Further Advances in Breast Cancer Genomics Needed to Optimize Treatment

Partner | Cancer Centers | <b>University of Wisconsin Carbone Cancer Center</b>

Mark Burkard, MD, PhD, discusses the steps researchers are taking to address questions on genomics in breast cancer and what subtypes pose the greatest challenges.

Mark Burkard, MD, PhD

Physicians must continue to explore breast cancer tissue to understand the differences among individual tumors, in an effort to predict disease outcomes and eventually develop treatments targeted at less-common subtypes, according to Mark Burkard, MD, PhD.

Currently, an ongoing clinical trial is actively recruiting to obtain 1100 DNA samples to undergo genomic testing for primary breast cancer (NCT01334021). The observational, prospective study will help determine whether researchers can use genetic testing on tumor samples to predict response to select treatments. The researchers also hope to establish whether certain genes become activated that will then make the tumors sensitive or resistant to chemotherapy or hormonal therapy.

Burkard, an associate professor of medicine at the University of Wisconsin School of Medicine and Public Health, provided insight on the genetic landscape of breast cancer during the 2017 OncLive® State of the Science Summit on Metastatic Breast Cancer.

“One of the biggest and most exciting points is that genomic analyses of cancers, which are available to every oncologist in the United States right now, can profoundly help some patients,” explained Burkard.

OncLive: Please provide an overview of your lecture on genomic testing.

What steps are being taken to answer these questions?

In an interview during the meeting, Burkard discussed the steps researchers are taking to address questions on genomics in breast cancer and what subtypes pose the greatest challenges.Burkard: When I say, “genetics,” I mean all of the genes that make a cancer different than the rest of the person. The incredibly exciting but frustrating thing of what we have discovered in the last decade is that almost every cancer is a little bit different. If you are trying to find another patient like the 1 in front of you to help them and know what’s going on, you have to sometimes look at hundreds of patients with breast cancer—and that’s a problem we need to somehow solve in the future. There are many exciting steps toward that goal. One is just to characterize what the diversity of genomics in human breast cancer is. That has been fairly well addressed through the advances and technology that allowed it. If you keep in mind, 5 years ago, it took $100 million and 10 years to do this for 1 human to look at all of the genes. Now, it can be done for a little over $1000 in 1 week. We now are doing that almost routinely for many different cancers.

We are starting to understand that, but we are also starting to see the complications—one of which is the tremendous diversity between individuals. Another thing I talked about is the recent finding that there is a diversity and changes, or evolution, in the genetics within a person over time.

The research we are doing now [involves] characterizing what is going on, taking patient samples, and understanding the biology. There is a lot of research on what do we do with particular characteristics, and what drugs we have available for these patients that make sense given their genomic characteristics.

When is the best time for a patient to undergo genomic testing?

One thing I am a big proponent of, and is a call to arms, is we have to be able to share information about our patients. We have a large enough cohort of people for which we know what genes they have, what happened to those patients, what medicines work, and what didn’t work. This is so that when “Miss Jones” walks into the clinic 1 day and you have all of this genomic information, you say, “Aha! You have this rare subtype of cancer that only 0.2% of people have.” You can find those 0.2% of people and figure out what happened to them and use that information to help her. That is something we and other people are working on. This is a little controversial right now. In lung cancer, it is becoming clear that it’s right away. In breast cancer, it has been less useful for individual patients, so there is some argument about that. When I would do it in my patients, or at least strongly think about it and discuss it with them, depends on the type of cancer they have. For the hormone-sensitive cancers, the best time is when the cancer becomes refractory to the antiestrogen therapies.

For the triple-negative breast cancers (TNBCs), these are cancers where the standard of care, right now, is just chemotherapy. Those are patients where you might swing for the fences right away and, if you’re lucky and hit a home run, you may come up with a drug that allows that patient to avoid chemotherapy, sometimes for a long time. Honestly, for most people, you’re not going to hit that home run.

Can a patient’s breast tumor evolve and develop resistance mechanisms?

For the HER2-positive cancers, it seems to be less useful right now. What that means is that they are driven by HER2 in such a profound way that they never think of anything else, if you want to anthropomorphize. They are so focused on the HER2 that you can still give HER2-directed therapy and have it be effective—even after the first 1 stops being effective. So, it is less useful in that area right now. Yes. That is 1 pretty profound result that we discussed, which is if you look at a patient’s tumor after the patient had their breast cancer taken care of with surgery and antiestrogen pills, and then showed up again elsewhere in their body, what you find are there are differences. In 25% of those cases, you see a mutation in the estrogen receptor gene itself that causes the tumor to be resistant to the antiestrogen medicine. This completely explains why that tumor was able to grow.

Will researchers continue to do further subtyping as research evolves?

In a small fraction [of patients], you find mutations in other genes. One interesting one, of course, is HER2. This is a mutation, not an amplification; it is not what we would classically call a HER2-amplified breast cancer, but a tumor that is now driven by HER2. The reason why that is important is because there are effective therapies for that small fraction of patients where you can block the HER2 gene. You wouldn’t see it if you looked for the standard HER2 amplification or overexpression test. There are going to be 50 or more different subtypes, but even the subtyping that we use clinically is driven in part by the drugs we have today. Although in research we can find many more subtypes right now—that’s not hard, really—in the clinic, most of those subtypes aren’t particularly useful to define. [We need] more drugs to become available. For example, [we need] for there to be drugs designed for a specific type, such as the luminal androgen receptor TNBC—that is most relevant right now. We have a drug that is effective for those patients, and that [could] be moved into the clinic.

The problem with this, though, boils down to: how do you define your subtypes? There are many different ways to lump or split groups of people. We can argue all day about what is the best way [to do so] in the lab. However, when it comes down to it, in the clinic, it is whether these lumps or splits matter in terms of what treatments we give our patients.

What subtype in breast cancer is currently the most challenging?

What should community oncologists take away from your presentation?

Yes, I think there will be 50 [subtypes] and there may be more than that, but we are not going to get there as fast in the clinic as we will in the research world, simply because it doesn’t always make a difference for our patients. That is TNBC, which, as we know, is kind of a grab bag of what is left when you sort out all the estrogen-driven cancers and all the HER2-driven cancers. That is what’s left behind. It is well recognized that it is sort of a mixture of things. It is not clear how to pull out the patients who benefit from a particular drug.It may feel frustrating as an oncologist if you only look at 2 or 3 patients, because that is not going to help most patients. However, a very small fraction of your patients could be profoundly helped [with genomic testing].

It is worth thinking about and discussing with your patients, especially when we are looking at chemotherapies and using the drugs from the 1970s. Your patients are coming into you and saying, “Aren’t you doing anything new?” Sometimes you can find something that makes an incredible difference in someone's life, so it’s worth thinking about and talking about those things.