Read the Full Video Transcript

Thomas Keane: Good afternoon, everybody. This is Tom Keane coming to you from the Medical University of South Carolina and from UroToday. Today, we are delighted to welcome Dr. Bruce Montgomery, who's a GU medical oncologist and professor of medicine and oncology at the University of Washington and the VA in Seattle, where expertise in translational medicine are relevant for this presentation. He has an exceptional interest in better targeting of DNA repair deficiency in prostate cancers and ways of finding those who can benefit from this technology. He's very kindly agreed to give a presentation on a recent paper, which was published in JAMA Oncology. It was entitled, Concordance of DNA Repair Gene Mutations in Paired Primary Prostate Cancer Samples and Metastatic Tissue or Cell-Free DNA. It's a pleasure to welcome you, Dr. Montgomery, we look forward to hearing what you have to say. Thank you.

Bruce Montgomery: Well, thank you, Dr. Keane, this really is a great opportunity and I really appreciate the opportunity to talk to you and your audience about something that I've been working on for the last couple of years. So, as they oftentimes say, the bottom line up front, we did this work mostly because we were trying to find a way to make it easier for oncology providers to find the men, who we're all trying to find, men who can benefit from targeted therapies because some of these targeted therapies can be exceptionally effective. And so finding those patients is proportionally just even more important, so finding as many ways as possible to get this therapy to them is something that we've been invested in for a while.

It's also important to consider this in the background of how the field has moved over the last several years. About 5 years ago, a number of groups participated in a project, which we also participated in, where we were biopsying metastases of men with advanced prostate cancer. And by sequencing those metastases, we found many very interesting and important alterations in DNA, RNA, and proteomics for these patients. But the one area that was an immediate epiphany was the idea that DNA repair deficiency was much more common than we had imagined before, and that was most important because there were already FDA-approved drugs to target those alterations, although they hadn't been approved specifically in prostate cancer.

So, to walk through what we had found and what we think is important is this is a very nice schematic from a colleague of mine, Dr. Heather Cheng, looking at the different types of DNA damage and the relevant mechanisms for repair. Here you can see the very many different ways of developing DNA damage, which results from exposure to many various things in our environment. And then I'm going to focus on two types of DNA repair, that is double-strand repair and mismatch repair.

In the double-strand, repair homologous recombination is the most important high fidelity pathway in the genes, mostly BRCA and a few others that are involved in that. Patients who have these alterations can benefit from PARP inhibitors and platinum chemotherapy, at least from the information we had from patients dealing with other malignancies that are BRCA related. And then mismatch repair and the genes that you, again, can see in the schematic, are really important because immunotherapy can be exceptionally effective in for these patients. And so, again, as I said before, finding them was something that we were very invested in and doing in the way that we could.

Once we had found those alterations in DNA repair, it was sort of a race to show that, in fact, in prostate cancer, these targeted therapies were also effective. We already knew actually, from earlier work, but then it was confirmed in a larger study from Hopkins, that if you used immunotherapy in patients with any malignancy with mismatch repair deficiency, up to 50% of patients will have prolonged, in some cases, complete responses, which obviously is extremely important for patients. And then two PARP inhibitors were FDA approved for targeting BRCA alterations, both rucaparib and olaparib.

So because these are very effective drugs and there were drugs now available for targeting these patients, the idea was how can we best find them? Doing metastasis biopsies, which we had done as part of the original study, is something that is not necessarily very straightforward. And although circulating tumor DNA is really the path forward I think in many situations, in men with prostate cancer, if they aren't having progression of their disease, it's oftentimes not a productive effort to try to get circulating tumor DNA because you won't find anything in circulation.

This is a schematic from a group, a very nice paper, in Cancer Cell several years ago, in which this idea of trying to figure out where along the pathway of cancer development these alterations occur. We think that we start with a single clone of an alteration or alteration, the tumor then actually starts to expand and starts to acquire these other alterations along the way. But the question, we were very interested in things like BRCA2, which is a DNA repair defect, does that start very early? Does that start later? Do we need to actually be looking very late in the course of the disease rather than very early in order to find these patients who can benefit? So that was the main question and that's what we set out to answer.

This was the paper that Dr. Keane was referencing earlier. This was just published in JAMA Oncology several weeks ago, and the title pretty much says it all. We were looking at what the concordance was in primary tissue, that is either a prostatectomy specimens or prostate needle biopsies, and the agreement between those specimens and either metastasis, biopsies, or circulating tumor DNA from the same patient. This was a collaboration across multiple institutions and collaborators. We worked with Foundation Medicine, we worked through the WU system and the VA.
We found 72 paired samples. 21 of them had to be excluded because some of them were germline, and in patients who have germline alterations, you're always going to find the alteration in every tissue that you sample. 12 of the patients had what's called CHIP, which is clonal hematopoiesis, which was pretty evident from some of the circulating tumor DNA samples, and so they were excluded because ship events do not reflect somatic alterations from the tumor. We ended up with about 50 patients and you can see the details of what the various tissue samples were in those patients.

The money shot here is this schematic, which you can see. Over on the left-hand side, those were the genes that we were looking at across all the samples. Each individual column represents a patient, and within each of these larger boxes, so P is primary, M is metastasis, and B is blood, in some patients, we had all three samples, in other patients we just had to. Trying to make a fairly complex schematic a little bit simpler, everything that is green agrees, everything that is red does not.

And so when we look at the patients in which we were unable to show concordance, it was interesting that essentially most of the samples that were not in agreement came from some genes, which actually, in retrospect, we probably shouldn't have included in the analysis. They are DNA repair genes, but alterations in these genes don't actually predict for responses to PARP inhibitors or platinum or immunotherapy. So over half of the non-concordant or discordant samples came in genes that aren't really important anyway, so at least 90% of patients who had an alteration in their metastasis or their circulating tumor DNA that could be targetable, had it in their primary prostate tissue.

So again, not a lot of figures in this paper, but I think we pretty clearly showed that if you're looking for homologous recombination deficiency that's targetable by using either platinum agents or PARP inhibitors, you can use primary prostate tissue to find those patients, which can be a real advantage because those samples are oftentimes in-hand already. Essentially everybody has undergone some sort of a diagnostic biopsy or a prostatectomy, and that's available essentially at any time. Whereas things, as I said, like circulating tumor DNA, are sometimes very difficult to interrogate unless patients are having progression, which is not when you want to be scrambling around trying to find an alteration to change therapy, or in the context of a metastasis biopsy, which can be impossible in a lot of patients.

We did have some mismatch repair deficiency patients in this study. I don't think the number was high enough to definitively say that we can use primary tissue to find all these patients, but at least in the patients that we had, there was very good concordance in the primary tumor tissue.
And then the final point that I think came out of this, which I think many people in the field are beginning to recognize, is that this issue of what's called clonal hematopoiesis, or CHIP, which is really alterations that occur as a result of living longer. That is, the older we get, we start to accumulate damage to our normal bone marrow, and these somewhat dysplastic clones start to grow out of our bone marrow. And so when that starts to circulate, some of those genes that are CHIP, are genes like ATM or BRCA1 or CHEK2, all of which are potentially targetable alterations. And so, if you find an ATM alteration from a previous paper in JAMA Oncology from our group, the likelihood that that ATM represents cancer is actually lower than it is that is actually CHIP. So it's very important when you look at circulating tumor DNA to evaluate the potential for that.

Thomas Keane: Thank you very much. Again, it gives things in new air, or there's a lot of questions that one would ask about this. Is this potentially a means to identify those patients who are likely to develop metastatic disease?

Bruce Montgomery: Yeah, that's a very, very interesting, and probably a much larger, question, I would say. We do know that, for example, patients who have underlying germline alterations are definitely at higher risk. So for example, patients who have BRCA1 or BRCA2, and those alterations are expressed in every tissue in the body of somebody who is a carrier of that alteration, those men are definitely at higher risk of developing prostate cancer and higher risk of developing metastatic disease.
For patients who have germline mismatch repair deficiency, it's a little less clear what the overall risk of metastasis and developing prostate cancer is. I think if people have that, it is in the spectrum of the Lynch syndrome cancers as well. We do know that germline alterations do increase the risk of developing metastasis in some of these patients. I think the hard part is, for example, in somebody who isn't a germline carrier, is there some sort of alteration that occurs in the prostate along the way in one of these genes that, just by its very nature, clearly increases that risk? I don't know that we have an answer about whether these tumors are clearly higher risk for metastasis. Although, there are a number of papers that would suggest that is the case.

I think looking for the alterations right now is obviously mostly focused in patients who have advanced disease, because there's something that we can do with the results. I think the question that you asked, which is a much bigger and probably much more important question is, how does that figure into the risk profile of patients who have localized disease? And can we use that information, for example, to try to reduce the risk that they will have a relapse as a result of micrometastases that have already occurred? There have been a number of studies that have been started looking at using neoadjuvant therapy, for example, on patients undergoing prostatectomy, using things like PARP inhibitors and platinum therapy. So, people are trying to answer that question about, can we leverage those findings in a way that's going to be really helpful to men, but I don't think we have the answer for that today.

Thomas Keane: Yeah, because that would be, to your average urologist, all we have is the Gleason grade and MRIs and basically imaging to show us what we think is high-risk prostate cancer, and then we occasionally find the 3+4 or the 4+4 that you didn't really think was going to do anything, that ends up killing the patient, but certainly it's 3+4 would surprise you. But this, your presentation, to a guy who is the field, if you like, I'm taken out of prostate for doing stuff like this, and I'd much rather have a clearer mechanism by which I could say you're likely to get into trouble, even if you're at least in greatest discordant with what we're looking at from your genetic makeup. That's the person that I really want to hit hard. And perhaps somebody who doesn't have that would be somebody that you might be more tempted to, perhaps, assume active surveillance. And again, as you say, we're at the beginning of this, but it certainly would be a huge potential and a massive gain in the field.

Bruce Montgomery: Absolutely agree and I think there are a number of projects going on nationally, focused primarily in detecting germline alterations in men who are being considered for various localized therapies. And by finding them, then finding their families and how they can all benefit from earlier detection and early appropriate therapy. So I think this is an area that people are very excited for all the reasons that you and I are. That this provides a potential point of leverage much earlier in the course of the disease and we all hope that that's really going to be something that ends up being something that we can use to cure more men and I think that's obviously what many of us are focused on.

Thomas Keane: Well, Professor Montgomery, thank you so much. It was a wonderful presentation. I really think it's maybe the beginning of a big change, and certainly being able to detect it in the original specimen is a huge advantage to trying to detect it further down the line. I wish you all the best with your future work and thank you for coming on UroToday.

Bruce Montgomery: Thanks so much for the opportunity. You folks do very important work and we would be happy at any time you want to invite us back to participate.

Thomas Keane: Oh, we will indeed. We'll be keeping an eye on things. Thank you very much.