Bladder Cancer Biomarkers Aid in Diagnosis, Monitoring

Badrinath Konety, MD, MBA

When it comes to the burgeoning arena of biomarkers in bladder cancer, the value proposition is compelling, although the utility and costs need to be better defined, said Badrinath Konety, MD, MBA, department chair of urology at the University of Minnesota. He made the remarks at the 2016 Large Urology Group Practice Association (LUGPA) Annual Meeting.

In particular, Konety singled out biomarkers’ diagnostic consistency and the ways they can aid cystoscopy. “I use them to arbitrate atypicals and for post-intravesical therapy,” he said. He added that they also show potential in the areas of screening for bladder cancer, revealing other, unrelated cancers, and in distinguishing among bladder cancer grades.

These benefits contrast with the drawbacks associated with cytology, which include the need to send samples out to a lab, readings that can vary by pathologist, and a low degree of sensitivity, he said.

According to Konety, the use of bladder cancer biomarkers among urologists increased from 11% to 26% between 2000 and 2015. Among the more commonly used biomarker assays are the BTA/BTA Stat, NMP22 (BladderChek), UroVysion FISH, ImmunoCyt, and CxBladder.

But do biomarkers actually improve tumor detection?

Konety summarized a study from 2010 led by Madelon N. van der Aa published in the Journal of Urology, noting that it was a single-blind, randomized study using microsatellite markers. In the control arm, cystoscopy was performed at 3, 12, and 24 months without the positive or negative urine test results being communicated to the urologist. In the intervention arm, the urologist knew the marker result so that cystoscopy was performed on patients who tested positive on their urine test. At follow-up visits, recurrences were detected in 14.5% of patients in the intervention arm, versus in 10% of control-arm patients. The false positive rate in the intervention arm was 15%, compared to 20% in the control, showing a meaningful difference.

Konety described the BTA test as an in-office assay that identifies a complement H-related factor present in urine. Its sensitivity is 68%, while its specificity is 74%. He also noted that the NMP22 test detects nuclear mitotic apparatus protein (NuMA) with a sensitivity of 67% and a specificity of 75%. He noted that situations in which patients shed large numbers of benign cells can yield a high rate of false positives.

He then turned to Urovysion’s FISH test, which identifies the number of copies of chromosomes or regions of chromosomes with a sensitivity of 79% and a specificity of 88%. The ImmunoCyt assay consists of three monoclonal antibodies (19A211, M344 and LDQ10) labeled with two fluorescent markers. It has a sensitivity of 77% and a specificity of 76%.

There can be problems related to the FISH test. For example, false negatives can be caused by degenerated cells, presence of a hyphal fungus, contaminated squamous cells present only, excessive lubricant, or autofluorescent bacteria.

False positives, on the other hand, can appear due to umbrella cells, polyoma virus infection with inclusions, and, rarely, seminal vesicle cells. Tetraploidy in reactive cells can also confound the results. Fortunately, correlation with cytology findings can exclude false positives caused by these complicating factors. Additionally, FISH testing is insufficient for the upper tract and situations with urinary diversion.

According to Konety, FISH testing is also useful after Bacillus Calmette-Guerin (BCG) treatment. “A post-BCG FISH predicts for both recurrence and prognosis,” he said. “FISH is also useful if post-BCG cytology is equivocal.”

Konety also discussed what he called “second-level” biomarker assays, those that are most helpful after atypical cytology to guide further care. One example is CxBladder, developed by New Zealand’s Pacific Edge, which Konety said has been validated in patients with hematuria. Using mRNA expression by PCR, CxBladder tests for CDC2, HOXA13, MDK, IGFBP5, and CXCR2. The test has shown a sensitivity of 97% for high-grade tumors and 69% for low-grade tumors, making it “better than cytology” (83% for high-grade and 28% for low-grade tumors): “CxBladder has a couple papers in review so it will be interesting to see how those data pan out.”

Methylation profiling is another area with the potential for clinical applicability. In a 23-gene set, it was effective on 4: APC alpha, TERT alpha, TERT beta and EDNRB, with an area under the curve (AUC) between 0.69 and 0.82, a sensitivity of 63% and a specificity of 58%. Other genes that have been studied include RASSF1A, p14, TWIST1, and NID2.

“But there are issues with methylation profiling due to their small discovery and validation sets,” Konety said. “Most have focused on hematuria evaluation and seem to be more closely associated with grade and stage.”

Konety, who served on the American Urological Association's International Consensus Panel, discussed its recommendations. Marker-guided follow-up and screening of patients with low-grade non-muscle invasive bladder cancer (NMIBC) were both deemed “attractive,” but not recommended at this time. The panel does not recommend biomarkers in the surveillance of high-grade NMIBC. “Consider that 75% of patients want a test with a sensitivity of at least 95% in order to be willing to forego cystoscopy, as uncomfortable as that test is,” he said.

Konety also reviewed the recently updated guidelines from the American Urological Association (AUA).¹ Key points included:

• In surveillance of NMIBC, a clinician should not use urinary biomarkers in place of cystoscopic evaluation.
• 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.
• In a patient with NMIBC, a clinician may use biomarkers to assess response to intravesical BCG (UroVysion FISH) and adjudicate equivocal cytology (UroVysion FISH and ImmunoCyt).

“The AUA is most concerned with specificity and the issue of false positives,” he said. “Biomarkers’ false positive rate is unclear and we want to avoid additional harms to our patients.”

As for what’s in the pipeline, Konety mentioned several promising biomarkers. These include Genomic Health, MDx (Assure MDx), Genome Dx, CertNDx (Predictive Biosciences), COXEN, Foundation Medicine, and Xpert (Cepheid), which has already been released in Germany. “We’re clearly headed toward more complicated tests that will give us a more sophisticated view of what’s happening with our patients,” he said. “We’re currently waiting to see which genes pan out. The bottom line is that new markers need large studies, and we’re not there yet.”

Chang SS, Boorjian SA, Chou R, et al. Diagnosis and treatment of non-muscle invasive bladder cancer: AUA/SUO Guideline. J Urol. 2016; 196(4):1021-1029. doi: 10.1016/j.juro.2016.06.049.

<<<

View more from the 2016 LUGPA Annual Meeting