Bladder Cancer BioMarkers: Optimal Utilization for Diagnosis and Recurrence
- Published in Everyday Urology - Oncology Insights: Volume 1, Issue 3
Voided urine cytology has been the gold standard for detecting bladder cancer since 1945. Its specificity nears 90%, meaning that a positive result is highly reliable. But cytology is unreliable for detection of low grade tumors such that only about 20% to 30% of low grade bladder tumors are identified using cytology. Furthermore, up to one-third of cytology specimens are considered atypical because of confounding factors such as instrumented sampling or recent therapy with bacillus Calmette-Guérin.13
By adding urinary biomarker assays to their diagnostic workup, urologists can overcome the weak points of cytology without sacrificing its powerful specificity. Evidence supports this combined approach: patients with non-muscle invasive bladder cancer (NMIBC) who undergo both cytology and urine marker testing have about 20% better survival than those who do not. Additionally, in a randomized trial, urologists performing cystoscopy identified significantly more recurrent NMIBC tumors when they knew patients had tested positive for microsatellite markers than when they did not (14.5% vs. 10.0%; p<.001), and without increasing their rate of false positives1
The use of urinary marker tests more than doubled in the United States between 2001 and 2009. That trend has continued, highlighting the potential for these assays to help evaluate therapeutic response, inform diagnostic decisions in the face of atypical cytology results, and identify low-grade tumors, including carcinoma in situ (CIS), which white light cystoscopy can miss. Newer marker assays also can shed light on tumor biology and may one day help individualize therapy for bladder cancer.
The NMP22 (BladderChek) test detects nuclear mitotic apparatus protein (NuMA). This test has a 67% sensitivity for CIS, substantially exceeding that of cytology.2 However, its specificity is only 65%.3 As a protein-based assay, BladderChek is subject to false positives if a sample contains excessive numbers of benign cells, which can result from instrumented sampling or non-malignant inflammation. False-positive rates were 74% with instrumented sampling and 38% otherwise (p<.0001) in a large study of patients with hematuria.4 When inflammation was present, the false-positive rate of 85% (vs. 61% otherwise; p<.0001).
The bladder tumor antigen (BTA) and BTA Stat tests detect complement factor H-related protein, which prevents activation of the complement cascade. Their performance resembles that of the NMP22 assay. These tests are no longer in common use.
The UroVysion test uses multicolor fluorescence in situ hybridization (FISH) to identify loss of the 9p21 locus -- the first change to occur in bladder cancer -- and aneuploidy for chromosomes 3, 7, and 17. A positive result is based on detection of at least 12 cells with 9p21 loss, or at least four cells with aneuploidy of multiple chromosomes. Sensitivity is approximately 79%, and specificity is 88%.
FISH is also useful for post-BCG surveillance. A positive result before initial BCG instillation that remains positive post-BCG is associated with about a two-fold higher likelihood of cancer recurrence, and with a higher probability of progression, than if a positive baseline FISH becomes negative post-BCG.
FISH results are considered aberrant when they do not reflect the findings of cystoscopy or upper urinary tract studies. False negative results are associated with shedding of degenerated cells, fungal hyphae, squamous cells, autofluorescent bacteria, and exposure to lubricant. False positives result from normal multinucleated umbrella cells, virus inclusions, recent radiotherapy, cellular tetraploidy, or seminal vesicle cells.10. The presence of only four copies of a specific chromosome also is typically interpreted as a false positive. Urinary diversion specimens and those from the upper tract can be insufficient for FISH analysis.
Clinicians can use cytology to help clarify aberrant FISH findings. If cytology is negative, FISH could be obtained and characterized as high or low risk based on the specific chromosomal aneuploidy. Aneuploidy of chromosomes 7 and 17 predicts bladder cancer and is considered a high-risk FISH result. If a result is positive only because of 9p21 loss or gain of chromosome 3, this is considered low risk and can generally be ignored. Loss of 9p21 is the least predictive of an actual bladder tumor amongst the four markers test in the Urovysion test.12
The ImmunoCyt test uses fluorescent-labeled monoclonal antibodies to identify glycoprotein antigens 19A211, M344, and LDQ10, which are shed by malignant exfoliated urothelial cells. A test is considered positive if even one of 500 cells has at least one of these markers.
Both ImmunoCyt and UroVysion (FISH) are useful as second-level tests to help arbitrate atypical cytology results. Patients with atypical cytology, equivocal cystoscopy, and negative FISH have a 92% chance (negative predictive value) of negative results on further diagnostic workup. When cytology is atypical and cystoscopy is negative, the NPV of FISH increases to 100%. ImmunoCyt performs similarly, with NPVs of 94% when there is no bladder cancer history and 82% when there is a history of low-grade bladder cancer.16
The CxBladder assay quantifies messenger RNA levels for the HOXA13, CDC2, IGFBP5, MDK, and CXCR2 marker genes. The CXCR2 gene functions in nonmalignant inflammation, and is included to identify false positives.8
In a study of patients with gross hematuria but no history of bladder cancer, CxBladder detected 82% of urothelial tumors overall, 97% of high-grade tumors, and 100% of tumors that were at least stage 1, given a predetermined specificity of 85%.8 When plotting sensitivity against specificity, the area under the receiver operating characteristic curve (AUC) was 0.87, indicating that the test correctly classified 87% of cases and controls. CxBladder also had markedly higher sensitivity than its forerunner, the uRNA assay, and when compared to NMP22. Data on using CxBladder to monitor recurrence are forthcoming.
METHYLATION AND GENE EXPRESSION PROFILING
The ASSUREMDx test, which is used for evaluating hematuria, analyzes three different gene mutations as well as DNA methylation of the TWIST1, ONECUT2 and OTX1 genes. Methylation of the promotor regions of these genes masks their “on” switch, leading to downregulation of tumor suppressor activity.11 This test yielded a high sensitivity and specificity in small data sets, with an area under the curve (AUC) of 0.95.
The CertNDx assay also is available for hematuria workup. This test analyzes TWIST1 and NID2 gene methylation, FGFR3 mutation, and expression of MMP2 protein, and showed high sensitivity and specificity and a 98% NPV in a study of 698 patients.
Urinary gene expression is also being explored for NMIBC detection and staging. In a validation study, a genetic signature consisting of 12 diagnostic genes and two genes for tumor aggressiveness had a sensitivity of 80% and a specificity of 86% for distinguishing cancer patients from controls. The test also distinguished high-grade and low-grade tumors with an AUC of 0.83.15 Additional work yielded simpler 2, 5, and 10-gene signatures, all of which included IGF2 and MAGEA3, with similar or superior diagnostic performance.15 Perhaps most notably, a streamlined 6-gene signature distinguished between low-grade and high-grade tumors with an AUC of 0.91.
Thus far, studies indicate that only analyzing methylation is less effective than also testing for gene and protein-level changes. Future tests probably will involve a combination of assays for methylation, gene mutations, and protein expression.
COST-EFFECTIVENESS AND USE
The increasing number of marker assays raises questions about balancing their clinical utility and costs. Combining flexible cystoscopy, ImmunoCyt, and photodynamic diagnosis (PDD) yielded the highest rate of true positives and the best survival in a meta-analysis.7 Cytology followed by white light cystoscopy was least effective. Among individual tests, sensitivity was highest for ImmunoCyt, while specificity was highest for cytology.
Other studies indicate that combining cystoscopy with FISH is significantly more effective for tumor detection than white light cystoscopy alone, but that the combined approach costs nearly three times as much.6 FISH can save substantial costs, however, when used to guide decisions about whether to pursue further workup. In a study of patients with atypical cytology and equivocal or negative cystoscopy, in-office biopsies decreased by 68% if urologists performed them only when FISH was positive.14 This approach also cut operating room biopsies by 88%. In each case, the strategy led to significant per-patient cost savings.
Urinary biomarker tests should not replace cystoscopy for NMIBC surveillance, according to 2016 guidelines from the American Urological Association (strong recommendation; moderate-quality evidence).5 The guidelines also state that clinicians also should not routinely use markers or cytology for surveillance of patients with normal cystoscopy and a history of low-grade bladder cancer. However, marker tests can be used to assess response to intravesical BCG (UroVysion FISH) and to adjudicate equivocal cytology (UroVysion FISH and ImmunoCyt) in patients with NMIBC.
These conclusions largely reflect those of an International Consultation on Urologic Diseases (ICUD) Consensus Panel, which described reflex marker testing after atypical cytology as an interesting approach, but did not endorse the use of markers for screening or to guide NMIBC follow-up9. Patients also have echoed these recommendations, stating in several studies that they would not forego cystoscopy unless the sensitivity of a marker test exceeded 95%.
The pipeline for urinary bladder cancer markers is extensive. As examples, methylation and gene profiling tests are in development by Genomic Health, MDxHealth (Assure MDx), Genome Dx, the Southwest Oncology Group (COXEN), and Foundation Medicine (whole genome sequencing). These tests aim not only to diagnose disease, but also to assess therapeutic response, and are hoped to offer features such as high NPVs in patients with asymptomatic microhematuria.
In summary, urinary markers allow for more accurate bladder cancer detection, but this gain is offset by moderately high rates of false positives. For now, these tests are best used to arbitrate atypical cytology results, assess BCG response, and guide post-BCG follow-up. Their costs and performance need more study before they are used for routine hematuria evaluation, cancer surveillance, or upper urinary tract monitoring. Profiling tests in the pipeline exciting, but will need further validation before entering routine urology practice.
Written by: Badrinath Konety, MD, MBA, Professor and Chair of the Department of Urology and holds the Dougherty Family Chair in Uro-Oncology as well as being the Director of the Institute for Prostate and Urologic Cancers and Associate Director for Clinical Affairs of the Masonic Cancer Center at the University of Minnesota
1. van der Aa MN, Steyerberg EW, Bangma C, et al. Cystoscopy revisited as the gold standard for detecting bladder cancer recurrence: diagnostic review bias in the randomized, prospective CEFUB trial. J Urol. 2010;183(1):76-80.
2. Landman J, Chang Y, Kavaler E, et al. Sensitivity and specificity of NMP-22, telomerase, and BTA in the detection of human bladder cancer. Urology. 1998;52(3):398-402.
3. Konety BR, Gertzenberg RH. Urine based markers of urological malignancy. J Urol. 2001 Feb;165(2):600-11.
4. Todenhöfer T, Hennenlotter J, Kühs U, et al. Influence of urinary tract instrumentation and inflammation on the performance of urine markers for the detection of bladder cancer. Urology. 2012;79(3):620-624.
5. Chang SS, Boorjian SA, Chou R1, et al. Diagnosis and treatment of non-muscle invasive bladder cancer: AUA/SUO guideline. J Urol. 2016;196(4):1021-1029.
6. 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):1119-1123.
7. 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, iii-iv.
8. O'Sullivan P, Sharples K, et al. A multigene urine test for the detection and stratification of bladder cancer in patients presenting with hematuria. J Urol. 2012;188(3):741-747.
9. Kamat AM, Hegarty PK, Gee JR, et al. ICUD-EAU International Consultation on Bladder Cancer 2012: Screening, diagnosis, and molecular markers. Eur Urol. 2013;63(1):4-15.
10. Tapia C, Glatz K, Obermann EC, et al. Evaluation of chromosomal aberrations in patients with benign conditions and reactive changes in urinary cytology. Cancer Cytopathol. 2011;119(6):404-410.
11. Ibragimova I, Dulaimi E, Slifker MJ, et al. A global profile of gene promoter methylation in treatment-naïve urothelial cancer. Epigenetics. 2014;9(5):760-773.
12. Bonberg N, Taeger D, Gawrych K, et al. Chromosomal instability and bladder cancer: the UroVysion(TM) test in the UroScreen study. BJU Int. 2013;112(4):E372-82.
13. VandenBussche CJ, Sathiyamoorthy S, Owens CL, et al. The Johns Hopkins Hospital template for urologic cytology samples: parts II and III: improving the predictability of indeterminate results in urinary cytologic samples: an outcomes and cytomorphologic study. Cancer Cytopathol. 2013;121(1):21-8.
14. Gayed BA, Seideman C, Lotan Y. Cost-effectiveness of fluorescence in situ hybridization in patients with atypical cytology for the detection of urothelial carcinoma. J Urol. 2013;190(4):1181-1186.
15. Mengual L, Ribal MJ, Lozano JJ3, et al. Validation study of a noninvasive urine test for diagnosis and prognosis assessment of bladder cancer: evidence for improved models. J Urol. 2014;191(1):261-269.
16. Odisho AY, Berry AB, Ahmad AE, et al. Reflex ImmunoCyt testing for the diagnosis of bladder cancer in patients with atypical urine cytology. Eur Urol. 2013;63(5):936-940.