Dr. Alicia Morgans is a medical oncologist at Vanderbilt-Ingram Cancer Center in Nashville, TN.
She specializes in treating advanced prostate cancer. She is particularly interested in addressing treatment side effects and in how men with advanced prostate cancer make treatment decisions.
Prostatepedia spoke with her recently about how genomics is—and isn’t—being used in the clinic.
What is genomics versus genetics?
Dr. Morgans: Genetics looks at different genes that can be passed on and inherited through different generations to give you certain traits or different features, like eye or hair color. Genomics for oncology is the study of the genes that alter a cancer’s behavior or growth. It involves our understanding of what is driving a cancer’s growth in a particular person.
Are we currently using genomics for screening prostate cancer?
Dr. Morgans: We use genetics in that we think about men with a first-degree relative with prostate cancer as being at higher risk. Some organizations suggest that men with a first-degree relative should start screening for prostate cancer at an earlier age, though with the United States Preventive Services Task Force (USPSTF) recommendations, things are a little bit murky right now.
We think about people with a family history of breast and ovarian cancer who have BRCA1 and BRAC2 mutations as being an additional group who may think earlier about screening for prostate cancer.
But we’re not necessarily using genomics in the screening setting yet. We use genomics after we have screened someone and taken a biopsy. Genomics can help us predict who may have an aggressive cancer and who may have a less aggressive cancer that can be followed with surveillance instead of treated immediately with surgery or radiation.
How are we using genomics to select initial treatments?
Dr. Morgans: There are some tests that men can use at the time of prostate cancer biopsy or after a prostatectomy that help them and their doctors think about how likely their cancer is to come back, to become metastatic, or to cause them to dieof prostate cancer. Those are tests urologists most commonly use.
Some are used when men get a biopsy and some, like the Decipher test, are used on a prostatectomy specimen. GenomeDX, the company that makes Decipher, just announced a biopsy product to predict the risk of metastatic disease, as well.
There is the Prolaris test which is done on a biopsy specimen. Prolaris uses genomics to risk stratify who is likely to have their cancer come back in an aggressive form and who will have low-risk disease and not necessarily need to have surgery.
How do we use genomics to predict who will or will not respond to certain drugs? For example, looking at AR-V7 mutations to predict resistance to drugs like Xtandi (enzalutamide) and Zytiga (abiraterone)?
Dr. Morgans: AR-V7 testing, which would potentially help us predict who would respond to drugs like Xtandi (enzalutamide) or Zytiga (abiraterone), is not in clinical practice yet. We do not yet have a Clinical Laboratory Improvement Amendments (CLIA)-certified test that can identify AR-V7 mutations in circulating tumor cells or in prostate tissue that will tell us whether or not a patient will respond to Zytiga (abiraterone) or Xtandi (enzalutamide). At some point, a test like that will likely be commercially available.
There are clinical trials that are attempting to use the presence or absence of AR-V7 to either enroll people, risk stratify them, or statistically stratify them in the analysis, but it is not a commercially-available and clinically-used test at this point.
Do you think that is the path genomics will take in the future?
Dr. Morgans: We would love to have a test that could tell us which drugs will work for a particular patient’s cancer and which drugs won’t. We don’t have those tests now, but if and when that information becomes available to clinicians, it will dramatically impact how we treat prostate cancer.
Do you think genomics will change how we design clinical trials?
Dr. Morgans: It already has to some extent. There are clinical trials that focus around certain mutations or alterations. There are studies that include only patients with AR-V7. There are PARP inhibitor studies that use the presence or absence of DNA repair defects to include or exclude patients. These advances in our understanding of genomics are definitely being integrated into clinical trial design.
If those clinical trials demonstrate a benefit, it will be another step forward for genomics. Genomics will then have to be incorporated into our practice over time.
So it’s just a matter of time before we demonstrate genomics’ usefulness and then it will quickly become integrated into patient care?
Dr. Morgans: I don’t know how quickly, but we’re trying. Changing practices in medicine doesn’t always move quickly enough for many of us for important safety reasons in place to protect patients. I would say that genomics has already demonstrated utility. Now we need to finish the clinical trials to really prove that it has a role in clinical practice to help men live longer and better lives. If the studies don’t show this, it won’t be integrated into patient care. I think we are all hopeful that genomics will provide meaningful direction for clinicians as we choose among treatments for men with prostate cancer.
To this point, there is a great New England Journal of Medicine paper by Dr. Joaquin Mateo and colleagues that performed biopsies on men with advanced disease who had had one or two lines of chemotherapy. (See Prostatepedia June 2016 for a conversation with Dr. Mateo about that study.)
The men in the study underwent biopsies and were then treated with Lynparza (olaparib), which is a PARP inhibitor. Those men who had DNA repair defect mutations appeared to have a very high response rate to Lynparza (olaparib) by the criteria included in that study.
I would say that genomics is already demonstrating the possibility of effectiveness. It just needs to be demonstrated on a larger scale. Mateo’s study only included 50 or so people, not hundreds, as we would have in a definitive, practice changing study. But early signs of clinical utility are definitely there.
Couldn’t a medical oncologist take that study by Dr. Mateo and his colleagues and apply the same principles to his or her own patient?
Dr. Morgans: Medical oncologists are doing that, but sometimes we’re limited by insurance. I have patients on Lynparza (olaparib) right now, and we are hopeful that all will go well. We commonly use genomic testing to look at mutations in patients’ prostate cancers, but whether or not an insurance company will pay for us to use an off-label treatment is another question.
It’s easier in a clinical trial setting, because pharmaceutical companies pay for the cost of the drug. It can be more challenging if you’re not in a clinical trial setting.
Are those drugs very expensive?
Dr. Morgans: One of my patients paid out-of-pocket for a week for Lynparza (olaparib), and I believe it was in the thousands.
What about genomic tests? Are they expensive?
Dr. Morgans: That varies by state, by medical institution, and by patient insurance. I have not had trouble getting genomic testing for my patients.
What if a patient has already been treated years ago. Is the biopsy tissue usually preserved or does a patient have to request it?
Dr. Morgans: The patient doesn’t have to request it if a biopsy or surgery was already done. The patient can talk to his doctor or his family could talk to his doctor and say, “We want to have genomic testing.” The doctor just has to request the test and identify the tissue sample that should be used. Sometimes the patient has preserved samples. If that patient has had a prostatectomy, for example, there may be a recently preserved sample that could be sent.
However, I would say the most useful genomic testing is done on a metastatic site at the time that a patient is progressing after first line treatment. The reason is that over time, after being treated with multiple medications, prostate cancers acquire new mutations that allow them to become resistant to different treatments.
If you sample the initial prostate tumor, you may find a set of mutations that don’t include the ones currently driving that cancer and allowing it to get around the treatments used at that particular time. If there is a way for patients to undergo a biopsy of a metastatic site, that tissue would give them a clearer picture of the mutational or genomic profile of his cancer at that time. There can also be some difference, or heterogeneity, between the genomic profile of different metastatic sites. The genomic profile of one metastatic site may not be exactly the same as another metastatic site. There are stronger driver mutations that we think really help that cancer grow and spread and are driving the growth of that cancer. We hope those mutations would be the same across different metastatic sites, but different sites do acquire different mutations. That has been demonstrated in multiple studies at this point.
If you get multiple biopsies over a number of years, should you have all of those tissue samples tested?
Dr. Morgans: You could. But I don’t know how many times an individual insurance company will pay for genomic testing. I have had some patients tested multiple times, usually separated by at least a year to two years because they were on one therapy and then another. These are not inexpensive tests. Sometimes an insurance company may say, “We just won’t pay for that.” But I have not had a problem getting a genomic test at least once for patients.
Is there anything else patients should know about genomics?
Dr. Morgans: Genomics is where we’re going. Being able to predict who will respond to which treatment is our goal. We want to prevent people from being exposed to treatments that aren’t going to help them and stop them from wasting time on therapies that are not going to help them to live longer and better lives.
I would say that, right now, while genomic testing is happening in the clinic, sometimes it takes a patient asking, “Can I have genomic testing on my tissue?” If you are really interested in getting cutting edge prostate cancer treatments, at least understanding which mutations are present in your personal cancer is the place to start.