Dr. Joshi Alumkal is a medical oncologist and Co-Leader of the Prostate Cancer Research Program at the Knight Cancer Institute at Oregon Health & Science University (OHSU) in Portland, Oregon. His lab at OHSU focuses on identifying ways by which prostate cancer evolves into the lethal form of the disease.
Prostatepedia spoke with him about how doctors hope to use genomics to guide treatment.
When and how do we molecularly profile tumors?
Dr. Joshi Alumkal: For patients with newly diagnosed localized prostate cancer, we have fairly standard therapies: radiation, surgery, hormonal therapy with radiation for higher grade cancers, hormonal therapy after surgery in patients who have lymph nodes involved, etc.
Otherwise, there isn’t good data to suggest that the use of molecularly targeted therapy or any other forms of adjuvant therapy are useful to patients.
Molecular profiling is primarily limited to metastatic disease. In that scenario, we’re thinking more about how an abnormality may guide therapy. We think about using investigational agents, which are associated with risks and side effects. These investigational agents may be worthwhile because we think the benefit of treatment outweighs the risks. For someone with localized prostate cancer, we don’t know if finding specific molecular abnormalities will impact their care—particularly because local treatment with surgery or radiation may be sufficient to cure them. If we found a molecular abnormality in a localized prostate cancer, it would be a stretch to recommend an adjuvant molecularly targeted therapy.
Molecular profiling of a tumor certainly offers an opportunity to identify patients for clinical trials, but a lot of those efforts are focused on patients with advanced metastatic prostate cancer.
Would it make sense for men with metastatic disease to get their tumors molecularly profiled?
Dr. Alumkal: It’s at least worth having a discussion with your physician about determining whether or not molecular testing is something that makes sense and if it could be done safely. Ideally, one would biopsy a new metastatic lesion rather than use an old, archived sample. That way you can get as much information as possible about what is going on at that point in time in a man’s tumor.
AT OHSU, we have been involved with a Prostate Cancer Foundation and Stand Up to Cancer-funded Dream Team Award over the past three years. We routinely perform metastatic research biopsies in patients. We have an internal 124-gene mutation panel that we commonly perform in patients.
That includes many gene alterations that are potentially targetable with specific drugs. In our practice, the most common use of molecular testing is in patients who have advanced disease and whose cancers have progressed despite one or two approved therapies—when we’re starting to run out of options.
We want to know if there are other things going on in the tumor that may be targetable with specific medications approved for diseases other than prostate cancer. There has been a lot of excitement recently because a variety of groups have shown that approximately 20% of patients with metastatic prostate cancer resistant to most approved treatments have mutations in gene involved in DNA repair in their cancer.
These mutations involve genes like BRCA1 or BRCA2, which are genes historically associated with hereditary forms of breast and ovarian cancers.
Preliminary data from a small clinical trial in the United Kingdom suggest that a class of medicines called PARP inhibitors—they specifically tested a drug approved for ovarian cancer called Lynparza (olaparib)—can work quite well in patients with mutations in these DNA repair genes. (See Prostatepedia June 2016 for a conversation with Dr. Joaquin Mateo about that study.)
That is one example of how a mutation found in molecular testing could lead us to recommend that certain patients join a clinical trial focused on PARP inhibitor treatment. A variety of companies that have developed PARP inhibitors have initiated, or will very soon initiate, clinical trials in men with advanced prostate cancer whose tumors have these abnormal genes. That is an exciting new avenue of research—looking at tumors and finding abnormalities that seem to correlate well with response to a class of medicines. We have some sense of who may benefit from these drugs. But by no means do we know for certain all the patients who are likely to respond. There is some refinement needed in understanding which genes are really important and which patients may be best suited for that type of therapy.
This is one of the first examples of a molecular assay for prostate cancer that is being used to make treatment decisions for prospective clinical trials. It’s a really exciting time, particularly since the frequency of these alterations is around 20% and quite common.
If you do molecular profiling on one lesion in a man with multiple lesions, does that testing give you information about all of his lesions? Or does each individual lesion have its own molecular profile?
Dr. Alumkal: The verdict is still out on that. There are a variety of groups trying to tackle this question. Most of these studies have been done through what are called through rapid autopsies, in which patients donated their bodies to science to allow researchers to evaluate tumors from multiple sites. There are certainly unique things that you will find in different metastases within a patient. In many cases, there are many features that are the same from metastasis to metastasis within the same patient. There have been a small numbers of studies done in this area. Some work was done by Dr. Steven Bova in Finland and more recently in Dr. Peter Nelson’s laboratory at Fred Hutchinson Cancer Research Center in Seattle, Washington.
Those studies have had some differing results. It is clear is that many of the alterations present within a patient are conserved across all their metastases. It’s not uncommon when we treat patients with therapeutic drugs to see a fairly uniform response or progression in patients. Clinically, that does suggest that many drugs’ behavior may be similar in different cancer lesions within a patient.
I think one way around what we call heterogeneity, or differences between tumor lesions within a patient, is to have tests that sample all of those tumors. That is where a lot of groups are developing bloodbased sequencing or liquid biopsies. Tumors shed DNA into the circulatory system. You may be able to sample DNA and measure DNA from a variety of different tumor sites in a blood sample. This is another set of tests being developed by several companies.
If those assays prove to be more helpful than a single biopsy of a single metastatic lesion, they may be useful in helping guide therapy.
Are liquid biopsies currently available?
Dr. Alumkal: There is at least one major company called Guardant Health (http://www.guardanthealth. com/), which has a blood-based assay for mutational profiling, including many of the genes relevant to prostate cancer involved in DNA repair. It is by no means a comprehensive set of genes, but that is probably the best known and most commonly used blood-based test on the market. There are several other companies developing tests similar to the Guardant assay that will soon be in this space. There has really been an explosion in interest in trying to find non-invasive ways to sample tumors from patients and to provide molecular or genomic information. It’s really hard to get sufficient material from metastatic biopsies to do genomic testing.
At OHSU, we have been fortunate enough to work with an excellent team of interventional radiologists who have honed their technique over the past several years. We’re now at the point where the vast majority of biopsies we’re performing provide sufficient tumor and DNA for informative molecular testing. This is not something easily done at most centers. Having an expert radiology infrastructure is really critical. That is really why, hopefully, bloodbased tests that can be done anywhere will be shown to be as effective, if not more so, than a single-site biopsy.
Are community oncologists and urologists interpreting and using this data effectively?
Dr. Alumkal: There are a variety of different companies that provide mutational testing. As I mentioned, we have our own diagnostic lab here at OHSU. Foundation Medicine (https://www.foundationmedicine.com/) is another company that provides these types of data. It is critical that the output of those tests and those reports be digestible, both by physicians who are at academic centers and by those in community practice. In many cases, patients themselves want these reports for their own records.
We need companies that can provide that information in a digestible fashion and provide recommendations on how those alterations may impact care and clinical trials. We need information that makes those genomic results most informative and actionable.
What are the barriers to genomics becoming more widely used to guide treatment?
Dr. Alumkal: Certainly cost, and whether payers will cover these tests, is critically important. It will be important for the companies that develop these tests to demonstrate that mutational testing can have a significant impact on patient care and decision making. Those are open questions—whether or not the technology will provide better information than clinical information alone. If they can demonstrate impact that could certainly increase the usefulness of this information.
I think we need prospective cohorts of patients to demonstrate what these tests’ characteristics are, what their false positives are, what their false negatives are, and how reliable they are. We want to know that if we were to order these tests, we’re likely to find things that will impact how we care for patients.
Does it make any sense for a man five years after surgery to have tissue banked after biopsy or prostatectomy molecularly profiled?
Dr. Alumkal: If there isn’t a new lesion that can be biopsied, one could consider using an old, archived sample from the same patient. Particularly if the patient has undergone multiple rounds of therapy that could dramatically alter the DNA of their tumor, then it’s probably preferable to get a fresh biopsy and test that new tissue for gene alterations.
If that isn’t possible, then I think there is value in going back and looking at that archived or prior specimen to see if something that was present then could be targetable or actionable. The hope would be that that abnormality is still present and that targeting it now with a specific medicine might make sense.
Those are obviously unknowns, particularly if five years have passed. We may have eradicated the tumor cell that was present to begin with in the original tumor sample and what has grown and emerged is an entirely different cell with a different set of alterations. That is why getting a fresh sample gives a better sense of what may be going on at this moment in time. We’ve certainly had patients interested in getting more information about their tumor, particularly if they wanted to have as much information as possible to guide certain therapies. In those cases, we have gone back to those archive samples and tried to determine if there are abnormalities that were present that suggest a certain medication might make sense.
In those cases, we refer patients for clinical trials, or in some cases prescribe medicines approved for other cancers to treat their disease.
Would it make sense to periodically redo molecular profiling as part of a monitoring plan or is that excessive?
Dr. Alumkal: Anything that doesn’t kill a cancer makes it stronger. We know cancers continually evolve. My hope is that in the future, we’ll get to the point where we can serially monitor patients with non-invasive blood tests to get a better sense of whether or not their treatment is working. If their treatment stops working: why?
How has the cancer changed? To the extent that we can determine that that is safe and useful and can guide decision-making and treatment choices, then I think we’ll see more widespread implementation of those sorts of tests.
It will take linking finding those abnormalities with knowing that if certain abnormalities are present certain medications may target those abnormalities. That will be paradigm-changing.
My hope is that in the very near future, we’ll approach advanced prostate cancer the way oncologists approach localized breast cancer. Oncologists evaluate the estrogen receptor status, progesterone receptor status, and the HER2/NEU status to get a better sense of whether or not certain targets are present to guide whether certain therapies may work.
We’ve seen similar changes in lung cancer. If you have certain alterations in a gene called ALK or in a gene called the EGFR, then specific medications that target those abnormalities make sense for those patients.
We don’t yet have those types of tests in prostate cancer, but I’m hopeful that through a variety of different research and commercial efforts we’ll develop testing that can guide how we treat patients rather than the shot-in-the-dark approach we take now. Today, we just take existing drugs off the shelf and hope they’ll work well in that specific patient sitting in front of us.
Don’t you waste time and money in that approach?
Dr. Alumkal: Yes, because in many cases what we prescribe doesn’t work and it takes time to figure that out. That is lost time for that patient who could have been on a more effective therapy or who could have avoided side effects.
Getting more information about what is going on in the tumor and coupling that with therapies that can target the abnormalities present is a major unmet need in advanced prostate cancer specifically and in oncology in general.
Is there anything else you think prostate cancer patients should know?
Dr. Alumkal: We’ve never known more about prostate cancer than we do now in 2017. That will only improve. As we understand more about what makes each individual patient’s cancer tick and as we find abnormalities that we can do something about, we’ll really be able to make a significant difference in the lives of patients. I don’t believe we’re many years away from that.