Charles Ryan: Hello from PCF 2019. I have a great group with me now. We're going to talk about PARP inhibition, DNA repair and the recent clinical data and what it tells us and what it might not tell us. I've got Elena Castro, who's a Medical Oncologist at the Spanish National Cancer Research Center in Madrid. I have Wassim Abida, who's a Medical Oncologist from Memorial Sloan Kettering Cancer Center, and Karen Knudsen who's a Professor and Director of the Sidney Kimmel Cancer Center at Thomas Jefferson University in Philadelphia. What a great group.

I want to start with the question that I have on my mind, which is, I'm looking at the PARP inhibitor response data and I'm a little bit surprised. The response proportion is in the 40% to 50% range, and I might have expected that it would have been a little bit higher. So Elena, what's your response to that?

Elena Castro: Well, I think these studies are not perfect, but this is what we have at the moment. So we're probably not selecting correctly those patients that are most likely to benefit. And maybe we may be mixing some that are likely to benefit with all those who actually don't have a totally DNA repair functional loss.

Charles Ryan: Right. So what would be an example, Wassim, of a situation where we're getting mixed results from our genomic testing?

Wassim Abida: Yeah, so I look at it in a glass half full kind of picture and I think the data is very promising. But we work with the biomarkers that are clinically validated that we have so far, right? So we use essentially genomic tests that identify a single alteration in BRCA2 where the indication is strongest, but we need to refine that biomarker in time. We have to identify is this a clonal event? Is it biallelically lost? And does it really result in what you expect that homologous recombination deficiency, which is what sensitizes tumor cells to PARP inhibitors. So it has to be refined in time, and it will be. And now that we have the data sets, we can start refining the biomarkers. 

Charles Ryan: Right. So there's sort of three variables in here, right? We have the mutation that we're treating, we have the drug that we're using, and the assay that we're using to assess. And all three of those, Karen, all three of those issues need to be sort of working in the right direction for the proper selection for precision medicine to really work. So as I'm thinking about the assays and sort of what we need to do, what are the limitations you see of the current state for testing in the clinic and the decisions we make based on those assays?

Karen Knudsen: Yeah, I think there are a few things going on there. I mean, I think if we look at that 40%, 45% response rate, that's not so different than selected trials historically, even in other tissue types. So I personally wasn't as surprised as you were, but I do very much agree with the optimism and that we can confidently say that pathogenic BRCA2 alterations may enrich for a better response. But we can also say that that's the tip of the iceberg. The flip side is we'd like to get those numbers to be the majority, not the minority of responders.

And I think some of that will come with exactly what was discussed, the complex mixture of genetics. Is it the stage of disease that we're treating, is that the right place? The right combination. But also are we really truly assaying all the things that are associated with PARP biology? So there's an entire discipline that's been studying PARP families for years and appreciating the roles in DNA repair, which are very important. And the synthetic lethality is clearly an important component of the response.

But other functions of PARP may play in. It has a known role in protecting the ends of chromosomes, a known role in regulating gene expression, and a known role in regulating a cell's ability to put itself to death. So all of those things may actually relate to who within even just the BRCA2 pathogenic mutant pool is a responder versus not. And it's this kind of collaboration amongst all of us, those of us who study DNA repair and PARP, and those are who are administering the PARP inhibitors and doing the genetics that will lead to the answer.

Charles Ryan: So Elena, you've done some work that shows that identification in an early stage of the disease of a BRCA2 alteration can predict, or be prognostic, I should say, for the eventual outcome of that patient who is receiving local therapy. So do you think based on your data that most of the benefits that we're seeing in these patients come from the primary, come from the early stages of the disease? In other words, is the genetic defect there early or are we perhaps going to be targeting many patients for whom it develops later? And that's going to be your question. So localized prostate cancer, the initial biopsy, is that the most informative case?

Elena Castro: For what it's been, perhaps at the moment I think in most cases the alterations in BRCA are already present in the primary tissue, even when there may be some cases that acquire it later. And possibly some of them will acquire a second event later on the progression of the disease, but probably Wassim knows better.

Charles Ryan: Well, when we looked at the results that had been presented, we haven't seen much of a difference, if I'm not mistaken, between the germline versus somatic alterations.

Wassim Abida: That's right, not much difference. But I do want to add to that, and Elena's work was extremely valuable. I do think that at least for BRCA, and I don't think BRCA is the only story, right? There are other genes, loss of RAD51 homologues, and PALB2 certainly, but at least for BRCA, it looks like even when it is lost somatically in those patients who have somatic loss, it's an early event. At least based on a case series of match samples, where the patient has the primary latent metastasis. So it does seem like if you detect a BRCA loss, that it is an early event. And Elena's data pointed out to really poor outcomes for patients who have localized disease with BRCA loss. I suspect she showed it in germline, I suspect some of the same probably will be true in patients with somatic loss. And so talking about new avenues for BRCA, in a case where it's certainly obvious, is probably we need to test PARP inhibitors upfront for those patients who have really bad outcome.

Elena Castro: Definitely.

Karen Knudsen: Yeah, I agree.

Charles Ryan: Yeah. I want to come back to that point and just ask the question, so if a clinician is listening to you and thinking, "Oh are you telling me that I don't need to do biopsies now of metastasis to rule out a BRCA alteration if I have the primary?"

Wassim Abida: You know, I think we need to learn more, but certainly if you don't have the ability to do a biopsy, I would argue that sequencing the primaries is helpful. Certainly, there's new technology... You can sequence a germline, but we know that sequencing the germline you'll miss half of the BRCA cases because it can occur somatically. So I think that sequencing the primary can be very informative in that situation.

Charles Ryan: But a negative test in the primary, does that mean you don't need to look further, it's a negative test?

Wassim Abida: You know, it's a little tricky to answer that question because this issue's related to how good of a primary sample do you have? Is it more than five years old? It has a high failure rate. So there are issues related to sample quality when it comes to assaying, which unfortunately are complex. And so it does come down to that a little bit. But if you have a high quality primary sample, I would argue that that is a good thing.

Elena Castro: I will add to that that it also depends on what is the primary sample you have. Because if it's a biopsy just with a limited number of cords, you may not be picking up the alteration.

Charles Ryan: Yeah, yeah. So, still a lot of work to do there to figure out that algorithm, so to speak.

Karen Knudsen: And so many variants of unknown significance that we're really not sure yet to handle, and that knowledge will get advanced based in part on our colleagues in breast and ovarian cancer and help inform us about something that today we might not pay attention to, but tomorrow could be informative.

Charles Ryan: So, will variants of undetermined significance, can they become of known significance if we understand the significance of them over time?

Karen Knudsen: They can. Hopefully everything that's unknown significance will ultimately be tested.

Charles Ryan: Become known.

Karen Knudsen: And when you're dealing with a gene as large as something like BRCA2, you can imagine that a genomically unstable tumor just by law of averages is going to have an alteration. So really helping to segregate what's truly pathogenic versus what is weakly associated with pathogenicity is going to be an important aspect of the field moving forward.

Charles Ryan: So one question I have is, we're talking about BRCA2 alterations, we're talking about DNA repair defects, and then we're kind of making the assumption that inhibiting PARP is the only way we can deal with this problem.

Karen Knudsen: Yeah.

Charles Ryan: And so my question for you is, are we right in thinking that or are there going to be better targets that we should be thinking about in the context of DNA repair?

Karen Knudsen: Well, I mean, one of the great things about a PARP inhibitor is how well, in general, tolerated they are compared to chemotherapy, and the known synthetic lethality with homologous recombination defects, and in prostate especially with regard to a pathogenic BRCA2 background. We already talked about the fact that we can do better. Are there other targets? I think there will be. We know that there are DNA repair factors that are downstream of androgen signaling and the androgen supported double strand DNA break, like DNA-PK, and trials ongoing in that space. ATM inhibitors I know will be examined.

So I think in the next few years we'll see a lot more trials reporting out on these, but I think what we'll also see is much more information rolling out from our colleagues in the breast and ovarian cancer space. And I think there's a lot to be learned there, especially with the concept in ovarian cancer that the best indicator of who might respond to a PARP inhibitor may be in fact responsiveness to platinum. And that that, in that case, outperforms the status of a detectable DNA repair mutation. So I feel like as a field we should watch that very carefully as well and determine which portion of that plays out in prostate.

Charles Ryan: So final question, and this is great. Final question is getting back to the point you made earlier, which is, if we detect DNA repair alteration early, how important is the androgen deprivation in the use of a PARP inhibitor? Because in other cancers they don't use androgen deprivation therapy and they use PARP inhibitors. So is there a potential window for us to use these without ADT, or is there something about the testosterone depletion and the PARP that's necessary?

Elena Castro: I think we need to have the results of those clinical trials that are addressing whether PARP inhibitors could be used as, for instance, adjuvant therapies or in advanced disease without concomitant ADT.

Charles Ryan: And those trials are ongoing. Okay, agree?

Wassim Abida: I can think of at least one trial that's ongoing. I mean, to be honest, I think there will be responses to PARP monotherapy. But unfortunately, PARP inhibitors still don't cure patients with BRCA2, so we're still thinking sequential treatment for now. And certainly, it seems that although patients who have BRCA2 alterations can have a shorter time to castration resistance, from Elena's data, they certainly respond about the same to next generation hormonal therapies once they're castration resistant. And so they do have similar responses to hormones, and hormones have to be part of the treatment paradigm for BRCA patients at this point.

Charles Ryan: Closing thought?

Karen Knudsen: I would absolutely agree with that. So we know that clinically there has been evidence that there is a benefit to having hormone therapy in combination with a PARP inhibitor. But it's also the case that we all know that castration or hormone deprivation is a great radiosensitizer, and the mechanisms behind that are really known. It's actually one of the only clinical radiosensitizers that we actually have is hormone deprivation. So I would suggest that biologically there's a lot of reason to believe why that would be the case, but more clinical studies are certainly needed. But from everything that is a state of knowledge today, I would say yes, I think that's a good idea.

Charles Ryan: Great. Well, what a great conversation. I wish we could continue. And although I started out on a slightly pessimistic note, I was just trying to get the conversation going. I'm very optimistic about all of this as well.

Karen Knudsen: I think we all are. Yeah.

Charles Ryan: So thank you so much for joining us and enlightening all the viewers on this really important topic.

Karen Knudsen: Thank you for having us.

Wassim Abida: Thank you, Chuck.

Elena Castro: Thank you.