Approximately 1.3% of individuals with microscopic hematuria and no obvious cause, and 20% of individuals with gross hematuria, will have bladder cancer (DeGeorge, 2017). Individuals with bladder cancer may also present with symptoms such as urinary frequency, nocturia and dysuria (NCI, 2020).
For individuals with incidentally discovered microscopic hematuria on urinalysis, clinical assessment is recommended, along with repeat urinalysis after treatment of any benign causes that were identified. If repeat urinalysis is also positive, individuals should be evaluated for renal disease and, if negative, should undergo cystoscopy and imaging of the upper urinary tract. Cystoscopy is recommended for all individuals presenting with gross hematuria and in individuals with microscopic hematuria age 35 years or older (DeGeorge, 2017).
Bladder cancer frequently recurs and therefore regular surveillance of the urinary tract is recommended after initial treatment. The NCCN Bladder Cancer Guideline (V2.2022) has offers several follow-up surveillance protocols depending on the type of cancer. For individuals with low-risk non-muscle-invasive bladder cancer, follow-up with cystoscopy is recommended and, for individuals with intermediate- or high-risk non-muscle-invasive bladder cancer, recommended protocols involve a combination of cystoscopy, upper tract imaging with or without abdominal pelvic imaging and urine tests at intervals that vary over time.
Gene expression tests have been developed to assist in the diagnosis, management and post-treatment surveillance of bladder cancer. Manufacturers contend that detection of purported biomarkers, and the use of predictive algorithms, can triage patients for additional workup, guide treatment decisions and obviate the need for some invasive tests (cystoscopies). Several gene expression profiling tests are commercially available for the diagnosis or management of bladder cancer. None of these tests have been cleared or approved by the U.S. Food and Drug Administration (FDA), as laboratory-developed tests do not require FDA approval.
The Cxbladder group of tests (Pacific Edge Diagnostics, Hummelstown, PA) involve processing urine samples to extract mRNA and using reverse transcription quantitative polymerase chain reaction (RT-qPCR) to amplify and measure the expression of 5 genes. Findings are combined with clinical characteristics and, using a proprietary algorithm, a risk score is calculated. Cxbladder Detect, the original test, is intended for the diagnosis of bladder cancer in high-risk individuals who present with gross hematuria. Cxbladder Triage, which also incorporates factors such as age, sex, smoking status and hematuria history, is intended to evaluate the likelihood of bladder cancer in low-risk individuals who present with hematuria. Cxbladder Resolve is designed to be used in conjunction with the Cxbladder Detect and Triage tests to identify individuals who are likely to have high-grade or late-stage bladder cancer and need a high-priority workup. Cxbladder Monitor is intended to be used in the surveillance of individuals with a history of bladder cancer to rule out disease recurrence. The tests are intended to reduce the need for additional clinical evaluation, in the case of Cxbladder Triage, or invasive testing such as cystoscopy in the case of Cxbladder Detect and Cxbladder Monitor.
The Decipher Bladder test (Veracyte Labs, San Diego, CA) is used in the management of bladder cancer, specifically to aid in the decision of individuals with locally advanced muscle-invasive bladder cancer who are considering neoadjuvant chemotherapy (NAC) prior to radical cystectomy. The test uses gene expression analysis to classify the molecular subtype (luminal, luminal infiltrated, basal, basal claudin-low, neuroendocrine-like) of tumor specimens.
The Xpert® Bladder Cancer Monitor test (Cepheid, Sunnyvale, CA) is used to monitor people with a history of NMIBC to determine the risk of recurrence or progression. The test measures the expression of five mRNA targets in urine samples. The test is intended to be used in conjunction with cystoscopy, and could potentially lead to a decrease in the frequency of cystoscopies.
Determining treatment decisions according to the molecular subtype of bladder tumors is not recommended by the NCCN. The NCCN Bladder Cancer guideline, V2.2022, recommends neoadjuvant cisplatin-based combination chemotherapy prior to cystectomy for individuals with Stage II muscle-invasive bladder cancer. For individuals who are not eligible for cisplatin-based chemotherapy, the guideline recommends cystectomy alone. As noted above, the NCCN guideline does not recommend different approaches to chemotherapy treatment of muscle invasive bladder cancer according to molecular subtype.
Gene expression profiling: A laboratory test that measures the activity of multiple genes at once for diagnostic or prognostic purposes. The test result is often reported as a proprietary summary score.
Hematuria: The presence of blood in urine.
Neoadjuvant therapy: Therapy given prior to the main treatment.
Non-muscle invasive bladder cancer (NMIBC): Cancer that is located in inner lining of the bladder, without growth of cancer cells into the deeper muscle layer of the bladder.
Urothelial: Cells that line the urinary tract, which includes that bladder, urethra, ureters and renal pelvis. Also called transitional cells. Bladder cancer that begins here is called urothelial cancer or transitional cancer.
The following codes for treatments and procedures applicable to this document are included below for informational purposes. Inclusion or exclusion of a procedure, diagnosis or device code(s) does not constitute or imply member coverage or provider reimbursement policy. Please refer to the member's contract benefits in effect at the time of service to determine coverage or non-coverage of these services as it applies to an individual member.
When services are Investigational and Not Medically Necessary:
For the following procedure codes, or when the code describes a procedure indicated in the Position Statement section as investigational and not medically necessary.
Oncology (urothelial), mRNA, gene expression profiling by real-time quantitative PCR of five genes (MDK, HOXA13, CDC2 [CDK1], IGFBP5, and CXCR2), utilizing urine, algorithm reported as a risk score for having urothelial carcinoma
Oncology (urothelial), mRNA, gene expression profiling by real-time quantitative PCR of five genes (MDK, HOXA13, CDC2 [CDK1], IGFBP5, and CXCR2), utilizing urine, algorithm reported as a risk score for having recurrent urothelial carcinoma
Oncology (bladder), mRNA, microarray gene expression profiling of 219 genes, utilizing formalin-fixed paraffin-embedded tissue, algorithm reported as molecular subtype (luminal, luminal infiltrated, basal, basal claudin-low, neuroendocrine-like)
Oncology (urothelial), mRNA, gene expression profiling by real-time quantitative PCR of 5 genes (MDK, HOXA13, CDC2 [CDK1], IGFBP5, and CXCR2), utilizing urine, algorithm incorporates age, sex, smoking history, and macrohematuria frequency, reported as a risk score for having urothelial carcinoma
Unlisted molecular pathology procedure [when specified as a bladder cancer gene expression test such as Cxbladder™ Resolve or Xpert® Bladder Cancer Monitor]
Peer Reviewed Publications:
Government Agency, Medical Society, and Other Authoritative Publications:
|Websites for Additional Information|
Xpert® Bladder Cancer Monitor
The use of specific product names is illustrative only. It is not intended to be a recommendation of one product over another, and is not intended to represent a complete listing of all products available.
Medical Policy & Technology Assessment Committee (MPTAC) review. Updated Description, Rationale, Background, Coding, References and Index sections..
Updated Coding section with 01/01/2023 CPT code changes; added 0363U.
Updated Coding section with 07/01/2022 CPT code descriptor for 0016M.
MPTAC review. Rationale, Background and References sections updated. Updated Coding section, added 81479 NOC code.
MPTAC review. Initial document development.
Federal and State law, as well as contract language, including definitions and specific contract provisions/exclusions, take precedence over Medical Policy and must be considered first in determining eligibility for coverage. The member’s contract benefits in effect on the date that services are rendered must be used. Medical Policy, which addresses medical efficacy, should be considered before utilizing medical opinion in adjudication. Medical technology is constantly evolving, and we reserve the right to review and update Medical Policy periodically.
No part of this publication may be reproduced, stored in a retrieval system or transmitted, in any form or by any means, electronic, mechanical, photocopying, or otherwise, without permission from the health plan.
© CPT Only - American Medical Association