As biomarker testing for genitourinary (GU) pathology plays catch up with the well-established, and well standardized, landscape in lung and breast pathology, we ask: what are the barriers to the ideal workflow? What happens when test manufacturers’ marketing targets oncologists rather than pathologists? Why are tests being ordered and repeated unnecessarily? Why should pathologists be the stewards of all molecular testing?
These issues and more were addressed by our roundtable panel of four GU pathologists in the US. Here is what they told us…
Meet the Panel
Ming Zhou - Vice Chair for Oncological Pathology, Director of Urological Pathology Service and Fellowship Program, Mount Sinai Hospital and Icahn School of Medicine, New York
Fang-Ming Deng - Professor of Departments of Pathology and Urology, Director of Urologic Pathology, New York University Langone Health
Manju Aron - Professor of Clinical Pathology and Urology, Section Director, Urologic Pathology, Keck School of Medicine, University of Southern California
Anil Parwani - Chair of the Department of Pathology, Ohio State University
How has next-generation sequencing (NGS) impacted the diagnostic and prognostic landscape for GU cancers?
MZ: NGS has significantly improved our ability to characterize cancer at the genomic level. For both prostate and bladder cancers, we've started using NGS for prognostic stratification and to guide targeted therapies. For instance, in prostate cancer, NGS can identify actionable alterations – particularly those with therapeutic and prognostic significance.
We know that tumors with mutations in homologous recombination repair genes, for example, may respond well to PARP inhibitors. Similarly, tumors with microsatellite instability are often responsive to immune-oncology based therapies like immune checkpoint inhibitors. Because NGS can pinpoint those types of tumors, it has transformed the therapeutic landscape in prostate cancer. This is particularly relevant in advanced stages, such as castration-resistant prostate cancer, where tumor-based NGS is frequently ordered to help guide treatment decisions.
In some cases, we also pursue germline testing. If a patient carries a germline BRCA2 mutation, for example, there are important implications. Screening should begin earlier – around age 40 instead of 55 for the general population – and if such patients undergo local therapy, they may face a higher risk of disease progression and potentially lower overall survival.
These factors should be discussed with patients, especially if they are considering active surveillance.
How are NGS tests typically ordered and implemented in GU pathology?
AP: In general, the tests we use are fairly generic – they’re designed for a broad range of cancer types. We don’t often have panels that are specific to GU cancers like prostate or bladder cancer. These are usually included as part of a broader solid tumor or pan-cancer panel, but not specifically highlighted or customized for GU applications.
There’s also a recurring communication gap between GU oncologists, urologists, and pathologists. Often, oncologists attend national meetings like the American Society of Clinical Oncology (ASCO) or regional society conferences, hear about new tests, and then come back wanting to order them. But not many labs are set up to offer those tests, and pathologists aren’t always looped in.
Some academic centers like ours do have in-house NGS capabilities, but many community pathologists rely on send-out testing. When that's the case, the ordering process can bypass the pathologist altogether. That’s a concern.
MZ: Yes, I agree. Even at our hospital, most of these tests are still being sent out. We recently validated a 500-gene panel that was approved by the New York State Department of Health – but it’s not yet in routine clinical use.
In practice, most NGS requests are still sent to large commercial laboratories. First, these tests are costly. Second, the turnaround time can be lengthy – at least two weeks. That can negatively impact patient care, because our clinicians are looking for results quickly to make timely treatment decisions.
F-MD: As Ming and Anil mentioned, most of the time we’re using general-purpose panels. And for identifying prognostic markers or therapy targets, that’s usually sufficient. But it can also add real value in challenging diagnostic cases. For example, when we’re dealing with carcinoma of uncertain origin – where we’re not sure about the exact cell lineage – NGS can sometimes provide additional clarity.
How well standardized are these tests across labs?
MZ: Since most of these tests are performed by centralized commercial labs, there’s generally good consistency in how they’re carried out. That centralization helps reduce variability across testing sites.
The challenge arises when we try to bring these tests in-house. Standardizing them internally is extremely complex and resource-intensive. That’s where we face significant hurdles – in validation, quality control, reproducibility, and regulatory approval.
Does faster turnaround time come at the cost of consistency or quality?
MA: Yes, that’s the trade-off. On the one hand, faster turnaround times are ideal for patient care. Some institutions – MSK, for example – do have their own in-house molecular testing platforms, which I presume are well standardized and tightly integrated into clinical practice. But at most small and even mid-sized academic centers, like ours, oncologists tend to send specimens to large outside reference labs and then base clinical trials or therapy decisions on those results.
But as Anil mentioned earlier, most of these tests are being ordered by oncologists. And there’s the issue of reimbursement – oncologists often order the tests directly and manage the billing and reimbursement process. They typically send specimens to large commercial labs, which offer validated, standardized platforms for clinical use. Based on those results, they can move forward with targeted therapies.
F-MD: Another issue – possibly more relevant to New York – is patient-driven test duplication. For example, the oncologist may send the tumor sample to the commercial lab. Then the urologist says, “Oh, we have our own 580-gene in-house panel,” and orders another test. Then the patient visits Memorial Sloan Kettering, and they run their own in-house platform again.
That kind of repeated testing adds unnecessary cost and effort. It can be a waste of both time and tissue.
MZ: That brings up a critical point – whether results across different platforms are actually comparable. And more importantly, what is the role of the pathologist in this process?
In many situations, clinicians bypass us completely. They send tissues directly to commercial labs without involving pathology. But our role is essential. As Manju mentioned, we need to ensure the tissue being submitted is representative of the tumor and that there’s sufficient tumor content to yield reliable results.
And when results come back, I always emphasize that pathologists should be involved in interpreting them. We should integrate the molecular findings with the pathology report and the broader clinical context. That’s key to ensuring results are meaningful and actionable.
To what extent are liquid biopsy tests being used in the detection and monitoring of GU cancers?
AP: In my experience, we’re not seeing many liquid biopsy requests for GU cancers coming through pathology. The companies offering these assays market directly to oncologists, promising rapid turnaround times. From a pathology standpoint, it feels more like a clinical chemistry test – it doesn’t offer any spatial perspectives on disease.
Right now, in my practice, there’s little demand from urologists or oncologists for liquid biopsy testing. But I suspect that may change as adoption grows. There are potentially actionable applications, especially in bladder cancer, but I’ll let my colleagues weigh in further.
MZ: Yes, I agree. At the moment, liquid biopsy testing – such as circulating tumor DNA (ctDNA) analysis – completely bypasses pathology. Blood samples are drawn in the clinic and sent directly to external testing labs without any involvement from the pathology department.
F-MD: That’s exactly what we’ve observed. Like Anil said, we’re not seeing many of these tests directly – but that doesn’t mean they aren’t happening. For example, in advanced bladder cancer, liquid biopsy is used before neoadjuvant therapy to assess disease burden. It’s also commonly used for post-treatment monitoring and recurrence detection.
MA: And that’s the real issue – it’s happening, but we’re not included in the process. These tests are ordered during routine clinical follow-up, often in outpatient settings. We only learn about them during multidisciplinary team meetings, when someone mentions ctDNA levels while discussing patient management.
MZ: Let me give you an example from when I was at Tufts in Boston. A patient with recurrent lung cancer had blood drawn for ctDNA testing, and the sample was sent directly to a third-party lab. The results were faxed to the ordering physician but never uploaded into EPIC. The physician didn’t receive them, and when the results were needed, no one could locate them in the system.
We were asked to investigate, and we discovered the test had been ordered outside of our standard workflow. The pathology department had no record of the test ever being ordered or reported. Worse, we found out the testing lab wasn’t even CLIA-certified. That’s a serious concern. Without proper oversight, important results can fall through the cracks.
What are the barriers to using urine-based biomarker testing in monitoring GU cancers?
F-MD: I think urine-based testing has significant potential. There are already many commercially available tests – not necessarily in the US, but certainly in Europe, China, and India. The issue, though, is that there are so many options. Choosing the right one becomes a challenge.
More importantly, these tests need to be rigorously validated. At this stage, a lot of companies are pushing aggressively into the market, but the scientific and clinical validation is still catching up.
MZ: I completely agree with Fang-Ming. There’s huge potential in using liquid biopsies – like ctDNA and urine-based tests – especially since these are considered noninvasive or minimally invasive techniques. That makes them attractive for both patients and providers.
However, as Fang-Ming mentioned, the problem is fragmentation. There are so many different tests, and most are only validated within limited patient cohorts. We don’t yet have broad, cross-institutional validation studies. That makes it difficult to interpret and compare results across centers.
Another major barrier is reimbursement. Not all of these assays are covered by Medicare or commercial insurance, so financial feasibility becomes a concern. Plus, with so many options on the market, selecting the most appropriate test often depends on the individual urologist’s experience and preference rather than standardized guidelines.
F-MD: Yes, and another important limitation is the size of the gene panels used in many of these assays. A lot of current urine-based tests are built on very narrow panels. That means sensitivity can vary widely depending on tumor type.
For example, some of the newer iterations claim improved sensitivity for bladder cancer. But for other GU malignancies, such as kidney or adrenal tumors, sensitivity remains relatively low.
Looking ahead, we’ll likely need to expand these gene panels or incorporate additional biomarkers to improve test performance. Sensitivity, in particular, still poses a significant challenge.
Is the situation similar for epigenetic biomarker testing in GU pathology?
MA: It’s definitely an evolving area. We’re starting to see more epigenetic markers identified – especially in bladder cancer – that appear to have predictive value. The field is growing, and I expect these assays to become more widely adopted in clinical practice.
In our institution, when oncologists request epigenetic or molecular testing, those orders typically come through pathology. As Ming mentioned earlier, we’ve set up a system to ensure appropriate tissue handling – verifying tumor content, confirming availability of adjacent normal tissue when required, and meeting all the pre-analytical criteria specific to each test.
Once results are received, they’re routed back to us. We log them into our laboratory information system, append them as addenda to the original pathology report, and ensure they’re integrated into the patient’s electronic health record. That feedback loop is important – it helps us understand how these tests are being used and whether they’re delivering value for patient care.
This applies not only to epigenetic markers, but also to NGS more broadly – for example, in urothelial tumors. These baseline assays are often done before patients begin neoadjuvant chemotherapy. Since the tissue is routed through us, we retain some level of oversight – unlike with liquid biopsy, which typically happens in outpatient clinics and bypasses pathology altogether.
That said, I don’t have much direct experience with clinical epigenetic testing. We do have a strong research lab that works with methylation markers, but the clinical use in our setting is still limited. There are commercially available tests out there, but uptake is variable.
MZ: I recently had a question from a sales representative about orders of their assay – an epigenetic test used for patients with persistently elevated PSA but negative prostate biopsies. The idea is to help assess cancer risk in these ambiguous cases.
The rep emailed me asking why orders from our institution had dropped significantly. They wanted to know if we’d stopped using the test altogether. So I asked our urologists – and the answer was simple: they’ve started using other tests that they believe offer better performance or clinical value.
That really highlights the competitive and shifting landscape in this space. There are multiple assays available, and no one test has a permanent hold on the market. Clinicians will move on if they find alternatives they trust more, or that offer better validation, ease of use, or reimbursement.
So yes, epigenetic testing has potential, but its clinical adoption is shaped by many dynamic factors – including test performance, provider preference, and market competition.
How do you view the broader landscape of biomarker and NGS testing in GU cancers – and the role of pathologists within it?
F-MD: Unfortunately, pathologists are often left out of the process. But we have to advocate for a more central role – particularly in specimen selection and quality control.
Let me give an example: with a proprietary assay for prostate cancer, commercial labs may request specific blocks – often the one with the highest Gleason score. But if that area is very small – say, less than 0.5 mm – it may not be a good sample for molecular testing, especially if it comes from a targeted biopsy.
Even if there’s another block with a larger volume of Gleason scoring tumor, the lab may insist on the highest-grade specimen. That often leads to test failures and repeated attempts. As pathologists, we understand which block is most appropriate. We’re better positioned than commercial labs to select the right specimen – and that’s where our expertise must be recognized.
AP: This is a global issue – not just a GU pathology problem. Biomarker and NGS testing are facing similar challenges across tumor types.
Imagine taking your car to the shop for a broken fuse, and the mechanic tells you the only solution is to replace the entire engine. That’s what’s happening right now in molecular testing. These broad panels are being marketed directly to oncologists, with minimal involvement from the pathology teams.
Companies are attending oncology meetings to promote these tests. They’re even starting to show up at pathology conferences – but their marketing is still largely directed at oncologists.
Every test report comes back with 500 or 600 genes analyzed, and the result is a 30-page document. No one has time to fully interpret it. And often, the clinical question is very specific – say, FGFR3 mutation status. That’s all the oncologist or urologist wants to know. But the report doesn’t answer that in a straightforward way. Instead, it creates noise and redundancy.
Sometimes, the same broad panel is ordered again, for the same patient, at another institution. What we need is better collaboration between oncologists, urologists, and GU pathologists. This isn’t just true for GU – it applies in GI pathology, lung pathology, and elsewhere.
The difference is that in lung cancer, melanoma, and colorectal cancer, there are clearer testing guidelines. GU feels like the “orphan child” of molecular testing. We lack both harmonized protocols and a coordinated multidisciplinary approach.
MA: I completely agree. As Anil pointed out, the fundamental issue in GU is that the field is still evolving. The data we have now are mostly from ongoing or early-phase clinical trials. It’s not mature enough to draw firm conclusions.
That makes it hard to say, “This specific gene reliably predicts survival” or “This mutation correlates with recurrence-free survival.” Until that evidence base matures, we can’t identify the most clinically meaningful alterations the way we can in lung cancer, for example.
Bladder cancer adds another layer of complexity. Intertumoral heterogeneity is a major issue – different areas of the same tumor may have very different molecular profiles. And we also have multiple competing molecular classification systems. It’s a very fragmented space.
So yes, it’s all still in development. To move forward, we need two things: first, clearer data on which alterations truly affect outcomes like survival or recurrence; and second, testing strategies that are tailored to those validated targets. Until then, GU molecular pathology remains a moving target.
What needs to change in order to improve the molecular testing workflow?
MZ: We need more involvement from pathology. That’s the starting point. But it’s not just about individuals – it has to happen at the hospital policy and regulatory level as well.
Anil, as a department chair, will appreciate that institutions should have clear policies that designate pathologists as stewards of all tissue- and fluid-based molecular testing. Regardless of who orders the test, the process should require pathologist oversight.
That means before any specimen is sent to an outside lab for testing, it should go through pathology. And once the results come back, they should return to pathology, be incorporated into our laboratory information system LIS, and be integrated into the final pathology report. That kind of structured workflow ensures both accountability and clinical relevance.
AP: I completely agree. But the reality is – it’s easier said than done. Anil already mentioned the marketing strategies for commercial tests. These clinicians are being heavily influenced to order tests that may not even be clinically necessary or meaningful in the setting.
MZ: True – and we as pathologists also bear some responsibility. In some cases, we’ve been hesitant to get involved because we feel we’re not adequately trained. We say, “I wasn’t trained to validate or interpret these tests,” and so we step aside.
But I think that’s a dangerous mindset. That reluctance creates a vacuum – and others step in to fill it.
AP: Exactly. This is just like what happened with earlier technologies – like PCR and even NGS itself. Initially, these methods weren’t reimbursed. Mainstream pathology labs didn’t want to invest in them. The infrastructure wasn’t there, and the costs were high.
So what happened? Commercial entities moved in to fill that void. And now we’re trying to reclaim the ground we gave up. Every time a new technology comes along – AI, NGS, liquid biopsy – you either take ownership of it, or someone else will. And if we don’t, we can’t really blame anyone but ourselves.
What can pathologists do to effect that change and ensure they have oversight of molecular testing?
AP: I think pathologists need to act on multiple fronts. It's not just about being involved in hospital-level test committees or engaging with C-suite leadership. We also need to be visible and active on the national stage – through our professional societies.
Within the GU pathology community, groups like the Genitourinary Pathology Society (GUPS), ISUP, CAP, and others should be issuing unified guidance. We need clear recommendations – best practices – for biomarker testing in GU cancers. Just as we’ve developed consensus guidelines for morphologic diagnosis, we now need to create structured frameworks for molecular testing.
If we don’t step up, we’ll continue to see a proliferation of companies offering redundant or unvalidated tests – like PTEN assays or FGFR3 panels – without any standardization. And once again, pathology will be sidelined.
This is an opportunity. We should be collaborating with our GU oncology colleagues and hospital stakeholders to ensure we’re shaping the future of testing – not reacting to it. Every time I attend a tumor board or a biomarker advisory meeting, I advocate for this.
We, as pathologists, must assert control over which tests are ordered, how they’re interpreted, and how results are integrated into clinical care. No one understands the morphology and features of diseases like prostate and bladder cancer better than we do.
If we want a seat at the table, we can’t just complain about being excluded – we need to step forward, show leadership, and claim our role as stewards of molecular diagnostics.
Newsletters
Receive the latest pathologist news, personalities, education, and career development – weekly to your inbox.
