Commercial Tests Offer Guidance in Bladder, Renal, Testicular Cancers

Brian Rini, MD

While biomarker-based tools designed to guide the diagnosis and treatment of prostate cancer are emerging at an extraordinarily rapid rate, they are not exploding in the same practice-changing way in all genitourinary cancers.

The development of molecular tests for bladder, renal, and testicular cancers is still in its infancy, but some tests have already reached the market, and doctors may want to consider the pros and cons associated with their use—while keeping an eye on promising biomarkers that could lead to the development of additional tools.

So far, most such tests in these disease states are focused in the area of bladder cancer. Some are diagnostic, while others are aimed at monitoring response to treatment. Meanwhile, in renal and testicular cancers, scientists are exploring a number of different biomarkers that may eventually provide guidance in both diagnosis and patient response to treatment.

Brian Rini, MD, a professor of Medicine at the Cleveland Clinic who specializes in genitourinary malignancies, is aware of that research, but says the clinical utility of biomarkers is less proven than in prostate cancer.

“These are four very different diseases. They are geographically related but quite different,” he said of prostate, bladder, renal, and testicular cancers. “In prostate cancer, there are some molecular tests around diagnosis, but other GU cancers really don’t have as much of an issue with diagnosis. The issue of overdiagnosis is not as big in the non-prostate cancers. As a result, if there are useful biomarkers in bladder, renal, or testicular cancers, they will be in matching specific drugs to specific patients, not so much diagnostic.”

Bladder Cancer

Bladder cancer is typically diagnosed by checking urine microscopically for the presence of blood via urinalysis, looking for abnormal cells via urine cytology, and/or through imaging and cystoscopy, according to the US Health and Human Services (HHS) in a report by its Effective Health Care Program. The report explores the research protocol for a planned review of “Emerging Approaches to Diagnosis and Treatment of Non-Muscle-Invasive Bladder Cancer.”¹

Biopsies for bladder cancer are not as difficult to interpret, and any lesion that exists can usually be visualized on cystoscopy, so additional diagnostic tests are not always considered necessary in the disease, for which there are no targeted treatments, Rini said.

“For bladder cancer, starting with non-muscle invasive disease, there are a number of urine cytology tests,” he noted. “Regular cytology would just look for cancer cells. Urine cytology, classically, is not always straightforward: The cells can look abnormal, but it’s hard to tell if they’re cancerous. There have been a number of assays that have attempted to develop other, more sensitive ways to look at cells that come out of that urine to determine if the patient has bladder cancer or recurrent bladder cancer, and there are some on the market.”

The HHS report cites a handful of urine-based molecular diagnostic tests that have been approved by the FDA for the diagnosis of bladder cancer or to guide the treatment of the disease, and some additional tests are also available.

Among FDA-approved tests, BTA stat (Polymedco) looks for bladder tumor-associated antigen (BTA) in office settings while sister product BTA TRAK identifies BTA in a laboratory; Alere’s NMP22 Test is an enzyme immunoassay that identifies nuclear mitotic apparatus protein in urine, and sister product Alere NMP22 BladderChek Test is a quick way to identify nuclear matrix protein 22 in-office; the Visys UroVysion Bladder Cancer Kit (Abbott Molecular) looks for chromosomal abnormalities via fluorescence in situ hybridization; and, according to HHS, ImmunoCyt (Scimedx) relies on monoclonal antibodies “to test for CEA and glycoproteins using an immunofluorescent technique.”

The CxBladder test (Pacific Edge Diagnostics USA), which does not require FDA approval, looks for five specific mRNA biomarkers. CertNDx (Physicians Choice Laboratory Services), also not approved by the FDA, is an assay of three tests: It uses DNA, immunohistochemistry, and protein biomarkers to assess hematuria, conduct molecular grading, and monitor recurrence, according to a description on the website of CGS Administrators, LLC, which provides Medicare-related services (http://tinyurl.com/n9sqvlg).

Investigated as possible alternatives to cytoscopy have been diagnostic cytology tests looking for fibroblast growth factor receptor 3; cytokeratin fragments such as CYFRA 21-1, TPA, and TPS; survivin; telomerase; vascular endothelial growth factor; aurora kinase; metalloproteinases MMP-2 and MMP-9; carcinoembryonic antigen; and mucin glycoproteins, the HHS states.

In more advanced bladder cancer, such as recurrent metastatic disease, Rini said, there are no blood- or tissue-based markers that are typically used to help predict responsiveness to chemotherapy or aggressiveness of disease.

“I’ve seen some data about tests to try to predict responsiveness to cisplatin, but in routine clinical practice, there’s nothing that we use,” he said. “There is a family of enzymes called ERCC1 that plays a role in repairing cells after chemotherapy damage, and people have looked at the tumor expression of this family of enzymes and how they will respond to cisplatin, which is commonly given in advanced bladder cancer [being ERCC1-positive may be a predictor of better response to treatment]. There is some data around that, but it’s not FDA-approved, and it’s not something I use in clinical practice. There are no prognostic biomarkers besides performance status and labs, but they are clinical, not molecular.”

On the other hand, many of the same biomarker-based urine tests being explored as diagnostic measures in bladder cancer may also turn out to be useful after treatment, to determine if bladder cancer has progressed or recurred, according to the HHS report. Approved in this space by the FDA’s Center for Devices and Radiological Health is the Vysis UroVysion Bladder Cancer Kit.²

Still, the report cautions that “The large number of available tests and testing strategies and potential trade-offs in diagnostic accuracy, risks, and patient preferences pose significant challenges in determining optimal testing and monitoring strategies. Tests with high false positive rates could lead to unnecessary invasive procedures for further evaluation and tests with high false negative rates could lead to missed diagnoses.”

Renal Cancer

As in bladder cancer, biomarker-based tools are not commonly used in the diagnosis of renal cancer, Rini said.

“It’s more the appearance of the mass on the CT scan,” he said. “It’s a disease diagnosed by imaging of the abdomen. If a mass is there, and there’s a question about differentiating between benign versus cancer based on how it looks on the scan, patients often go to biopsy or straight to surgery. But to my knowledge, there’s no blood-based marker to differentiate benign from malignant renal masses in routine clinical use.”

While there is an array of approved drugs for the treatment of advanced renal cancer—many of them targeted—physicians tend to use the medications based on empirical evidence during the course of the disease, Rini said, “because there are no blood- or tissue-based biomarkers reliably associated with response.”

Information about prognosis is usually generated using criteria established before the targeted-therapy era, by either Memorial Sloan Kettering Cancer Center or the Cleveland Clinic, or newer criteria developed by Heng et al in patients treated with targeted therapy, according to an article on the website of the 2015 Genitourinary Cancers Symposium written by Ulka Vaishampayan, MD, chair of the Genitourinary Multidisciplinary team at Karmanos Cancer Institute at Wayne State University.³

But biomarkers for prognostic risk, and to help guide treatment decisions, have been “extensively explored” in renal cancer, Vaishampayan wrote.

Many researchers are studying assays that might be useful in diagnosing or treating kidney cancer, with some working to develop tests meant to be sold commercially. At least one is available now: The Rosetta Kidney Cancer Test, approved by the New York State Department of Health and available throughout the United States, uses microRNA expression to differentiate between three types of renal cancer—clear cell, papillary, and chromophobe; it also identifies oncocytoma (http://www.rosettagenomics.com/pros-kidney).

Informing this type of work are various findings about biomarkers that appear important in the development and spread of renal cancer.

In the prognostic arena, Vaishampayan stated, it has been found that expression of hypoxia-inducible factor 1-alpha (HIF-1 alpha) at levels >35% “is associated with significantly worse survival rates (median survival, 13.5 months) when compared with lower expression (< 35%, median survival, 24.4 months), respectively (P=.005). In a multivariate analysis, HIF-1 alpha and carbonic anhydrase IX (CAIX) expression were the strongest independent prognostic factors in metastatic clear cell kidney cancer. The ccA and ccB subsets assessed by consensus clustering, on multigene analysis, have emerged as robust indicators of prognosis.^”

But those markers apply only to the anticipated course of the disease, and not to a patient’s response to therapy. Tests that predict response—possibly in the form of companion diagnostics that are marketed along with specific therapies—are sorely needed in renal cancer, Vaishampayan pointed out. In fact, the doctor has studied an assay based on multiplexed protein analysis of tumor tissue in renal cancer, finding that it “is capable of providing clinically applicable information to help guide therapy.”⁴

A current challenge in that area is the development of a biomarker that will predict a patient’s response to high-dose interleukin-2 (IL-2), Vaishampayan wrote. While the SELECT trial failed to definitively prove that CAIX can play a role as a predictor of response, the study offered glimmers of hope that it eventually could. Meanwhile, “VEGF levels, HIF-1 alpha and Von Hippel Lindau gene status in association with sunitinib therapy, and PTEN expression in association with temsirolimus have been explored and disappointingly revealed lack of predictive capacity of the respective markers,” she pointed out.

More promise has been associated with recent study results demonstrating that variants in genes related to angiogenesis may help identify patients who should not take anti-angiogenic therapy, such as pazopanib, because they will not experience any benefit, Vaishampayan wrote. Another report showed a link between levels of serum VEGF and neutrophil gelatinase-associated lipocalin and response to therapy with sunitinib, she stated.

Yet another novel approach, Vaishampayan noted, is to use side effects as predictors of response to treatment.

A patient who is taking axitinib or sunitinib and develops hypertension, for example, is likely to fare better than a patient who doesn’t experience that side effect, the doctor wrote.

“Patients who developed hypertension on sunitinib therapy demonstrated a 2- to 5-fold increase in response rates, as compared with those for patients without hypertension,” she wrote.

She cited a study of Rini’s that found that, in patients taking axitinib, those who developed systolic blood pressure in excess of 140/90 demonstrated a higher response rate and longer overall survival (OS) as compared with the non-hypertension group.

That held true in patients whose axitinib dose levels were above and below median, Vaishampayan added. “Patients with above-median levels had a median OS of 131 weeks compared with 43 weeks for those with below-median levels, and a median OS of 120 weeks and 42 weeks, respectively, was shown for those with and without hypertension. The neutrophil count was found to be a prognostic characteristic, with an elevated neutrophil count being a harbinger of worse outcome per the Heng criteria. Interestingly, neutropenia from sunitinib therapy was predictive of improved PFS [progression-free survival], regardless of dose adjustment of the therapy.”

Rini added that an exciting idea in the pipeline is the eventual treatment of renal cancer with checkpoint inhibitors, which are still experimental. As part of that effort, he said, “there are attempts to develop biomarkers based on the tumor expression of a molecule called PD-L1. There is some preliminary data from trials, but no FDA-approved drugs for kidney cancer yet that are affecting clinical practice.”

Testicular Cancer

When it comes to commercially available molecular diagnostic tests, the landscape of testicular cancer is also fairly barren—at least, so far, Rini said.

In this disease, he said, diagnosis is made by physical exam, so there are “no screens for these young, healthy men. There are serum blood tests for the tumor markers AFP [α-fetoprotein], HCG [human chorionic gonadotropin], and LDH [lactase dehydrogenase] that have been used for decades. They are incorporated into staging schemas and have been used in treatment to assess response to therapy. In general, all patients get the same chemotherapy with advanced disease, so you’re not choosing one therapy over another, but you are assessing the markers for response to therapy. If somebody has just finished chemotherapy and his scans look good but he has rising tumor markers, that patient likely would need more therapy.”

In general, Rini said he doesn’t see a dire need for biomarker-based tests in the pipeline for testicular cancer because “the response for this group of patients is generally so great, which is a good thing.”

But the FDA did approve an AFP-based test in 2013, the Architect AFP Assay (Abbott Laboratories), for use in monitoring disease progression in nonseminomatous germ cell cancers of the testis (NSGCT)—as well as for checking amniotic fluid for evidence of fetal open neural tube defects (http://tinyurl.com/owj5vxb).

There are also biomarkers beyond AFP, HCG, and LDH that may turn out to be helpful in classifying and treating the most common forms of testicular cancer in adolescents and adults—seminomas and NSGCT, according to the National Academy of Clinical Biochemistry (NACB) in its 2008 Guidelines for the Use of Tumor Markers in Testicular Cancer.⁵ These serum tumor biomarkers may be used in the “diagnosis, staging and risk assessment, evaluation of response to therapy and early detection of relapse,” according to the guideline.

The authors of a 2010 paper in the Indian Journal of Urology⁶ commented on why new biomarkers could be valuable.

“Although AFP, HCG, and LDH are the most commonly used serum markers for management of testicular germ cell cancer, these markers are not very specific and they are only detected in approximately 60% of men with testicular cancer,” the authors wrote. “In addition, the sensitivity of these markers is limited, and the levels of these markers are usually ‘normal’ in about 40% of men with disease recurrence.”

An early warning sign may come when a precursor to seminoma or NSGCT announces itself through the presence of carcinoma in situ (CIS) cells, which can be detected on biopsy, according to the NACB guideline. Once a tumor is present, classifying it as either seminoma or NSGCT can be aided by the detection of placental/germ cell alkaline phosphatase (PLAP), which is present in most seminomas and some NSGCTs, but almost never in a teratoma NSGCT, according to the guideline. However, the guideline authors noted that PLAP may be increased up to tenfold in smokers, which can lead to a falsely positive diagnosis and has represented an argument against the biomarker’s use.

“Of the other tissue markers, the stem cell factor receptor (c-KIT) has been used mainly to detect [carcinoma in situ] and seminoma, CD30 to detect embryonal carcinoma [a form of NSGCT], and AFP to detect yolk sac tumors and a 10-20% subset of embryonal carcinomas and teratomas [also forms of NSGCT],” the guideline authors added. “Recently, a potentially valuable marker OCT3/4, also known as POU5F1, has been identified,” although the authors of the paper in the Indian Journal of Urology noted that it is also expressed in normal adult stem cells, making it more difficult to use as a marker for cancer. “Neuron-specific enolase (NSE) is elevated in about 30-50% of patients with seminomas and less often in NSGCT patients,” the guideline authors continued, “but in spite of these promising results the use of NSE is limited.”

The Indian Journal of Urology authors added that analyzing expression profiles of high mobility group proteins HMGA1 and HMGA2 can help differentiate between testicular cancer types, as may SOX proteins.

When it comes to prognostication, potential predictors of metastatic spread in stage 1 NSGCT include vascular invasion, lymphovascular invasion, “predominantly embryonal carcinoma histology, the presence of undifferentiated cells, and the absence of yolk sac elements in the primary tumor,” those authors stated. In seminomas, predictors of aggressive disease might include CDK10, p53, MIB-1, and serum cell-free circulating mitochondrial DNA, they wrote, adding that tumor size and rete testis invasion can help predict relapse.

Tumors likely to demonstrate resistance to cisplatin may contain a defect that has resulted in a microsatellite instability, the guideline added, and a defect in caspase-9 function may also affect treatment sensitivity and resistance.

There are a number of ways to apply this knowledge of biomarkers in clinical practice, the guideline suggests.

In the diagnosis stage, physicians can consider the levels of the three gold-standard biomarkers. HCG tends to remain below 300 U/L in seminoma patients but can rise above 1000 U/L in NSGCT, particularly in choriocarcinoma. LDH is elevated in about half of patients with either seminoma or NSGCT. And elevated AFT can lead to the reclassification of a tumor from seminoma to NSGCT, the document states. Higher elevations of any of those three serum markers typically indicates a worse prognosis, and can affect treatment choices, the guideline authors added.

Since tumors classified as pure embryonal carcinoma are most likely to lead to early metastatic disease, determining the prognosis for patients with this condition is important, and there is evidence that “a pattern with high Ki67, low apoptosis, and low p53 is associated with better survival than other patterns,” the guideline states. “If these results can be confirmed, this could provide a tool for more precise tailoring of therapy.”

Since therapies for testicular cancer are so effective already, another useful focus for novel tests or assays would be to guide in the reduction of treatment toxicity for patients, Rini added.

Currently, he said, “there are none that we use clinically to determine which drugs to use.”

References

1. Effective Health Care Program. Emerging approaches to diagnosis and treatment of non-muscle-invasive bladder cancer. US Department of Health & Human Services website. http://effectivehealthcare.ahrq.gov/index.cfm/search-for-guides-reviews-and-reports/?productid=1941&pageaction=displayproduct. Published July 21, 2104. Accessed September 24, 2014.
2. Center for Devices and Radiological Health. Nucleic acid based tests. FDA website. http://www.fda.gov/MedicalDevices/ProductsandMedicalProcedures/InVitroDiagnostics/ucm330711.htm. Updated June 17, 2014. Accessed September 24, 2014.
3. Vaishampayan I. Prediction of response or survival in kidney cancer: the crystal ball of biomarkers. Website of the 2015 Genitourinary Cancers Symposium. http://gucasym.org/prediction-response-or-survival-kidney-cancer-crystal-ball-biomarkers. Accessed September 24, 2014.
4. Vaishampayan UN, Sethi S, Chen W, et al. Multiplexed tissue protein assay to predict patient response to targeted therapy in advanced renal cell carcinoma. J Clin Oncol. 2013;31(suppl 6; abstr 382).
5. Stenman U-H, Lamerz R, Looijenga LH, Bosl GJ. National Academy of Clinical Biochemistry guidelines for the use of tumor markers in testicular cancer. Clin Chem. 2008;54(12):e11-79.
6. Leman ES, Gonzalgo ML. Prognostic features and markers for testicular cancer management. Indian J Urol. 2010;26(1):76-81.