Expert Discusses Biomarker Development in GU Cancers

William Oh, MD

Although new treatments with immunotherapy and targeted therapy continue to push the field of genitourinary (GU) cancers forward, oncologists are still looking for biomarkers that can help them determine which treatment is right for which patient.

“Two of the most promising are AR-V7, which is a truncated form of the androgen receptor, and BRCA2 mutations, which may determine whether patients are benefitting from either androgen targeted therapies, PARP inhibitors, or platinum chemotherapy,” said William Oh, MD.

OncLive: Can you give an overview of the presentation?

How are those decisions made?

In an interview with OncLive ahead of his presentation at the PER^® 1st Annual International Congress on Oncology Pathology™, Oh, chief, Division of Hematology and Medical Oncology, professor of Medicine and Urology, Mount Sinai School of Medicine, reviews some of the biomarkers under investigation in GU cancers and possibilities for major advances in the field over the next few years.Oh: I will be speaking at today's meeting about how the use of biomarkers helps oncologists decide how to treat patients with advanced GU cancers, and whether we can use molecular changes in tumors to help us better treat patients with GU cancers.As everyone knows, we are in a molecular revolution. There is a lot of new data on genomic alterations in cancer but the real problem for oncologists is that we haven't yet implemented this in our day-to-day practice. In prostate cancer, we have many new therapies, such as hormonal type therapies, chemotherapies, bone-targeted therapies, and immunotherapies, but we don't yet know how to choose which patient should receive which treatment. However, this is starting to change.

What challenges come with trying to decide which therapy is best for an individual patient?

I am going to be discussing some of the molecular markers that might be helping us to steer patients towards one type of treatment versus another. Even though GU cancers may start in one organ, they may metastasize to the bone or other locations. The fact is that these cancers behave in very different ways. Some patients with metastatic prostate cancer, for example, may live for many years whereas some patients may only live for a few months. We believe that many of those drivers are related to genetic changes, either in the tumor or in the patient’s own immune system. However, we don't have good ways of predicting that.

Can you share more details about your presentation?

There are some biomarkers that are currently being investigated. Two of the most promising are AR-V7, which is a truncated form of the androgen receptor, and BRCA2 mutations, which may determine whether patients are benefitting from either androgen targeted therapies, PARP inhibitors, or platinum chemotherapy. A few years ago, a paper came out suggesting that AR-V7 or a truncated form of an androgen receptor detected in the circulating tumor cells of men with advanced prostate cancer might help predict whether patients would respond to enzaludamide (Xtandi) or abiraterone (Zytiga). Enzaludamide and abiraterone are next generation androgen targeted therapies which are widely in use. What this group from Johns Hopkins Medicine showed was that if you have AR-V7 in your circulating tumor cells, you were very unlikely to respond to enzaludamide or abiraterone.

That was exciting because, if we could negatively select patients who wouldn't respond to these drugs, we could minimize exposure of a very expensive, toxic treatment for patients who would not otherwise respond.

That was published in the New England Journal of Medicine in 2014 but, 3 years later, we still don't have this test in widespread clinical use. This is partly because it has been hard to replicate and hard to commercialize. We expect a commercial test to come out later this year.

Why has it been so hard to replicate an AR-V7 test?

There has been more data published which has been slightly mixed and not as definitive as the original studies of AR-V7. However, I think this could be helpful in certain circumstances and we're going to discuss what type of circumstances we may use this information for. I think one of the reasons why a test like AR-V7 may be hard to replicate is when studies come out of small, single institutions, they may be testing small populations of patients with a very specific technique in the lab that could be difficult to more broadly generalize when you start testing in a larger group of patients. That's why validation of any biomarker test is important.

Are there ongoing studies right now?

Another important thing about commercial tests is that they must be easily accessible, relatively inexpensive, and the average doctor has to be able to get it. I think the fact is that I don't know why we don't all have AR-V7 testing readily available for us but I think those are some of the factors. There was a very large clinical trial of a drug called galaterone made by Tokai Pharmaceuticals. Galeterone was intended to work whether a patient had AR-V7 or AR. This was a big biomarker study where patients were specifically selected to have AR-V7 in their circulation. In fact, that study did proceed and did select these patients, randomly assigning them to either galaterone or enzaludamide based on AR-V7 expression. A press release confirmed negative results for the study, which killed some of the prospects for the idea that AR-v7 could be used to select patients for a specific drug type, in this case galaterone.

However, we do know that if AR-V7 testing became more widely available, which it may later this year, than maybe patients who have high levels of AR-V7 would be better candidates for chemotherapy instead of an androgen receptive targeted treatment. Unless a biomarker is 0% versus 100%, you can't always make your clinical decisions based on the presence or absence of a biomarker.

What would you want the overall main takeaways to be from your talk?

Many of these biomarkers are not perfect and that is why we're still struggling to come up with better data that validates some of the earliest studies. For the average clinician, the main takeaway of my talk will that biomarkers are starting to change the way we treat these patients. I think more research is needed to prove when we should use a specific biomarker to determine a specific therapy.

However, it's too slow for the doctors and the patients, but I think we're making some progress. We may see more use of AR-V7 once a commercial test is widely available and possibly receive additional information about how best to use it.

Perhaps BRCA testing will help us to understand better candidates not just for PARP inhibitors, but also drugs like platinum chemotherapy. The average oncologist needs to be patient but I think we’re going to see more breakthroughs in GU cancers in the next few years.