Breast Cancer Biomarkers

This podcast episode is made possible, in part, by a grant from Lilly.

Jamie DePolo: Hello. Thanks for listening. Our guest for this episode is Dr. Debu Tripathy, professor and chair of Breast Medical Oncology at the University of Texas MD Anderson Cancer Center in Houston. A noted researcher, Dr. Tripathy joins us to discuss breast cancer biomarkers, the molecules that help doctors distinguish breast cancer cells from healthy cells. Biomarkers help you and your doctor make decisions about the treatments that are best for your unique situation.

Dr. Tripathy, welcome to the podcast.

Dr. Debu Tripathy: Thank you for having me today.

Jamie DePolo: We're very excited for you to explain this very complicated topic to us. I gave a brief description of what biomarkers are in the introduction, but I'm wondering, how do you talk about biomarkers when you're talking to your patients?

Dr. Debu Tripathy: Well, I tell them that the field of cancer, in general, has gone a long way. Breast cancer and many others are now driven by...in terms of treatment decisions, are driven by what we term biomarkers, and these are characteristics of the tumor that alter their behavior or govern their behavior, particularly in the context of being able to choose a treatment. We now know of many, although not all, of what we call drivers. These are changes that occur to cells that make them go from normal to malignant.

Obviously, we know -- we've known for many years -- that one of the common abnormalities is that they grow more rapidly, and they don't know when to stop growing, but now we know what actually governs that and what drives that, to a certain extent. We obviously have much more to learn, but we've started to apply that knowledge in different ways, particularly not only to prognosticate, but more importantly, to choose the therapy.

In many cases, we've known about the targets for many years. Like the estrogen receptor was actually first postulated over 100 years ago when a surgeon named George Beatson, a Scottish surgeon, made the connection between the breast and the ovaries, only because of menstrual periods and the pattern that women had and also pregnancy, and postulated that maybe that it was some factor that the ovaries were making. He had no idea what it was, but he did oophorectomies on patients and found that, for some of them, it really helped.

And then, of course, it was many decades later that we discovered estrogen and other hormones and then the estrogen receptor being on cancer cells in the '60s and '70s, and we've been using that as a biomarker since then. So, that's one of the older examples, but for the most part, they are either proteins -- and more recently -- genetic alterations, changes in our genome, different what we call mutations, is when the base pairs are altered, and that actually governs how cells grow and how fast they grow.

And a lot of it is linked to biology and new discoveries, but these are now standard assays that we do, and they vary among different cancers. For breast cancer specifically, there's a long list of them, and we'll review them later today, but at a very basic level, they are functional proteins or genes, and what I mean by that is they alter the way a cell behaves, and many of these are necessary to test for in order to choose the right treatment.

Jamie DePolo: Okay. Thank you. That makes sense, and I know sometimes there's a little confusion about biomarkers, because we do seem to have a lot of names for them, kind of depending on what the biomarker is.

As you’ve said, it could be a protein. It could be a gene mutation. So, I know sometimes biomarker testing is called tumor marker testing. It could be called genomic profiling. It could be called genomic assays, genomic testing, but really, if I'm understanding this correctly, they're all biomarkers that you, as an oncologist, use to figure out what's the best treatment and how is this cancer going to behave?

Dr. Debu Tripathy: Yes, that's right. There are many ways to classify biomarkers, and they are different types. One of the more common ways we classify them is if they are the DNA. That is the central template of information that is...that governs life in general. Everything is coded off DNA, which is made up of four base pairs, and the sequence of the DNA tells the body how to build the building blocks of proteins, how to assemble the amino acids.

So, when we're looking at the DNA, we can pick up mutations that may be important in cancer, and more recently, a lot of our biomarkers are actually a sequence of DNA, either mutations of DNA where it changes the behavior of the protein that is encoded by the DNA, or sometimes the DNA is lost. We call those deletions, and sometimes the DNA is amplified. We call those amplifications, and it leads to making more of the protein.

Now, sometimes we test the proteins itself, and we can use special stains that look at the proteins on the surface of cells, or sometimes they're inside the cells, and these stains are antibodies that either have a color on them or some way for the microbiologist or the pathologist to actually be able to discern. DNA requires that we get some material from the tumor to sequence.

But now, we can actually find small amounts of tumor DNA just circulating in the blood, and we call those liquid biopsies. And then, finally, there's RNA. DNA is encoded into RNA, and then RNA is encoded into proteins. RNA is very unstable, and it's hard to test, but we can sometimes run RNA tests on the tumor, as well, and there are some cases in which that is the best way to actually pick up a genetic abnormality, is through the RNA.

It gives you a better look at things that you may not pick up from sequencing the DNA. I won't go into the biology as to how DNA is processed, but suffice it to say that between DNA, RNA, and protein, that generally constitutes what are most of the biomarkers that we test, and they can be tested, generally, directly on the tumor itself, or sometimes you can measure it in the blood.

Jamie DePolo: Okay. So, let's talk a little bit about breast cancer biomarkers specifically. I would group them, please tell me if this is correct, into sort of three broad categories. We have these biomarkers that pretty much every breast cancer is going to be tested for, and then we have a group...now, we may test for these, depending on what, like, the stage of the cancer or size of the cancer, other characteristics like that. And then I feel like there are some biomarkers that are more done in studies.

Like we're not really sure how to use them yet, so there are a lot of studies going on. So, if you agree with that, I'd like to discuss them in that order. So, we'll start with the biomarkers that every breast cancer is tested for, and to me, that means the estrogen receptor status, the progesterone receptor status, and HER2 status, and I think those all fall into what you were talking as proteins, if I'm right? So, could you talk a little bit about those, and why is it that those are always tested for?

Dr. Debu Tripathy: Yes, that's absolutely right. There are some tests that help us with both early-stage and advanced-stage, and they're very basic, and we've been doing them for many years. The basic ones that you mentioned, the estrogen and progesterone receptors, and HER2, which is also a receptor, and then there's another one called the proliferative antigen called Ki-67, which is a protein that is seen when a cell is dividing.

So, it gives us a sense as to how aggressive a cell might be, and these are tested in every case of cancer, as you mentioned, including early-stage. The estrogen receptor tells us whether it is more likely than not that that cancer actually grows under the drive of estrogen, that estrogen actually promotes its growth, and that drugs that block estrogen binding to the estrogen receptor makes the cell grow, and that tells us that hormonal therapy is going to be an important treatment.

And in patients that have early-stage breast cancer and advanced-stage breast cancer, that's an essential protein that we always check for, and I'll also add that sometimes that can change. When patients develop a recurrence of their cancer, it sometimes...all of the receptors, and even the mutations, may be different in the cells that metastasize, and the reason that that happens is because of what we call evolution. Just like evolution of the species, cancers can evolve in the sense that one of a million cancer cells may develop a new mutation.

For example, it may be a mutation that makes it lose its estrogen receptor, and that particular small number of cells or single cell spreads and starts to grow, and let's say the patient's already on hormonal therapy. The hormone receptor-positive cancer cells are going to be checked in their growth, and the only ones that are going to grow are the estrogen receptor subset, but eventually, they take over the tumor. So, you do see shifts in biomarkers. So, when patients either recur or progress, we may repeat those biomarkers.

So, many times, we're doing them several times. The estrogen and progesterone receptors together tell us that hormonal therapy's going to work. Estrogen receptor's probably the more important one, and HER2, which stands for the human epithelial receptor, is a protein that is expressed in about 20%, maybe 25% of all cases, and that growth factor was discovered later in the mid '80s, and antibodies against that receptor were found to slow down the growth of those cells.

And we now use that as...to tell us whether the patients may benefit from anti-HER2 therapies, like trastuzumab, also known as Herceptin, and we use that both in early-stage and advanced-stage. So, that's why those are the basic biomarkers that we're always going to get, and sometimes, like I said, if we're doing a second biopsy, we may repeat it. Let's say a patient has a recurrence and that recurrence is many years later, we are almost always going to test it, or even if someone has their cancer well controlled, but one part of the tumor is growing.

Let's say someone has advanced breast cancer and one of the liver lesions is growing much more rapidly, we may actually biopsy that and see if that's different from the last set of results we got. So, those are, in fact, the basic ones that we are doing all the time, and the last one I mentioned, the proliferative index is something we're not quite sure how to use that test yet, but the more rapid cancer is growing, it'll have a higher percentage of cells that make this Ki-67 protein.

And there are some drugs that we use for early-stage cancer that actually depend on how high that Ki-67 number is. So, it has also become one of the basic four proteins that we test, primarily in early stage. We don't use Ki-67 so much in advanced stage, but certainly, estrogen, progesterone, and HER2 receptor retest on recurrence and progression.

Jamie DePolo: Okay. I do have one question about the Ki-67, the proliferative protein. If somebody was diagnosed with early-stage disease and that level was high, would that make it more likely that chemotherapy might be recommended because the cancer would be considered more aggressive?

Dr. Debu Tripathy: Yes. We are starting to use it in very defined ways. One thing about Ki-67 is there's a lot of variability in how different pathologists will read it, because not only does it have to be stained in a consistent way, but the judgment of the pathologist in terms of how deep the intensity is makes it a very subjective test. Nowadays, many of the immunohistochemical testing...so, what immunohistochemical testing means is we're using an antibody.

Usually, it's an antibody that is tagged with a color so that the pathologist can easily identify it, and these tests are now being read by an automated reader, a digitized reader, and that makes the reading a little more consistent. So, I would say that, over the last few years, there's a little more uniformity in reading Ki-67, and we are using it, but we use the Ki-67 in a very defined way, because there's still some concern about the variability in how it's read.

When...we don't use it just on its own, even though higher numbers generally indicate higher proliferation indexes, and the reason is that we have some better tests. One of the biomarkers that we're now using more often in early-stage breast cancer, in addition to, ER, PR, and HER2, is what's called gene profiling, and this is an RNA-based test, because RNA correlates very well with the amount of protein there is, but it's easier to measure, especially when you're measuring multiple different proteins, and over the years, we've started to use gene profiling to tell us which cancers are more likely to recur.

This is for early-stage cancers that are generally treated with surgery and maybe medical therapy, and most patients are cured, but some patients will develop a recurrence, and looking at multiple genes at the RNA level...remember, DNA is transcribed into RNA and RNA into proteins. So, at the RNA level, that intermediate level, can let us look at many genes, and we can recognize patterns of these gene expressions that have been, over the years, validated to show higher risk of recurrence or a better response to chemotherapy.

Many patients don't need chemotherapy for early-stage breast cancer. In the years past, we probably were overtreating patients, but when we started to look at large collections of patient cases and the tumors that we had then, we were able to start to go back with this new technology looking at RNA and be able to discern which ones had a higher risk of recurrence, and now, these are the commercially-available tests that we tend to run particularly in hormone receptor-positive and HER2-negative cancers, which is about two-thirds of all cancers.

We now use the gene profiling assay to identify those that are at higher risk. So, Ki-67 has fallen a little bit by the wayside when it comes to that particular measure, but it is still used in cases where we're now looking at adjuvant therapy with some of the newer drugs that've entered into the field. We used to use, primarily, hormonal therapy and chemotherapy, but now we're using some of the biological drugs, like cyclin-dependent kinase inhibitors, that were only used in advanced-stage breast cancer.

But in the last few years, have been tried in trials that are showing lower risks of recurrence when you take these drugs for two to three years, along with hormonal therapy, and one of those trials actually used Ki-67 to determine higher-risk patients, and so, that is a test that we use in that case. There are some other newer areas that are not quite ready for us to use, although, in some parts of the world, they are using it.

And that is, interestingly enough, when someone gets hormonal therapy for a brief period of time and their tumor is biopsied after the hormonal therapy, the Ki-67 drops if that tumor is very sensitive to hormonal therapy, to the point where they may not need chemotherapy. So, I won't say this is the standard of care yet, but it is an area where a drop in the Ki-67 with hormonal therapy may tell us that someone we otherwise might've used chemotherapy with may not need it.

But today, really, the most important test we use is just the anatomic stage. How big the tumor is and whether the lymph nodes are involved, but in some cases, like node-negative breast cancer, or when you have one to three lymph nodes, we're using the gene profiling assay called Oncotype. There's another one called MammaPrint. There's another one called the Breast Cancer Index.

There are several of these that've all been shown to be helpful in not only prognosticating, but discern who may not need chemotherapy. So, I would say that, in addition to the major biomarkers, we mentioned estrogen, progesterone, and HER2 receptor, and to some extent, Ki-67. The gene profiling now is something we're using in early-stage breast cancer. We're not using it in advanced breast cancer, and that's an RNA assay.

Jamie DePolo: Okay. Thank you. You must've been reading my mind, because that was going to be my next question, that these genomic assays, genomic tests, like Oncotype, like MammaPrint, those are done...they're common, but they're not done on every single tumor, because my understanding, in most cases, not all, I think there are six total genomic assays if you count the DCIS, the Oncotype DCIS test. They're done on, as you said, hormone receptor-positive, HER2-negative disease. I think there's one that can be done whether it's hormone receptor-positive or hormone receptor-negative.

So, yeah, so, I lumped those into...those are tests that might be done. They are more common than some of the other ones, but it really does need to be done on that early-stage, hormone receptor-positive, HER2-negative disease. So, thank you for explaining that.

I'd like to move onto some of the other biomarkers that might be tested for, and I'm going to list them and ask you about them. I hope that's okay, because there's quite a few. So, the first one I think that is probably...people have heard about more because of Keytruda, would be PD-L1 status. So, if you could talk a little bit about that?

Dr. Debu Tripathy: That's right. This is one that we've been using relatively recently. PD-L1 is a protein that is known as a checkpoint. Over the years, we've learned that the immune system actually can fight cancer. In fact, the immune system probably protects us from cancers that we may not even know, and there may be one or two cells in our body that the mutation was enough to actually start the very beginnings of cancer.

But the body and the immune system can recognize it as a foreign protein, just like as though it was a bacterium or a virus, but the immune system has a much harder time picking up cancer cells, because most of the proteins are the same as our own bodies' proteins. A few of them are different. They're subtly different, and then the immune system may kick in, or it may get overwhelmed, and people that are immunosuppressed have a higher frequency of certain types of cancers.

The field of immunology and breast cancer has been around for a very long time, but we never had the tools to really understand it well. We've known for decades that sometimes we see a lot of lymphocytes around the tumor, almost like the immune system is trying to do something, but not able to, and sure enough, that is the case. As we've developed more sophisticated tools to look at proteins in genes, we've found that some cancers survived because they adapt to the immune attack.

What they do is they use parts of the immune system that have always been there to protect us from over exuberant immune reactions. When we get an infection, the immune system kicks into high gear, and it causes a lot of inflammation. It actually causes some tissue damage and irritation to wipe out the infection, and if that were to continue, it might harm the host. We all know that when we get a virus, you get the fevers and chills and that. You feel crummy, and that's when the virus is replicating, and you have that kind of illness.

But then the longer phase of it is the recovery phase when you have the runny nose and the cough, and by then, the virus has been cleared, and what you're dealing with, really, are the after-effects of that exuberant immune system and all the, what are called, cytokines. These are chemicals that can kill bacteria and viruses, and they do cause damage, and in order for that to wane away quickly, the body has what are called checkpoints, and they make the immune system recede. Well, you know, cancer is survival of the fittest, right?

So, if a cancer cell happens to have a particular mutation that allows it to activate these so-called checkpoints, it's going to give it an advantage in growing, and so, we find that many tumors actually have these checkpoints upregulated, and it protects them from immune destruction. Well, now we have tools to now attack those checkpoints. PD-L1 is one of them. PD-1 and 2 is another one, and there's many others that we're still investigating, but the ones that attack that particular checkpoint system, PD-1, PD-2, PD-L1 system, are the ones that were first now tested and now available for many types of cancers.

So, PD-L1 is the ligand that binds the PD receptor, and when tumor cells express that and the immune cells express that, that makes it more likely that that tumor will respond to immunotherapy. So, there are some situations in which we need to check for that checkpoint before we prescribe immunotherapy. In breast cancer, the subtype that responds best to immunotherapy is the triple-negative subtype, which is about 20% of cancers.

So, for triple-negative breast cancers that are advanced, we check for PD-L1 to see if there's a chance that patients might respond. When you're PD-L1-negative, the chances of responding are very low, although some people do respond, and when...even when it's PD-L1-positive, not everyone responds, but those are the only groups that we treat, because immunotherapy has sides. We wouldn't want to use it indiscriminately.

Now, interestingly, immunotherapy was first approved in metastatic triple-negative breast cancer, advanced triple-negative breast cancer, in around 2019, and it required PD-L1 testing. We knew that the...that it made a difference, but in early-stage breast cancer, we did trials to add immunotherapy to standard chemotherapy, and found that patients did better.

And in 2021, it was actually approved for early-stage breast cancer, but interestingly, it was equally effective, regardless of the PD-L1 status. So, for early-stage, triple-negative breast cancer, if it's stage II or higher, we use chemotherapy along with immunotherapy, and we actually use it before surgery and then do surgery afterwards, and we don't check for PD-L1 in that situation, but for advanced breast cancer that's triple-negative, we generally do, especially for first-line therapy. We tend to use it as the initial treatment.

Jamie DePolo: Okay. Thank you. Now I want to talk about BRCA1 and BRCA2 mutation status. I know there are medicines called PARP inhibitors. So, if somebody has advanced-stage disease, you test for that. So, could you talk a little bit about those?

Dr. Debu Tripathy: Yeah. So, this is a whole category of mutations that we call germline mutations, and these are inherited, and this is a very important distinction, because many of the tumor genetic testing we do on the tumor itself is only on the tumor. But BRCA1 and 2 are germline mutations that confer a familial risk. So, when we see patients who have developed breast cancer at a young age or those that have a family history of breast or ovarian cancer, which also tracks with some of the mutations that we see, we recommend what's called germline genetic testing, and that can be done with any cell.

You don't need tumor cells. In fact, it's better to have normal cells. That way, you won't be confusing it with mutations that may only be in the tumor, and you can either test the blood or a cheek swab, just get some cells from the mouth. Those are common ways to do that type of testing. And the first susceptibility genes to be identified was BRCA1 and then BRCA2, which gives you a very high risk of breast and ovarian cancer, but now, we've identified many other genes. You know, there's probably 600 or 700 tumor-susceptibility genes.

Some are very rare and infrequent, but now that we can do panel testing and sequence a whole genome, we order a lot of these multi-panel genes to test these in patients who have a family history or are very young or have multiple cancers. But because the tests have become more precise...and we used to get a lot of these variants of unknown significance. So, alterations that we didn't know if they were a disease-causing mutation or just a benign variant, but now that we have followed many patients over many years, these numbers of VUS, variants of unknown significance, are much less.

We feel more confident in testing patients, even if they have a lower chance of having one of these mutations. So, we always test patients that have a strong family history or are at a young age, but we're also testing patients with more advanced breast cancer, because when these genes were first identified, it was mostly to know who in the family had the cancer and then to initiate surveillance, or even, in some cases, preventive surgery, like mastectomy or oophorectomy, once the patient got to a certain age.

But as we've learned more about the biology of some of these germline mutations, we're also learning that the mutation leads to other cellular abnormalities that can be targeted with certain drugs, and BRCA1 and 2, for example, are involved in DNA repair, which sort of makes sense. If you don't repair your DNA well, you're more susceptible to get mutations, including those that can be cancer-causing, and because there are several ways we can repair our DNA, it was discovered that people who have BRCA1 and 2 mutations their tumor cells can't repair DNA.

So, if you knock out another type of DNA repair, now the cell is left with no way to repair its DNA, and it's what's called a synthetic lethal defect. In other words, the fact that the BRCA mutation already exists in those cancers and now you're knocking down another DNA repair pathway, it's an effective drug, and those cancer cells can be killed or damaged, to a point where these PARP inhibitors now are like other drugs we use.

They can cause...bring about remissions and control cancers, and so, we now want to test patients, in addition to patients with strong family history. Anyone with advanced breast cancer now, we check just to make sure, because it turns out that a strong family history or cancer at a young age doesn't pick up all of the people that have BRCA1 and 2 mutations, and now there's a whole bunch of other mutations, PALB2, CHEK2, there's a whole list of them that will be in a test score, but so far, PARP inhibitors are only approved for BRCA1 and 2.

There's some evidence that they work for some of these other mutations, and eventually, they may be approved for them, as well, but there's a new generation of drugs now called DNA damage repair inhibitors that are being tested for people that have a variety of these mutations, and so, you know, the field marches along. It probably won't be long before we have these drugs available for these and other tumors, and that's why this field is so important, because I know that we're going to get into this later, but I'll just bring it up now.

That, in addition to testing for clinical need, where we have a proven reason how we're going to use that information, we're building it into many of our clinical trials, and we are starting to learn now, when a new drug is being studied and hopefully will become approved, the FDA now is expecting the maker of the drug to not only show that the drug is effective, but to also define which biomarkers predict who's going to respond and who isn't.

And we've been...as scientists in the field, we've been urging drug companies to invest the money in doing biomarker research, and they were very reluctant to do that, because it raised their costs, and they didn't see the immediate benefit, but after years of showing these benefits, they've flipped, and now they want to do it, and they're running it, and they're investing in it, and of course, science is the star and we're collaborating with worldwide consortiums, especially for rare tumor or some types of breast cancer, to really understand which genes are driving cancer, not only for diagnostics, but eventually, knowing the genetic makeup and how the proteins are effective allows us to develop better drugs against people that carry specific mutations.

And when it used to be that we only tested breast cancer patients for the mutations we knew of that were important in breast cancer. We are now testing for mutations that are rarely seen in breast cancer, but are more seen in melanoma or renal cell cancer or some other cancer, because some of the drugs are now being approved in what we call a tumor-agnostic fashion. That means that if you have that mutation in your tumor, no matter what cancer you have, you might benefit from that drug. So, not only are we now interested in testing germline mutations, as I was mentioning, but also tumor mutations in patients in a much broader sense.

Jamie DePolo: Yes. That's very exciting, and I think this next one falls into that, that PIK3CA mutation. I think that's a tumor mutation, right, not a germline mutation?

Dr. Debu Tripathy: That's correct. That's an acquired mutation, and thank you for making that distinction. Most of the mutations we pick up in cancers are acquired mutations, and that means they're not in the germline. They're not in the DNA that you carry in your body and pass onto your children. A few of them are, like BRCA1 and 2, but most of them are acquired, and the reason that that happens is that, as I mentioned earlier, cancer is a selection of the fittest.

And just like bacteria and viruses, if they develop a way to become resistant, they start to then, you know, overpopulate and survive, and they can become difficult to deal with and difficult to treat, and most of our cells are very tightly regulated. If you look at...in an adult body, the number of cells that are dividing is well under 1%. Most of the cells are just sitting there, doing what they're supposed to do. They're either an enzyme or they're structural. They're holding the body together, but they're not dividing.

You know the...obviously, the bone marrow cells are dividing. Some of the skin cells are, but they're very tightly regulated. We have a lot of checks and balances in our body that tells cells to stay still, don't grow, and then, when it's the right time to grow, certain signals go on. Well, those can be co-opted if there's a mutation, and if a... Mutations happen all the time. They're random events, and especially as we get older, more mutations happen.

There are certain things that can prompt more mutations, like tobacco smoking and sunlight, and so, those are carcinogenic exposures, because they increase the rate of mutation, but if one in a million mutations actually makes a cell grow faster...most mutations lead to death of a cell and you never hear from it again. But every now and then, it happens to be just in the right place, like a growth factor, and it makes the cell grow faster.

Well, guess what? Those cells now have a selective advantage over all the other ones, and they're going to be selected for it. So, it's almost.. you would think everybody will get cancer over some time if you have billions of cells in your body multiplying, and the fact is we do, but our immune system...not only our immune system. There's other systems in our body that can sense these mutations and auto-disrupt the cell. We actually have suicide pathways built in the cells.

It's called apoptosis, and when you have a mutation, that happens, but every now and then, all the defense mechanisms, you know, by random chance, might be overrun. Now, sometimes it's not just random chance. Some people have a variety of inherited abnormalities in DNA repair genes, like BRCA1 that I mentioned earlier, but others, as well, but for the most part, our body's pretty good at picking them up, and when cancer develops, it's because there was a breakdown either in the immune system or the DNA repair, and sometimes it's just chance.

The most common cause of cancer is just random chance. Now, obviously, these are certain predisposing factors, either environmental or genetic. Because of that, we can catalogue now large numbers of mutations just by observing this, and one of the benefits of the Human Genome Project, which had been a goal from the '80s onward...and it took forever, but as soon as next-gen sequencing...which was an amazing invention in the early 2000s, it allowed us to sequence the genome much more quickly.

And now we have a template of what the gene should look like in normal patients, and the next effort was the TCGA, the Cancer Genome Atlas, which was funded and finished in around 2010. It's still going on, really, in different forms, but what it was is a snapshot of tumors at diagnosis. What are the mutations? Where are they? And this was a quantum leap in our knowledge to catalogue all the mutations across different tumors.

One of them was PIK3CA. PIK3CA is part of a very important growth factor pathway, which is employed by many growth factors. When a growth factor is activated, it has to transmit that signal then to the nucleus and the other parts of the cell to coordinate growth, and those signal transduction pathways, as they're called, involve a chain reaction with many different proteins, and PIK3CA is one of them, and a mutation in that protein can stick that pathway in the on position.

And happens to be one of the more common mutations in breast cancer, particularly in hormone receptor-positive breast cancer. If you look at newly-diagnosed patients and sequence PIK3CA, about 40% of them will have a mutation in that particular gene. We don't know why that particular type of breast cancer...and you do see it in triple-negative cancers, and you see it in other types of cancer, as well, but that's notable as one of the more common ones, and there's different types of cancers where mutations in this pathways are more prevalent.

So, upon the discovery...and this really came around the time that the TCGA was published, which was around 2013 or so, and we started learning about that, and then we started developing drugs that actually work against mutated PI3 kinase or the downstream pathways, and it became important now to test patients, and now that's one of the common genes that we test for. Of course, nowadays, we're testing a whole panel of genes, and that's going to be one of the ones that’s included.

But we generally test that in patients...we're only using it in advanced breast cancer at the current time. So, anybody with advanced breast cancer that's hormone receptor-positive, we're generally going to be doing next-generation sequencing and testing for that. There is a drug now approved called alpelisib for patients that have certain types of PIK3CA mutations, and not all mutations are activating mutations.

So, when we do a gene sequencing test, one of these next-generation sequencing tests, it's very important to have what's called annotation. You don't just get the raw data here, the mutations, but you get some explanation as to is this mutation one that is expected to actually be pathogenic to make the cell grow in an abnormal way and actually be an indication for a drug or not. And sometimes, it's not clear, because it's a mutation that just hasn't been reported.

And then they have to sort of deduce what the sequence would do to the shape of the protein, and then they can give you a readout on that. So, it's a very tricky area to have good annotation, but the science has really advanced to the point where we're using a lot of computer-assisted technology and artificial intelligence to sort out what the meaning of these mutations is, and that is part of the report. So, when an oncologist is sending next-generation sequencing, they're getting troves of information.

These reports are hundreds of pages long, and the mutations have to be annotated and explained in such a way that it's useful to the doctor, because there's going to be these rare mutations I mentioned that would be an indication for a drug we commonly don't use. You see them occasionally in breast cancer and colorectal cancer, and then we use...we use those drugs, these melanoma drugs, of all things, in breast cancer, rarely.

And so, we have to really understand the full report and what the nature of the mutation is in order to interpret it. Again, just to remind you, this is all in advanced breast cancer, but that's why it's important to send genomic testing off in anyone who has advanced breast cancer and is interested in treatment, because the results can sometimes really change the course of what we're going to recommend.

Jamie DePolo: Absolutely, and there's another mutation that's relatively new, the ESR1 mutation, and that's the estrogen receptor, and if I'm correct, people with advanced-stage disease and they've been receiving hormonal therapy, like, it's very high. I want to say, like, 75% to 80% are going to develop a mutation in this estrogen receptor, this ESR1, and then they stop responding to hormonal therapy. Am I understanding all that correctly?

Dr. Debu Tripathy: That's correct, and it's very specific to a particular type of hormonal therapy, and this is what we call estrogen-deprivation therapy, when you're lowering the estrogen levels, and one of the common drugs we use, even in early-stage breast cancer, is a class of drugs called aromatase inhibitors, and as we've known, as I mentioned earlier, estrogen is an important driver, particularly in hormone receptor-positive cancers.

And suppressing ovarian production of estrogen was one of the important types of treatments that we still use today, but after menopause, there still is some estrogen in the body, and that comes from androgens that are getting converted to estrogens. Because the ovaries are no longer producing estrogen, but the adrenal glands are making androgens, and they can actually be converted to estrogens in the bloodstream or in the fat tissue, where an enzyme that does is called aromatase.

It converts it, and you can block that enzyme with aromatase inhibitors, and that lowers the estrogen level from still lower than pre-menopausal, but down to an even lower level, and that can make a difference in microscopic cells that may be hiding in the body. So, we use it in early-stage breast cancer to lower the risk of recurrence, because, you know, when we're...the reason we treat patients with early-stage breast cancer with things like chemotherapy and hormonal therapy is to address those microscopic cells we can't see, which, years down the line, can lead to recurrence.

And when you lower estrogen levels, it is very effective at those cells, but every now and then, again, just it's selection of the fittest. If, by random chance, you get a mutation in the estrogen receptor...that makes it now not need estrogen, then what happens is you get these particular mutations in the estrogen receptor, allow estrogen to be activated, even without binding estrogen. Because, normally, when estrogen binds the estrogen receptor, it changes the con... It changes the shape of the estrogen receptor, and it translocates to the nucleus where it then transcribes other genes and causes the cell growth. That's the natural role of estrogen, is to cause growth of things like breast tissue, but other tissue, too, bone marrow and everything.

But when you happen to have an estrogen receptor-positive cancer, then you...that amount of estrogen can actually lead to cell growth. So, selection of the fittest again. Those that have this estrogen...cancer cells that have the estrogen mutation are now going to grow under cases of estrogen deprivation.

So, you're right in what you said earlier. In patients who specifically are on aromatase inhibitors and they develop a recurrence, 30% to 40% of those are actually going to have a mutation in the ESR1. And we've known about ESR1 mutations for decades, but they were thought to be a rarity, and no one really thought that much about it, until we sequenced them and realized what was going on is that the estrogen receptor mutations do make the estrogen receptor work autonomously, even without estrogen.

So, now, when patients progress, especially on aromatase inhibitors, that is one of the panels we check. Of course, it almost gets checked automatically now because it's on all the panels. So, when we do see an ESR1 mutation, number one, we know that aromatase inhibitors aren't going to work, but the drugs...there are drugs that can work, and these are drugs that degrade the estrogen receptor. They're called estrogen receptor degraders or downregulators.

Faslodex, or fulvestrant, is one that we've had around for a while, and we have known that sometimes it can work with these ESR mutations, and when you're on Faslodex, you generally don't tend to get an estrogen receptor mutation like you get when you're on aromatase inhibitors, but it's not a very strong downregulator, and there are some newer, more potent, downregulators, including one called elacestrant, which was just approved in January for patients who have ESR1 mutations. It's a more effective hormonal therapy, and there are some newer drugs in development that are even more...that work even better. So, you know, we expect some new therapies for patients with ESR1 mutations that are going to be approved.

Jamie DePolo: Right, and I know with elacestrant, or Orserdu, it's...my understanding, it's a little bit more convenient for folks because it's a pill, as opposed to Faslodex, which is a poke in the butt, basically. Yeah, so, that's nice.

Dr. Debu Tripathy: That's more effective, too. The study that led to its approval compared it to Faslodex, to fulvestrant, and found it to be more effective.

Jamie DePolo: Right, which is also a very, very big plus. You know, we've been talking a lot about these mutations. So, I know one thing that is sometimes checked for is something called tumor mutational burden, which, my understanding is the amount of mutations that's actually in the cancer. But when would you test for that, and what does it mean when you get the results?

Dr. Debu Tripathy: Yes. There are some people whose tumors have just a larger number of mutations than others, and we now realize that there are certain genes that are involved in this and can lead to just more overall mutations, and they have to deal with how DNA repairs itself and how...and maybe other factors that can lead to higher rates of mutations.

So, there are some cancers that naturally have a higher tumor mutational burden, and those tend to be cancers that arise in situations where they're getting external insults, carcinogenic insults, like melanoma, getting UV -- ultraviolet radiation -- or in tobacco smokers who get lung cancer, those tend to have higher tumor mutational burdens.

And it's postulated that when you have a lot of mutations, that the immune system may recognize these tumors more. So, they also tend to be those that are more immunogenic, and that's why melanoma and lung cancer were among the first to really benefit from immunotherapy, is, in part, because they have a higher tumor mutational burden. There's probably other factors, as well, but basically, it's a measure of how many mutations you have along the entire genome.

So, with next-generation sequencing, you automatically get that information. And you can also get it with liquid biopsies, although liquid biopsies can't sequence the genome as entirely...as well as a tumor biopsy. So, the tumor mutational burden, even though you get a readout from the liquid biopsy, it may not be as good as one that you get from when you do it off the tumor itself, but because the tumor mutational burden, higher ones...higher TMBs, as we call it, have better response to immunotherapy, there was actually a study that looked at patients with high tumor mutational burden, and did a randomized study with two of the checkpoint inhibitors that we have now, dostarlimab and pembrolizumab. And found that these patients responded, regardless of what tumor they had. So, a high tumor mutational burden now opens the door to immunotherapy, even though you may not have PD-L1 or have the other reasons to get immunotherapy.

And so, that is something that you commonly get as a readout whenever you're doing a genomic test, is the tumor mutational burden. You also get another somewhat similar test looking at mismatch repair, which is another way that we repair our DNA. And some people are deficient in mismatch repair, particularly in colorectal cancer, uterine cancer, certain brain tumors. These have a high proportion of mismatch repair, but you can sometimes also find it in breast cancer, not very common.

But if you see that, that's another indication for getting immunotherapy, regardless of PD-L1. So, these are some of the benefits of next-generation sequencing, is sometimes you pick up a needle in a haystack and you get a drug that you never would've thought of for that patient, only because of the genomic testing, and now that genomic testing is so widely available, and all of the companies that do it now are using similar technology, they've almost gotten indistinguishable.

They used to be some of the ones that we all preferred over the others, but to be honest with you, they're all pretty good right now. What's more important now than just the sequencing...because, I mean, high school students are doing next-generation sequencing now. It is really more the informational analysis of it is more important than the actual technical part of it, is what to do with the information.

You know, oncologists are getting reams of information, new drugs. It's really hard to be an oncologist and stay on top of these things if it weren't for all the tools that we need to have. You can go online. The report itself has to be readable and digestible, and it does benefit patients, and we're going to find more and more things that we learn about how to connect the genomic information to new drugs. So, we're only scratching the surface in this podcast.

Jamie DePolo: Yes. Definitely, and scratching a lot of surfaces. And I just want to clarify, too, though, about the tumor mutational burden. So, from what you said, it sounds like something that that would sort of be part of other information you were getting when you're getting the genomic profile of the tumor. Then the oncologist could decide, like, oh, this is very high. Maybe we should consider some of these treatments. So, it wouldn't necessarily be a test that you would just say, let's just look at that. It kind of comes with the other information.

Dr. Debu Tripathy: That's correct. When you send the test, you're looking for everything, because you can develop a new mutation that may not have been picked up on an earlier test. It could be that the mutation was there, but it was so infrequent and present in such a small fraction of cells, that it didn't get picked up. Over time, those cells are growing. The other ones aren't.

You rebiopsy. Now you find it, but you're also going to get the tumor mutational burden. That doesn't shift as much, but it can shift also from one test to another. So, it's pretty much standard now. They run it most of the times when you're in next-generation sequencing. It used to be that you'd have to order what gene you wanted, but now they're just sequencing the whole thing, and you get it whether...you just get everything.

Jamie DePolo: Okay. Okay. And then, lastly, at least in my mind lastly, for things that might be tested for, you mentioned it earlier. Circulating tumor DNA, and I guess I'm a little unclear. Like, when would you look for that? I know liquid biopsies...I've been hearing about them for a while. I still hear that maybe they're not ready for prime time. Some people are doing them, some people aren't doing them. When do you do a circulating tumor DNA test, and what do you do with the information?

Dr. Debu Tripathy: Well, circulating tumor DNA, you're looking at the same mutation that you would on a tumor, and the cells, tumor cells release them into the bloodstream, and you ...you can sequence them. It's not as sensitive as you might imagine, because you're getting much more DNA from a tumor than you are from a small amount that's in the blood, but sometimes, it's just a more convenient way to do it. Sometimes it's not safe to get a biopsy because of where the tumor is.

And so, we do send the liquid biopsies. It's easier to send, less invasive, and as the technology gets better, it'll probably get closer and closer to approaching what a real biopsy gives you, but when you have an opportunity to get a real biopsy, it's probably better, because there are certain mutations, like PIK3CA, where, if you don't see it in the blood, yet you've got hormone receptor-positive breast cancer where you definitely want to know, then you actually go to tumor to get it.

Now, ESR mutations, oddly enough, are easier to pick up with blood than tumor. It's one of the few that goes that way, and that's because ESR1 mutations tend to be subclonal. They tend to develop later, so only some proportion of the cells are going to have it, and if you have multiple tumors of the body, it's possible that some have ESR1 mutations and other don't. But the ones that have ESR1 mutations may be, you know, more aggressive, right? So, you still want to know, even if it's a small population. You still might use elacestrant, for example.

So, in that situation, it's actually better to do a liquid biopsy, but for most of the other mutations, we tend, at least at some point in the patient's trajectory, to get a tumor biopsy, but because we're now sequencing more often and upon recurrence because new mutations might develop, the blood one seems to be more convenient. The best time to use a blood assay is when a patient is having progression, because once they get treated, the amount of circulating tumor DNA goes down, and so, that's when we tend to do it.

If someone's progressing, we might get a liquid biopsy at that point in time. Or sometimes, when one area is progressing, then we'll actually get a biopsy, because we...and of that area because that may be where the action is, and we may want to look at that. This whole idea of tumoral heterogeneity, that not all the tumors are the same, is a complicating feature, but we're recognizing that that's the reality, and our technology needs to adapt to be able to widely survey everything that's in the body.

Jamie DePolo: Okay. Now, what is the difference if you're going to test for circulating tumor DNA versus circulating tumor cells? Because I've read about both. It seems like DNA would give you more information, but that's just my impression as a non-oncologist. So, would you order both tests for one person, or is one more preferable than the other? Does it give you different information?

Dr. Debu Tripathy: It does give you different information. First of all, circulating tumor cells are less common in patients with cancer. The technology to detect them has improved. We can detect, you know, much smaller numbers, but only 5% to 10% of patients may have circulating tumor cells. That number's going to be higher when they're progressing versus when they're on a treatment that's working.

Even in early-stage breast cancer, we sometimes see circulating tumor cells, but circulating tumor DNA is a little more efficient, and you can sequence it more easily. If you happen to see circulating tumor cells, you can sequence them, but it's much harder. So, circulating tumor cells are helpful in prognostication, but we're not using them so much in diagnosis anymore.

They're still important research tools, and there's a lot of interesting research around them, but I would say that that technology is falling out of favor due to its complexity and the fact that it's not capturing everything. But it is prognostic, and there still is a role for it, and certainly, in the research area, it's an important tool.

Jamie DePolo: Okay, and then a couple other experimental...at least as far as I understand, biomarkers are tumor-infiltrating lymphocytes, and you talked about lymphocytes earlier, that it's a type of immune cell, and then stromal tumor-infiltrating lymphocytes. So, if you could tell us just a little bit about each of those, and I guess they're more being used in research, is my understanding, than sort of a test that you might order commonly for somebody diagnosed with cancer. If you could talk a little bit about those?

Dr. Debu Tripathy: Yeah, that's right. It's not being used, really, in routine clinical practice, but we've known for years that people who have lymphocytes around their tumor have a better outcome. They actually do better with early-stage breast cancer and especially triple-negative breast cancers that are getting standard treatment, which has been chemotherapy for many years. The outcome of those patients is better.

Now we know that that probably has to do with the fact that these tumors are more immunogenic, although it may be more complicated than that, and not only are you getting lymphocytes in the area of the tumor itself, but you get it in the peritumoral area, too. And so, you'll get stromal TILs, or tumor-infiltrating lymphocytes, and both of those are prognostic of a better outcome. Even without immunotherapy, even if you just get chemotherapy, those patients do better.

We’re not routinely using it yet in decision analysis. At some point, we probably will, once we learn more about the outcomes of these patients, and it may help us decide...sharpen a little more who should get this treatment versus that treatment, whether it's immunotherapy or chemotherapy or certain types of chemotherapy, like platinum-based chemotherapies, which seem to be more effective in the more tumors as well as in BRCA-mutated tumors, which also, by the way, tend to have more immunogenicity.

And we tend to see infiltrating lymphocytes more in tumors that have DNA repair deficiencies, and again, it may be because they have more mutations. But in any event, it still is a research tool, but a lot of people are building it into their prognostic models. So, I suspect, at some point, we will be integrating it to make treatment decisions as we get more data.

Jamie DePolo: Okay. Dr. Tripathy, we have talked about a lot of biomarkers. Are there any important ones that you feel like we've missed? I feel like there are probably more experimental ones that we could talk about, but are there important ones that are more commonly tested for that I just have overlooked or that we didn't talk about?

Dr. Debu Tripathy: No. You've done a great job in covering all of these. I think we've talked a lot about, you know, the basic ones that are used in decision-making and even the ones that are sort of in the research field, but I think may, someday, be important. So, I will think about that question longer, and I may come back to you later with more comments on that, because the field is moving so rapidly.

But certainly, we will be developing a bigger framework with more markers over time and better support tools so that physicians and patients alike can know what to do with this information, and it really becomes a responsibility, really a societal responsibility for us to educate everybody and to make sure it's widely available. The good news about genomics is that it's actually relatively less expensive than many of the other things you do.

I mean, in China, I visited some of the hospitals there. They've got Illumina 500 sequencers on a whole floor of the hospital just running nonstop. They're sequencing a lot of their tumors. Yeah, they're just starting to, but they're...you know, and other countries are, too. So, I think that this is going to become even more broadly available and accelerate and really turn the lights on the road for us to really see where we're driving.

Jamie DePolo: Dr. Tripathy, thank you so much for helping us understand this very complicated topic. I really appreciate your insights.

Dr. Debu Tripathy: Thank you for having me.