Evaluating Risk Scores for Triple-Negative Breast Cancer in Black Women

Podcast Transcript

Dale Shepard, MD, PhD: Cancer Advances, a Cleveland Clinic podcast for medical professionals exploring the latest innovative research and clinical advances in the field of oncology. Thank you for joining us for another episode of Cancer Advances. I'm your host, Dr. Dale Shepard, a medical oncologist here at Cleveland Clinic directing the Taussig Early Cancer Therapeutics program and co-director of the Cleveland Clinic Sarcoma program. Today, I'm very happy to be joined by Dr. Holly Pederson, professor of medicine and part of the medical breast team here at Cleveland Clinic. She's here today to talk to us about polygenic risk score as a predictor for early onset triple negative breast cancer. So welcome.

Holly Pederson, MD: Thank you so much. Thanks for having me.

Dale Shepard, MD, PhD: Absolutely. So we're going to talk about polygenic risk score. We're going to talk about triple negative breast cancer. This is really rooted in some research that you discussed at the recent ASCO conference. Let's start really basic though. What is triple negative breast cancer and why is it such a problem?

Holly Pederson, MD: So the majority of breast cancers, about two-thirds of breast cancers are what we call estrogen receptor positive and HER2-negative. There are three receptors that we check on breast tumors, the estrogen receptor, the progesterone receptor, and a protein HER2. And the different combinations of those three receptor types help to determine therapy response to therapy and give us some idea of the biology involved.

There's a particular type of breast cancer called triple negative breast cancer where the estrogen receptor, the progesterone receptor and the HER2 protein expression are completely negative. They're very aggressive type cells that are not very, what we call well differentiated, and it's a type of cancer that we see twice as commonly in Black women as we do in white women and particularly at early ages.

Dale Shepard, MD, PhD: I guess to elaborate on that when we say... I know in many tumor types. There's kind of a new focus on early onset cancers. When you're talking about early onset, what kind of age range are we talking about?

Holly Pederson, MD: When we were going back and forth with that screening debate in mammography, it's important to realize that about 19% of all breast cancers in women in this country are diagnosed under the age of 50. And in the Black community it's more like 23%. And as I mentioned, the Black community has about twice the rate of triple negative breast cancer as white women do. What we're seeing is that most of the time not linked to family history. We try our best to help and identify young women who are at risk and screen them more carefully like with breast MRI, but we don't identify women who develop breast cancer probably about 60 to 70% of the time. Something else is going on and we think we might be on to a little bit of what that something else might be.

Dale Shepard, MD, PhD: I guess the problem is we have a population who's younger, they don't have traditional risk factors. Is there anything currently that's being done to assess risk and maybe start mammograms earlier?

Holly Pederson, MD: Well, that's a great question. Most women don't become concerned about breast cancer or breast health until their 40s when the general public begins screening. We still recommend annual mammography starting at the age of 40, but there are a number of young women and young Black women in particular that develop very aggressive cancers well before screening would be initiated.

Dale Shepard, MD, PhD: At this point, the problem is we don't have a good idea who those patients are that are going to go on to develop cancer.

Holly Pederson, MD: Right. And you asked about how are they being risk assessed? And in primary care practices and OB-GYN practices, there are so many time constraints and other issues at hand that the primary care providers need to deal with on a given day that oftentimes the family history is not elaborated upon, nor is even the idea of genetic counseling entertained if it's not relevant to the visit at hand.

I think that as we go forward, artificial intelligence will probably be really helpful for primary care providers in identifying patients who at least have family history where it might be a red flag that they need to be watched more closely. But as I mentioned, 50 to 60% of women who develop early onset breast cancer don't have any family history.

Dale Shepard, MD, PhD: So, we're going to talk a little bit about risk assessment using this polygenic risk score. Maybe let's start with what exactly is a polygenic risk score?

Holly Pederson, MD: That's a great question. So breast cancer has really been one of the leaders in the explanation of how diseases can develop. And we are all going to be identified with gene changes or mutations that predispose us to developing all sorts of different diseases. But breast cancer has sort of been head of the pack in terms of identifying those genes. And in fact, in 1994 and 1995, BRCA1 and BRCA2 were identified. But it really wasn't until 2013 when Angelina Jolie so bravely came forward with her story that most people became more aware of these genetic syndromes and she had a genetic mutation in the BRCA1 gene.

There are about six or seven genes that are what we call highly penetrant, meaning you're more likely to get cancer than not in your lifetime. Those are the situations where we start screening at 25, where we have discussions around risk-reducing surgery, where we discuss assisted reproduction. There are genes that are really, really risky in terms of breast cancer development. Then there are moderate risk genes like CHEK2 or ATM, which are definitely linked to the development of breast cancer but are not felt to be as causal.

They act in concert with the patient's environmental exposures, food and interaction with other genes. And when I say other genes, there's a third level of genetic risk evaluation in terms of breast cancer. And these are common genetic variations called single nucleotide polymorphisms or SNPs. And there's over 300 SNPs that have been identified as being linked to the development of breast cancer, but they confer very small levels of risk individually.

If you add the effects of the different risk alleles in a weighted way such that certain of them have more significance than others. That's what a polygenic risk score is, is sort of a weighted average of the effects of these SNPs or common genetic variants. We all inherit two copies of every gene, one from our mother's side and one from our father's side. So you can see if there's 300 of these SNPs that are predisposed to breast cancer, each child could have a completely different composition of alleles, which are the places at the gene where there may be these gene changes or mutations.

And the alleles are inherited independently. So different siblings, even in a single family can have very different polygenic risk scores. We found that polygenic risk scores are useful not only in predicting breast cancer, but other types of cancer and other disease types. It's really being looked at in every disease type, diabetes, things like that. It's extremely interesting as it may go on to explain a lot that we have not been able to explain to people in terms of why they developed something.

We found that the polygenic risk score in breast cancer can not only substratify risk in women who have family history, but in patients who carry the highly penetrant or moderately penetrant gene mutations. The polygenic risk score can also affect the expression of those genes. And in newly diagnosed breast cancer patients has been shown in three different studies to predict contralateral breast cancer risk or development of a breast cancer on the other side.

Ultimately, hopefully we'll be able to identify low risk women who may not need to be as vigilant. At the present time, we can't identify low risk women, so we advise all women to start screening at the age of 40. So you've got patients with genetic mutations, patients with family history, but no genetic mutation identified, newly diagnosed cancer patients and everyone else.

So at some point we believe that this will be part of standard risk assessment, and our dream is that every young woman by the age of 30 should have a comprehensive risk assessment that's mainly family history based, but perhaps in the future, a polygenic risk score to see if a woman is at increased risk.

Dale Shepard, MD, PhD: I guess historically, has the polygenic risk score done a very good job specifically with triple negative breast cancer?

Holly Pederson, MD: Well, that's a great question. So most of the polygenic risk scores in this space were developed and validated in European women. And the test for that reason is most accurate in Caucasian European women. In fact, if you use the test that was developed and validated in that population, in other ancestral groups, there is a market overestimation of risk. And so some of our prior work involved calibrating the polygenic risk score to account for differences in ancestry.

It's felt that there are probably four like source ancestries, European, African, East Asian, and Amerindigenous. These are just four different areas where you can be part this, part that, part this. And each ancestral component has different weighting of the SNPs. And so you need to determine a woman's genetic ancestry before you form a polygenic risk score that's accurate for women of all ancestries. I hope that that's clear. It's kind of a weird concept.

Dale Shepard, MD, PhD: But that's sort of the origin of this multiple ancestry polygenic risk score. And then I guess as part of the data you presented, you use this modified polygenic risk score to assess risk for women with triple negative breast cancer. Is that correct?

Holly Pederson, MD: Right. So what we've found over time is that the discriminatory accuracy or the ability to separate high risk from low risk women is less impressive in women of African descent because the SNP and their risk association are not as well established. If you think about it, Africa is just incredibly genomically diverse and the cradle of humanity essentially. And there were really very small numbers of people that went to different parts of the globe to populate those areas, leaving the areas that were populated very homogeneous.

Whereas Africa is very difficult to characterize genomically and the polygenic risk scores have been validated back to 2015, but are still not available in clinical care because although they are useful in American Black women, they're not as useful as they are for American white women. So what we decided to do, in addition to continuing to gather that data, to have that number be essentially equivalent, we decided to look at it the other way around.

We looked at young Black women who presented for genetic counseling over a one-year period and there were over 17,000 young Black women who presented genetic counseling and testing through myriad genetics. We always need to partner with these large labs because they have such incredible data sets like this. Amongst those 17,500 patients, over 3,000 had a personal history of breast cancer, and the remainder had enough family history of breast cancer that they were eligible for genetic testing. That may not have been only breast cancer, but it could have been other types of cancer as well that made them eligible for testing.

What we wanted to see was is there a difference between the young women who have developed breast cancer as compared to the young Black women who were at increased risk due to family history but have not developed breast cancer. And all of these patients, the ones who had cancer and didn't have cancer, had negative genetic testing in the classical sense of the word testing for those highly penetrant and moderately penetrant genetic mutations. And so their cancers were really unexplained.

What we found is that in looking at the polygenic risk score, it really did help to predict which young Black women were going to develop breast cancer period and which might develop triple negative breast cancer, which is biologically one of the most aggressive types.

Dale Shepard, MD, PhD: If we come up with a test like this sort of multiple ancestry polygenic risk score, for instance, and we can identify high low risk like put people in risk categories, what is the most effective thing to do clinically with that? Is it starting to screen at a younger age? Is it the test we use to screen with? Is it the frequency? How do we use that to then guide clinical care?

Holly Pederson, MD: So what would we potentially do with this information? I think that the most important thing is early and more sensitive screening. It's been shown in our BRCA population that women in their 30s, the sensitivity of their screening mammograms, meaning that the chance that you will find cancer if it's there is about 30% meaning we're going to miss 70% of breast cancers in women's 30s on a mammogram due to the breast density typically. And breast MRI is a highly sensitive test that could be offered to young women annually to catch things early.

I mean, it makes all the difference in the world to identify cancers at their earliest, most treatable and curable stage. And remember, these are young women who develop cancer before screening would even be undertaken, and often they aren't taken seriously by their primary care provider when they bring up a concern because they're so young. And this potentially could identify people who could undergo enhanced surveillance or screening with breast MRI.

There's a medication tamoxifen in some patients that may be useful to reduce risk and with a significant family history, one might even consider risk reducing surgery not on the basis of this test, but on the basis of their other risk factors, including their family history. It also would give us an opportunity with these young women to stress lifestyle modification. About 30% of cancers in this country are caused by the obesity epidemic, and that hits the Black community even harder than the Caucasian community. And it really is an important risk factor.

Alcohol consumption is increasingly being identified as an important risk factor as well. But I worry most about the weight and I think if we could create a standardized risk assessment at the age of 30, particularly for Black women and those of Ashkenazi descent, we may be able to identify a lot more people who could be watched more closely at minimum.

Dale Shepard, MD, PhD: And so, since risk of breast cancer is both family history and germline changes due to exposures, 30 is probably a good spot in terms of assessing that risk?

Holly Pederson, MD: Between 25 and 30. As I said, with BRCA and some of the other highly penetrating mutations, we would start screening at 25. No earlier than 25, but between 25 and 30, it would just be wonderful if we could identify young women even just on the basis of family history who could use additional help.

Dale Shepard, MD, PhD: What are the next steps? Where are you going from here?

Holly Pederson, MD: So, we will continue to validate this multi-ancestry, polygenic risk score in other data sets. And others, of course, will continue to validate different polygenic risk scores, which largely include the same SNPs. There's not a large difference between, say, the 313 SNP polygenic risk score that's mainly used in Europe as compared to the one that we're using. But the difference with the one that we're using is really that it's calibrated for ancestry and it's very accurate for women of different ancestries, including those in whom we may not ever have enough data to have as accurate of numbers as we'd like to have.

Dale Shepard, MD, PhD: Well, that's great. So a very serious disease that, we, up until now don't have a great way to identify risk. And it looks like you're doing some great work to try to solve that and appreciate your insight.

Holly Pederson, MD: I mean, that's what we're trying to catch women before they develop breast cancer. That would be our dream is to try to identify those so that we can make interventions before somebody gets sick. I think it could be a paradigm changer. There's nothing worse than a young woman who comes in with breast cancer well before screening would begin, who has no family history and no reason to have developed breast cancer. And so this year at our breast cancer summit, we're focusing on the care of young women.

Our predecessors, Dr. Joseph Pro who started the breast center was one of the pioneers in nipple sparing mastectomy, and really did a lot to help young women both treat and prevent breast cancer. So it's been a great space to work in. And the polygenics combined with breast density, I think, and patterning on mammograms are probably going to be areas that are going to be more helpful going forward.

Dale Shepard, MD, PhD: That's fantastic. Well, thank you so much for being with us today.

Holly Pederson, MD: Well, thank you so much for having me. I appreciate it.

Dale Shepard, MD, PhD: To make a direct online referral to our Cancer Institute, complete our online cancer patient referral form by visiting clevelandclinic.org/cancerpatientreferrals. You will receive confirmation once the appointment is scheduled.

This concludes this episode of Cancer Advances. For more podcast episodes, visit our website, clevelandclinic.org/canceradvancespodcast. Subscribe on Apple Podcasts, Spotify, or wherever you listen to podcasts.

Thank you for listening. Please join us again soon.