Introduction: This Week on RTP

DR LOVE: Good afternoon, everyone. I’m Neil Love from Research To Practice and welcome to PARP Inhibition in the Management of Prostate Cancer – Where We Are Today and Where We’re Heading. We have a great faculty today, Dr Johann de Bono from the Institute of Cancer Research and the Royal Marsden NHS Foundation Trust in London, and Dr Fred Saad from the University of Montreal Hospital Center, Montreal Cancer Institute in Montreal, Canada.

If you have any questions or cases you'd like to run by the faculty, just type them into the chat room and we'll bring up as many of these as we have time.

If you're into audio programs, check out our Oncology Today podcast series including a recent program on metastatic CRPC with Dr Yu.

We do webinars all the time now. Next week, we’re starting a new series on CML with Dr Cortes from MD Anderson, and another new series on non-small cell lung cancer in patients with EGFR mutations with Dr Neal. Then, we’ll continue our ovarian cancer series with Dr Matulonis on July 6^th. And the next day, starting a new series on hepatobiliary cancers with Dr Abou-Alfa from Memorial. Then, we’re going to continue our upper GI series with Dr Klempner from MGH. And then, we’re going to do a special webinar with Dr Stone from Dana-Farber looking at what happened at ASCO and EHA in AML and MDS. That’ll be on, let me see here, what’s the date? Yeah, on June 23^rd. But today, we’re here to — oh, that’s actually today. Anyhow.

Today, we're here to talk about PARP inhibitors in prostate cancer. And as we do for many of our webinars, I met with each faculty person separately and recorded a 45-minute presentation and discussion. Professor de Bono talked about some translational work that he’s done leading the field and the biology of the use of PARP inhibitors and then trials in advanced second and third line therapy. And then, Dr Saad presented more recent data, including data he presented looking at the PROpel trial as well as the MAGNITUDE trial that was presented at the GU Symposium.

I just want to start out with a couple other facts before we sort of jump into this. And, Johann, we saw, I think it was today actually, a press release saying that dabrafenib/trametinib now has approval that’s tumor agnostic. So we have another. I think we have one with NTRK and, of course, we have it with MSI. We have it with high PD-1 with pembro. Now, it looks like — well it is going to happen with BRAF. Johann, I’m just kind of curious, is that the path you see PARP heading on?

PROF DE BONO: First of all, I’m delighted to see this approval. This is a big deal for the non-melanoma BRAF mutant tumors. And I’ll think there will be further combinations that will further improve this subset of diseases, combination with FAK inhibitors, for example, or maybe even VEGF inhibitors have been muted and are ongoing, those studies. So I think for the whole field, there is increasing interest in indication agnostic, I guess, approvals, mismatch repair and pembro being the first one to lead that space. With PARP, I’m not so sure. There are data from Sloan Kettering published in Nature that BRCA mutations aren’t all the same. BRCA mutation in lung cancer may be quite different to a BRCA mutation in, say, ovarian, breast or prostate. So we’ll need to see.

DR LOVE: So I was just corrected. I guess Dr Cortes has moved to the Medical College of Georgia. Well congratulations, Dr Cortes. Congratulations to them. But that’s very interesting in terms of what you were saying in terms of PARP across tumor types. This comes up all the time, Fred. We get people with non-small cell lung cancer. Who knows? Bladder cancer, they run out of options and there’s an NGS report with a BRCA pathogenic mutation. I hear people trying PARP inhibitors. I occasionally hear about them working. Any thoughts about whether we are going to move in that direction?

DR SAAD: Well I agree with Johann. I think we need some data to suggest or to prove that it is agnostic because until we do that, the risk is we might be exposing patients to drugs that are unlikely to help. But the thought of being biology driven rather than tumor tissue driven I think is very attractive. And we’ve been talking about this for as long as I can remember, of treating patients based on their biology rather than on their histology. That is not always the best approach for many cancers.

DR LOVE: It is interesting though if you think about BRAF though. Like in colon cancer, they have to add an EGFR antibody to get a response. So maybe it’s not going to play out exactly the way we’d like it.

Here’s where we’re heading on this symposium. There’s some interesting things that happened in some webinars we did this week that I just want to run by you to get started because I think it might be relative to what we’re talking about today. Then, we’re going to get into what I call my top 10 list. I made a list of the 10 questions I’m hearing the most from docs in practice about PARP inhibitors in prostate cancer, including questions I had as I was sitting there listening to your presentations. And those presentations are posted in the chat room, incidentally. So if you want to check them out afterwards, they’re there. But I put together a list of the top 10 questions. I’m going to kind of run through them. These are the practical clinical questions. And then, we’re going to sort of pick through some of the data you both presented and a couple cases you presented.

But first, I want to start out with a couple things that happened this week I just want to get your take on. So first of all, we had our ovarian cancer series on Tuesday, 2 days ago, and Dr Westin from MD Anderson did that. And, of course, we talked about the big ATHENA PARP trial that was just presented at the ASCO meeting. There were 3 big Phase III trials that really have driven therapy in ovarian cancer prior to this with olaparib and niraparib. But this is now a new Phase III trial. I was all excited. It looked positive. It was kind of what you would expect. Not that much different than what was seen with olaparib and niraparib respectively. But then to the surprise of many of us, all of a sudden, a few days after ASCO, we see this FDA thing that the FDA actually is not going to review this because of concerns that a prior trial in the later line setting, ARIEL4, there was an increasing death rate in a subpopulation there that concerned the FDA. Particularly since you already have olaparib and niraparib, you can understand why they’d be conservative. So they’re going to hold off until they have some survival data.

But, Johann, the reason I want to bring that up to you is one of the things that you got into in your talk, and there’s so much great translational discussion in your talk, is the overlap between platinums and PARP in terms of cross resistance. And the fact that since this “mortality deficit” was seen in people resistant to platinum, to me, that makes it a different story. It’s almost like second line therapy in a way which is really a different question, but certainly maybe not expect to see any benefit in those patients. Any thoughts? And anything you want to say about this crossover between platinums and PARP, Johann? And also, the strategy that you use in ovary which is to give platinums first and then use PARP if they’re not progressing which is very different than prostate, pancreas and breast. Any thoughts though, Johann?

PROF DE BONO: Well first of all, the strategy used in ovarian cancer is quite reasonable. In patients who have been deemed to be platinum sensitive who have responded to platinum, at the end of their course of platinum, if they’re still doing well, they get a PARP inhibitor as a maintenance. And that sounds quite reasonable because the platinum re-sensitivity itself is a predictive biomarker that also tells you the tumor may be PARP sensitive.

I’m interested in the FDA’s call really. I think the FDA are saying there in that press statement, that actually they want to see overall survival. And I do think that is important. And one thing that I keep thinking about with these and other drugs is we need to focus on doing no harm and not claiming victory unless we’re sure we understand about the degree of benefit in these populations. But certainly, PARP inhibitors work in ovarian cancer. They’ve improved outcome. Obviously, ovarian cancers have a very high proportion of tumors with these DNA repair defects to the BRCA1 which is uncommon in prostate cancer. But as you rightly said, you cannot really compare the ovarian cancer trials to the prostate cancer trials. It’s a very different ball game.

DR LOVE: So one other thing that happened, I was curious, Fred, what your thoughts are. So last night, we did a symposium on upper GI cancers with Dr Van Cutsem and one of the things I was asking about, one of his papers which looked at, you know, a therapy that’s been out there a while, antiangiogenic ramucirumab plus paclitaxel. But the thing that I found fascinating about it was they attempted to plot out, using sort of a waterfall approach, what happened to people’s symptoms. And this graphic is actually a little more complicated than it seems. But without getting into the details, I just want to go back to you, Fred, and ask you where things are. Quality of life is so important. They have these scales, like the FACT scale, et cetera. I read that and I can’t figure out what it really means to say to a patient. I like the idea of trying to get more information from patients about, you know, self-reported symptoms, even to the ability to see individual patients as they try to do here. Any thoughts about that, Fred, you both have been involved with big Phase III trials, about trying to capture quality of life data in that situation?

DR SAAD: Absolutely. It’s more of a challenge than people realize. Quality of life is much harder than pure efficacy, survival, progression free because you can lose subtleties in favor or in disfavor when you give these large questionnaires. So I actually liked the idea of focusing on specific quality of life issues and repeatedly asking those questions to get a sense of, over time, how things are going. And when I look at those kind of waterfall plots, it gives you very quickly the idea, are you doing harm to patients by adding therapy? Because that’s the first thing we would like to reassure. And then, are you having benefit actually with the addition of more therapy? And we did some kind of analysis like this in the nonmetastatic CRPC setting where patients are very well when they come in and then you’re adding a drug like apalutamide. And so we specifically focus on some of the adverse events that people most often complain about and looked at between the placebo and the treatment. And it was reassuring that upfront, there wasn’t much difference, but other time, you started to see a difference in favor of actually treating the patient because as the disease progresses, the placebo arm is starting to do worse. But I think you have to do more and more of this kind of — you need the overall picture of quality of life, but if we want to get some specifics, you need to focus on specific issues that are of importance.

DR LOVE: I just think that patients would love to see more information that they could really relate to about what to expect, particularly if we’re going to say, hey, it may be a little bit rough for the next couple months with this PARP inhibitor, GI, et cetera, but you’ve got symptoms from the disease and maybe that’s going to get better with time. All right.

PARP Inhibitors in Prostate Cancer — Top 10 List

DR LOVE: Well let me jump into my top 10 list. And I’m just going to kind of go through this like we’re making rounds here. We know a lot of people end up listening to our webinars so they can multitask and exercise. I’ve heard about people raking the leaves. So if you’re out there raking the leaves right now, here we go. So we’re going to start with the workup for patients with metastatic prostate cancer, Johann, both patients with family history and without. The message that we get, of course, now you see not only with ovary, but now breast, adjuvant PARP inhibitor, adjuvant olaparib, so the genomic workup becoming more and more important. Johann, BRCA — or germline panel is something I think a lot of people agree on. But what about somatic testing, NGS? Some people do it on reflex. Some people send it all at once so they can get it quicker. Some people don’t do it. From your point of view, Johann, what’s the ideal workup genomically?

PROF DE BONO: Well I think that getting germline and somatic sequencing today in a man with metastatic prostate cancer is, you know, a mandatory thing to do. I guess the real question is when should we do it? It partly depends on funding availability. One could, if funding was not an issue, argue the case for doing this both at diagnosis and for somatic, later on in the disease. Because actually, resistance to hormonal therapy, particularly abiraterone, enzalutamide, et cetera, can involve, as we’ve shown in a paper published by Sumanasuriya et al in European Urology, you can actually have evolution of loss of chromosome 13 that can sensitize to PARP inhibition evolving with hormone therapy. So ideally, you want to do it at least once in the patient’s disease history, probably early on. However, if the patient develops, for example, liver mets with more neuroendocrine-like disease, I do think there is rationale for getting a further biopsy if you can and repeating the genomics even if at baseline, there was no evidence of a a DNA repair defect.

DR LOVE: Fred, I’m curious how you’re approaching genetic evaluation. I always thought that more of that should have been done online with Telemedicine but, of course, now the last couple years, we’re seeing an explosion in Telemedicine. A lot of the research centers will have an in-house geneticist. But, of course, in private practice, that’s not the same situation. Any thoughts about genetic counseling and how it’s been affected by Telemedicine?

DR SAAD: Yeah. Well it’s still a huge issue in Canada. If we have to get a genetic counselor involved, we’re talking about months and months of delays. And so in that case, we need to do it way before there’s an indication to actually act on the findings. We’re trying to take — we obviously try to get germline and somatic simultaneously. We try to streamline so that we’re not overburdening the genetic counselor where we find only about 1 in 10 germline mutations in metastatic disease. So I think we’re coming to some compromise. But I totally agree with Johann, the earlier we can test, not only does it help us to determine or know ahead of time what we can give them, but it also modulates a little bit of the follow-up because these patients don’t do as well on the therapies that we offer them when they’re metastatic or metastatic castration resistant. And so following those patients maybe closely helps us to determine resistance earlier and get them on the proper drug as early as possible.

DR LOVE: So, Johann, in the chat room here, Abdul is asking do you repeat biopsy or genomic evaluation with every progression? And that kind of ties a little bit into my point #2 which is role of liquid biopsy. If you could comment on that, Johann. And in particular, if you could comment on one of the striking things that you said in your presentation the I have not heard before which is that you can miss, you know, false negative genomic evaluations, particularly with plasma. And what really scared me is that you can get a false negative in the people who respond the most which is people with homologous deletion. Can you comment on that a little bit?

PROF DE BONO: Yes. So first of all, BRCA loss can be caused by either a mutation of one allele and het loss of the other allele or it can be a result of deletion of both alleles. Now these tumors become resistant to PARP inhibition by getting a second mutation in the mutated BRCA gene, putting that BRCA back in frame and restoring BRCA function. And that restoration of BRCA function usually happens within about 6 months to maybe 18 months of treatment with PARP inhibition. And the cancer finds it relatively easy to restore BRCA function if you’ve got a mutation. Now if you’ve deleted both BRCA genes, the cancer finds it much harder to restore BRCA function. And the reason for it — or what that results in is that the patients with homozygous deletion, i.e. deletion of both BRCA alleles, in our publication in Cancer Discovery, clearly have a much, much longer time on drug as single agent olaparib, you know, 3 or 4-fold longer compared to the patients that have the BRCA mutation and second hit by heterozygous loss. So those are what we call the super responders. If you have a BRCA2 homozygous deletion, you have deleted both alleles, you are a super responder.

Now the problem with all of these tests is that when you’re sequencing either a plasma sample of a null biopsy, there is a lot of DNA in there that’s normal DNA, non-cancer cell DNA. And that makes the assay quite difficult to call because normal cells, the BRCA sequencing is normal. And if you’re trying to look at that smoothie or soup, whatever you want to call it, where you have a little bit of carrot in a sea of potato or whatever, it might be quite hard to detect loss of something in that carrot, if you get the gist of it. And actually, this is the issue, particularly in null biopsies, we’ve got degraded DNA and in prostate cancer, unlike other cancers, you have a high stromal content, non-cancer cell content. And this is a bigger issue in plasma. And in plasma DNA, it really is a nightmare. In fact, many of the plasma assays are not approved for detecting homozygous deletion. So I should say, there is hope on the horizon. There are other assays developing from plasma, particularly things like low-pass whole genome, that I think are going to get better at detecting BRCA2 homozygous deletion, PALB2 homozygous deletion, ATM homozygous deletion, because they’ll sequence the whole genome. So these are things that we have to be aware of. And as always, it’s not that it’s a bad tool, it’s just how we use. It’s the user’s, I guess, application of the tool that we have to be aware of.

DR LOVE: So, Fred, I love all the translational — I always have since I first heard the word synthetic lethality but, of course, I’m not sure I understand 90% of it. But I’m a practical person. If you tell me there’s a super responder that I’m going to miss, I think that kind of gets my attention. And it leads into the next, really next 3 topics which is really the use of genomic evaluation and deciding whether or not to use a PARP inhibitor. Obviously, it depends a lot on the individual situation, where things are in general. I will point out, Fred, that this ATHENA study of rucaparib in ovary also showed a benefit in HR proficient just like they saw in ovary with niraparib and just like you saw in PROpel. And we’ll talk about that when we get to PROpel and MAGNITUDE. But just, in general, what are the mutations, whether they’re germline or somatic, Fred, that at this point, you believe we have evidence that PARP inhibitors work? Obviously, BRCA is one thing, but there are a bunch of other biomarkers, ATM, PALB2, et cetera. Any comments, Fred?

DR SAAD: When you look at these studies, they’re all such small numbers outside of the BRCA mutated patients, including ATM which are a larger number and others. And we’re seeing a mixed bag of responses and, obviously, the expert is Johann, but I think we’re going to need to start doing some multiple analyses of different studies to start getting numbers that are reasonable to make sense of it. Because almost every one of these other deletions or mutations have some sort of response. And it’s not like there’s no response in these other mutations, it’s just that the numbers are so small that the confidence levels are wide and it’s hard to make sense. And that’s why, unfortunately, for many countries, they’ve only approved PARP inhibition in prostate for BRCA mutated patients. Some have accepted ATM, but most haven’t gone the US route of expecting the whole — of accepting the whole panel that was included in PROfound which is unfortunate because I think some patients do respond. And I think Johann has some really nice experiences in patients that were not BRCA mutated that had spectacular responses. And it’s always very, very disappointing when you can’t even get access to at least try and see what kind of response you would get after 3 months of exposure. Because it was definitely beyond just BRCA that were responding in PROfound.

DR LOVE: And that brings us also to this issue of these scores that come up which I’m not sure exactly what they’re measuring. I hear them described as scarring or, you know, mutation panels. LOH is one you hear that is used in ovary a lot. Johann, what about these scores? You have a patient that had a fantastic response to platinum with a high LOH score and you couldn’t find any mutations. What are these scores actually looking at? How reliable are they? How predictive are they?

PROF DE BONO: They’re really measuring genomic instability. When you have a DNA repair defect, your genome is much more genomically unstable. And they’re looking at things like large scale transition, so the number of areas in the genome that have major increases or decrease in copy number over the baseline. I think the main message with these is that in some cancers like ovarian and breast cancer, these scores seem to be working fairly well although they’re not perfect by any stretch of the imagination. They don’t seem to be working very well in prostate cancer, and I don’t think we’ve figured out quite why that is yet. It may be the same issue that there is a dilution of the tumor DNA genomics by the stroma much more than the other cancers. But I think the main simple message today is that these are adjunct readouts you can look at in your report. They’re given as a free extra generally. But I think in prostate cancer, they’re not particularly good, especially if it’s a negative result. The negative results are not really very meaningful.

DR LOVE: So, of course, my #1 priority in terms of the top 10 list and the #1 question in prostate cancer is what’s the implication of the MAGNITUDE and PROpel data to clinical practice? And, Fred, in your talk, of course, you went through all the data. We’re not going to go through all the slides, at least initially. But maybe you can comment a little bit about the thinking behind it. When I first heard about it, I was thinking 1+1=2, but actually it sounds like maybe their hope is that 1+1 might equal more than 2. Can you comment a little bit, Fred, about the background to this work?

DR SAAD: Yeah. So the background is some preclinical evidence that suggested that PARP inhibition might actually increase the efficacy of a novel hormonal agent like abiraterone, enzalutamide or all the others that we have around through some AR-dependent transcription. But I think where a lot of the data is, is that the novel hormonal in itself might induce an HRR-type environment deficiency that would increase susceptibility of PARP inhibition where in normal circumstances, the cell might not be very responsive to PARP inhibition. And so this preclinically, looked like it was doing something to make both of the drugs more effective when combined. And we actually did a Phase II study that addressed this taking an all-comer population that had progressed after docetaxel. And it turned out that patients with or without detectable mutations seemed to be getting as much benefit, at least in terms of radiographic progression, in a randomized Phase II study of about 140 patients getting abiraterone plus olaparib versus abiraterone alone in the post-chemotherapy setting.

DR LOVE: So another relevant practical question, Fred, is why was the decision made to combine olaparib with just abi as opposed to anti-androgen, enzalutamide, for example.

DR SAAD: Yeah. So because there was — based on the Phase II study and some other studies looking at drug-drug interaction through the pathways, it looked like abiraterone was probably the safest combination, at least for olaparib. And it allowed full dose olaparib to be combined with full dose abiraterone. There was no need to reduce the dose of either drug to make it stable and not induce PKs that were strikingly different than each drug used alone. So that was reassuring because at least in patients that are not mutated, we felt that the highest dose of PARP inhibition might lead to better outcomes. And the highest dose of abiraterone was essential because you wanted the standard of care arm to be really the standard of care that is being used in metastatic CRPC. And this is —

DR LOVE: You were telling me there was a talazoparib trial out there that hopefully is going to report fairly soon, TALAPRO-2, kind of similar to PROpel and MAGNITUDE, first line CRPC if I understood you correctly. But that’s combining it with enzalutamide, Fred?

DR SAAD: That’s combining it with enzalutamide. And I think in terms of efficacy, they’re, I would say, equivalent drugs, abi or enza. The choice was here to use enza. But combining it with enzalutamide, the talazoparib dose had to be reduced because there were more adverse events when using the full dose, 1 mg of talazoparib in combination with full dose enzalutamide. This was — it’s not unique to enzalutamide. It was seen with apalutamide in combination with niraparib where the adverse event rates were really too high. And I actually led the Phase I study that the adverse event rates were too high when combining apalutamide, even at reduced dose of niraparib. So that’s why the niraparib was combined with abiraterone which was more tolerable for MAGNITUDE.

DR LOVE: Really interesting. And, Johann, this is a slide that you discussed in your talk about your thoughts about sort of the biologic underpinnings of the PROpel and MAGNITUDE strategy. Do you want to comment on that?

PROF DE BONO: I think first of all, the first issue is that when you block androgen receptor, PSA goes down. And certainly, in our New England paper that we published, Mateo et al, in 2015, we did not really see PSA falls in prostate cancers that did not have a DDR defect, a DNA repair defect. Therefore, at this juncture, I really am not convinced that PARP inhibition blocks AR signaling in patients.

The second thing is that if AR blockade by a drug like abiraterone did cause what has been described as BRCAness, that HRD defect, then you would see after abiraterone treatment a major increase in sensitivity and response rate. So, you know, if abiraterone’s response rate is, for example, usually 40, 50%, if the abiraterone causes BRCAness when you add the PARP inhibitor, you’d expect that response rate to go to at least 80%, let’s say, for example. We don’t see that with these trials. Therefore, I think in my opinion, that is unlikely to be the case with the combination of abiraterone and olaparib. And the challenge with this combination is that if the patient is responding to that combination, you don’t know if they’re responding to just the abiraterone or getting benefit from both because abiraterone, of course, can work very well.

DR LOVE: Yeah, that’s a really good point. Of course, they’re going to be on it for a long time. Let’s talk about the 2 big trials. And, Fred, here is the PROpel design. Can you talk a little bit about how the trial was designed, how it was executed? And maybe just, without going through the slides, just summarize your take on the main findings.

DR SAAD: So the design was relatively simple. This is an all-comer population in patients getting first line mCRPC treatment, but were allowed to have had docetaxel in the hormone sensitive setting. They could not have had prior abiraterone, obviously, but were allowed to have visceral metastases. So it was really an all-comer population. And it gave us the opportunity to — and when patients were screened, obviously, we got tissue whenever it was available and we got ctDNA to analyze to be able to try to determine the status of whether they were HRR mutated or not. And what we hope is that this is going to be the opportunity to actually look for biomarkers that might not be known and might help us for future studies. And by powering the study as it was with about 800 patients, we had calculated based on what we know that we would have enough patients that were mutated to be able to have some analysis of how these patients responded.

And what the study showed was that we were able to get about an 8.2-month improvement in rPFS with the combination over abiraterone alone. And this is identical, basically identical to the 8.2 months that was seen of abiraterone compared to placebo in the pivotal Phase III study of COUGAR-302. So that was interesting. And I agree with Johann that the evidence is not 100%, but these patients did respond for a longer amount of time in a randomized fashion to the combination than abiraterone alone.

And I just want to recall that all the drugs that we’re using right now in mCRPC were compared to a placebo or a non-life-prolonging agent. And there are some subgroups that really looked like they were benefiting, patients under 65. All the subgroups were actually benefiting if you want to say being on the side that we would want them to be on, but some were actually doing much better than you would expect like visceral metastatic disease, post-docetaxel therapy that these patients progress very quickly. And we ended up having a 39% reduction in progression which turned out to be almost 14 months improvement in delaying progression. And what we hoped to see was actually seen, that patients that were not mutated were actually getting a 24% reduction in progression. Obviously, the ones that were mutated were getting the largest amount of benefit where we haven’t reached the median rPFS. But you see that the patients on abi alone are progressing faster than the average patient with mCRPC. And this includes all mutated patients that were detected without any enrichment for BRCA mutated patients in this study. So I think going in the right direction. I agree with Johann. I think we need to continue to look at the maturity of this data. But clearly, with an 800-patient study with a control arm that’s very effective and considered one of the standards of care, I think we are doing better.

DR LOVE: So, Johann, of course, the most controversial issue, and you see this at the bottom of this curve, is the response or benefit based on mutational status. It kind of reminds me of the PRIMA study in ovary. That was in niraparib, incidentally. But in any event, here, you see with the mutant disease a hazard rate, as Fred was saying, of 0.5. But then in the non-HRR patients, you have a hazard rate of 0.76. That is statistically significant. So, Johann, what are your thoughts about this finding, theoretically, but also clinically? If you could, in what patients would you use this strategy? And we’ll talk about niraparib in a second. But in terms of this dataset, putting aside reimbursement and cost issues, would you use it in this situation for, let’s say, BRCA germline? Would you use it in somebody who had nothing?

PROF DE BONO: I think first of all, it’s quite apparent that the patients that benefit most are the patients with the DNA repair defects. That’s quite clear. I think secondly, it is incontrovertible for me that the non-HRR subset will definitely have some false negative cancers. I think one can, I guess, guess what that number would be. It’s impossible to know for certain. But there will be false negatives in a non-HRR group. But it’s going to be particularly interesting to see what the FDA says. And remember, clearly this trial was not powered to look at these subgroups. All we can say overall is that we need to go and see what the survival data will look like. And what I want to see particularly is what’s the response rates of the 2 subsets. And I think, you know, we’ve seen that data. It’s not that different for abi alone versus abi and PARP, even in the non-HRR. So it just worries me a bit that perhaps we do need a bit more data to be sure this is working beyond HRR defective. So I think there are patients for sure that are not BRCA that are benefiting, for example, PALB2 mutated tumors, I guess our classic subgroup, ATM loss for definite. But I do think we need more data.

DR LOVE: Fred, I tell people, I always quote Johann as calling PALB2, BRCA3. People like that. But anyhow. The other one he does is p53. What do you call it? The guardian of the genome, Johann?

PROF DE BONO: Yes, sir.

DR LOVE: I like those understandable type of terms. What are your thoughts, Fred, about how you would like to apply it? Again, putting aside regulatory issues. What are your thoughts in particular, obviously, about who would you apply in to, both in terms of people with mutations outside of BRCA, people with high LOH, people with nothing? We did a satellite a couple hours after you presented the data, so I know what you were thinking then. I’m guessing you’re thinking the same thing now. But what are your thoughts?

DR SAAD: Yeah. So Johann led the PROfound study. So we know that PARP inhibition, and olaparib in particular, works in patients who have failed abi or enza or some of the other novel hormonal agents. So in patients that are mutated, for me, it’s clear that these patients need to be treated early because even within PROfound, it looked like patients who got olaparib earlier did better than the ones that got it subsequent to their progression. And I think this adds to that information because the rPFS is really quite outstanding in those patients. And those patients will need it eventually if they are mutated. So here, we have an opportunity of maybe making a big difference upfront.

Now in the nonmutated patients, very much like triplet therapy for chemotherapy, I don’t think every patient is going to be getting a combination. Absolutely not. That would be ridiculous. That would be, I don’t think, personalizing care. But we have patient populations that worry us, visceral metastatic disease, patients that biologically look like they might not be very good responders to abi. The post-chemotherapy patients that got it in hormone sensitive don’t do generally very well when they get onto a novel hormonal agent. That’s been our experience and that’s what we saw here in PROpel. The younger patient, for some reason, we’re trying to explore why they did so much better. So I think there are some biological parameters, some clinical parameters that we’re going to start to use, and mutations are going to be one of them when we can get access to them. And as Johann eloquently said, sometimes we have issues even detecting mutations, getting access to these testing mechanisms. And unfortunately, we might be missing some of the best responders based on the limitations of what we have. So if we’ve got other clinical parameters, I think I would be considering the combination because this is our opportunity to do better in these patients in the first line setting.

DR LOVE: So, Johann, I’d like you to comment on some of the tolerability toxicity data from the PROpel study and also the issue of MDS and AML. But there’s a question in the chat room actually from a doc that I did a case session on the day, Dr Bank, who is asking what’s the role of PARP inhibitors for earlier stage prostate cancer, even in the neoadjuvant setting? And maybe you could also, Fred has made this point a number of times when I worked with him over the last couple of years, about the idea that the earlier use of PARP inhibitors will end up with greater benefit. And kind of he cited your PROfound data in terms of patients who got PARP on — who were initially randomized to not get PARP and then got it later and how that affected — the fact that it did affect survival, suggesting it maybe earlier. So, Johann, maybe you can comment on what you saw there with the crossover with PROfound and Dr Bank’s question about where are we heading. Hormone sensitive prostate cancer is an obvious one. But what are your thoughts?

PROF DE BONO: Well I think there’s clearly a commercial driver to move these drugs earlier in this space. But I do think we have to be cautious. Our responsibility as physicians is to do no harm. And I do worry that starting these drugs in patients that have a very long lifespan, sometimes 5, 10 years or more, does carry risks. These drugs are DNA damaging agents. They do come with some risk of secondary myelodysplasia and leukemia, although small, and the drugs, I guess these drug’s primary toxicity is anemia, hematological toxicity. So I think we have to be thoughtful about how we use these drugs.

But on the other hand, PARP inhibitors are much better tolerated than chemotherapy. And in men with aggressive disease, I think that particularly there’s a family history which is where I’ve really been quite keen to use these agents, in, say, a BRCA mutation. There is no doubt there’s a large benefit. So I would have no concern in, say, a BRCA germline patient of trying to avoid castration, and I have done this in patients before, and give a PARP inhibitor on its own, without hormonal therapy. And I’ve done that successfully for some period of time. But there, I know the benefits are really quite large and I am certain of the likelihood of making a difference for that patient. Clearly, when you start combining these drugs, for example, abiraterone, you really don’t know is it the PARP inhibitor that’s working. And that’s kind of what concerns me. You’re giving a drug where you don’t know what benefit that it’s giving in the background of a drug like abiraterone that actually can work very well. I’ve had patients on abiraterone for over a decade, you know, treatment with abiraterone in this space in my clinic. So are we really going to give a PARP inhibitor for a decade? Maybe. Maybe we are. But that definitely does come with some risk.

DR LOVE: Of course, that’s a big debate now in breast cancer because they’re using it adjuvantly in breast cancer right now. So we’ll see how it lands in terms of the ultimate duration. Fred, maybe you can comment a little bit also on the MAGNITUDE study, a Phase III trial of niraparib. Here, they stratified it. Really, it had 2 different sections. One that was HRR negative. And there, they bailed out. It looked like it didn’t work, interestingly enough. I just mentioned it seemed to work. It’s indicated in HR proficient ovarian cancer. But you see the flip side here. But it looked, in terms of the positive mutation patients, pretty similar, of course, difficulty trying to compare between 2 trials, but hazard rate of 0.53 in BRCA. Not that much different from PROpel. Any thoughts about this study and why, at least it looks like, it lines up with the mutated, but not the unmutated, Fred?

DR SAAD: Yeah. So this study has a few differences than PROpel. Now I know Johann and I have talked about this before, whether dosing makes a difference. I think in ovarian, reducing the dose in proficient patients actually looked like it was less effective. And here, the dose had to be reduced in combination with abiraterone. It couldn’t be used at the same dose as was used in monotherapy for patients with prostate cancer. So maybe dosing makes a difference, especially in patients that don’t harbor mutations or don’t harbor BRCA mutations because what we saw here was that the BRCA mutated patients had a 0.53 hazard ratio in specifically BRCA mutated. But that went up to 0.73 if you added the non-BRCA mutated patients. And when you looked at the forest plots of the different subgroups, the non-BRCA patients really don’t look like they’re benefiting a lot from the niraparib combination. So I think there are differences. I think niraparib, like other PARP inhibitors, are extremely effective against BRCA. I think where it gets more complicated is when we’re talking about non-BRCA mutated patients or patients that don’t harbor mutations at all where there’s still a lot of work to be done to show why exactly it might be working. But dosing might make a difference. The combination might make a difference. The timing might make a difference because, surprisingly, if patients were just exposed for a few months to abiraterone, they seemed to lose the benefit in the study. And so there might — that timing might make a difference. It’s all theoretical. We’re still trying to work things out.

DR LOVE: So, Johann, a couple of quick questions from the chat room. One, when you’re getting a biopsy from bone, is that a problem in terms of genomic assays like NGS? Second follow-up question. So that one is from Richard. Saranya has a patient who is progressing on abiraterone for metastatic CRPC, has a CDK12 mutation. “Should I start olaparib? Should I start immunotherapy? Should I start enza alone?” And then Dr McKenna wants to know, she thinks that she sees better responses to PARP with people who have smaller tumor mass. Do you think that’s the case too? So this is a memory challenge, Johann. Can you remember all 3 questions?

PROF DE BONO: Yes, sir. I think bone biopsies can work and genomics can be acquired easily from bone biopsies, but you will have to make sure that your lab that you work with can do that. Certainly, our laboratory does the genomics from bone biopsies without a problem.

With regards to CDK12, the data for CDK12 on olaparib is not impressive, but there are some responses in PROfound to olaparib in patients with a CDK12 alteration. Work led by Wu et al that I’m a co-author on has shown that if you have biallelic CDK12 alterations, and these are usually mutations, that these tumors have a very high genomic alteration rate, usually rearrangements, very high neoantigen loads, high tumor mutation burden, largely rearrangements, and can respond very well to immunotherapy. So I guess if I had a CDK12 biallelic mutation, normally that would have to be biallelic, I would love to get on a trial with an immunotherapy type agent, a PD-1/PD-L1 targeting drug. And that would probably be my preference. But if it’s a monoallelic mutation of CDK12, that would be less attractive and you have to look at all the other possibilities. If you are not sure that it’s monoallelic or biallelic, then you can try and get the lab to do low-pass whole genome and you can see what’s called a tandem duplication phenotype or genotype with lots and lots of spikes in the DNA panel. So that’s the second question.

The third question was a patient — I’m just trying to think what the third question was. Someone progressing —

DR LOVE: Smaller tumors. Smaller tumors respond better.

PROF DE BONO: Smaller tumors. Yeah, I think overall, if you think about population statistics, when you have smaller tumors, the likelihood of getting cells emerging to resistance are smaller. So whenever you give drugs earlier, you are less likely to get resistant clones. And obviously, it’s the length of a piece of string, if you have the same hazard ratio early or late, you get a more significant benefit when you give these drugs earlier. But clearly, getting drugs in when there’s less disease, you can get better responses. And this is particularly more profound for, I think, immunotherapy. Actually, that is a big issue for the immunotherapy field although today, to-date outside MMR and CDK12, PD-1/PD-L1 immunotherapy targeting has not really been very effective in prostate cancer.

DR SAAD: Neil, the only thing I would add — Neil, the only thing I would strongly recommend against is going from abi to enza in a mutated patient.

DR LOVE: Absolutely.

DR SAAD: We showed some data from PROfound where patients, regardless of whether you went from abi to enza or enza to abi, in the mutated patient, really there was no difference in rPFS. The difference was really in the patients who got upfront olaparib. And in survival, it was just about 10 days difference in survival between abi to enza or enza to abi. So really, that is to be discouraged in somebody that you’ve detected an HRR mutation. You should go straight to some other more effective line of therapy like Johann has mentioned.

DR LOVE: So one of my top 10 things was in relapsed disease, and we’ll get to that in a second. But just to finish out on where things might be heading, Fred, this is a slide you showed in your presentation about what to expect. There’s the TALAPRO-3 trial that we were talking about there in the middle with talazoparib and enzalutamide. But you have a really nice graphic here showing kind of the natural history and where the trials are being done. Can you comment a little bit about what you expect for the next 3 to 5 years in terms of as we see these trials mature, Fred? Where do you think we’re going to land?

DR SAAD: Well hopefully, we’re going to continue to learn of who actually benefits. But I totally agree with Johann. If we’re going to start giving it earlier, and I’m a proponent of early therapy, but only in patients we’re sure are going to benefit or have good evidence. So in the hormone sensitive setting, I would be very, very much against giving it to an all-comer population because those patients can respond for a long time. So the studies are actually targeting patients with mutations that are detectable. And even there, we need to get a sense for this because those patients are going to be exposed for much longer than the mCRPC state, even if they’re deficient or mutated. But in the mCRPC state, the average patient is really only responding for about a year to 15 months, even though abiraterone, I agree, is very effective. And that is going to be the question. In a super responder, should we pull back after a couple of years? And I’ve actually stopped abiraterone in patients who have been in complete response for several years and we said we’ll take a break. Because I agree, 10 years of exposure is a long time for any drug.

DR LOVE: And you both have cases in your presentations of patients with long responses. And we hear these all the time, although I’m not saying that that’s the usual, but it seems like it definitely happens.

So just going back to our top 10 list here, I want to come back to you, Johann. We talked a little bit about MDS and AML. It’s interesting. This doc who asked the question about that, he presented an ovary case of a patient who got AML. And actually, the patient never got a PARP inhibitor. They just got chemotherapy. So chemo can cause that as well. But getting back to the more common acute toxicities, Johann. I’m just kind of curious if you have any special approaches clinically. Of course, we hear a lot about GI effects, particularly in the beginning, and obviously, cyteopenias. Any sort of clinical pearls about getting people through longer term therapy?

PROF DE BONO: Well I think these agents are generally reasonably well tolerated. Anemia is the biggest issue. And this can become a major issue. I have proposed that instead of dose reducing these patients if we see major anemia, we give them drug holidays and maintain a 300 mg dose, but give some time off drug every month, some time for the bone marrow to recover. And that can work well. For example, in some patients, I’ve given the drug 2 weeks on, 2 weeks off. And that can continue for some years although that was a single agent, not with abiraterone. And the patient continued to respond well, it was a BRCA patient, for some time.

And I want to remind everybody about platinum. If the patient cannot afford to buy olaparib which is clearly an expensive new agent, the platinums, particularly carboplatin, can work very, very well for these BRCA altered and DNA repair defective tumors and should not be forgotten about really. And these drugs, although not as well tolerated as PARP, do benefit these patients, I’m sure.

DR LOVE: So, Fred, what about the issue of PARP rechallenge in patients who get it earlier? And who knows? Maybe as time goes on, we’ll see, again, now breast cancer has popped through with adjuvant therapy and they’re asking this question. There’s already a study from the SOLO1, olaparib, of ovary patients of repeat olaparib and relapse after they’ve stopped it and they see responses. So I’m just kind of curious what we know at this point, Fred, particularly in the relapsed setting, about, Johann talked a little bit about some of the mechanisms, but also the issue of PARP inhibitor rechallenge in somebody who has had it previously. Any thoughts about that, Fred?

DR SAAD: I really don’t have much thought about that. We’ve been giving it in such advanced disease that that would be a great problem to have, such a great responder that we have the opportunity to rechallenge. I don’t think we’re anywhere close in prostate cancer to having that good problem of being able to take them off because, unfortunately, most of the time, they’re started so late that they go through them and they progress and, unfortunately, they die. Even in PROfound, it was only about 7.5 months to progression in patients. So that’s why I think we need to see in appropriate patients if we can bring them a little bit earlier where their response might be longer. But the reality is we don’t have that problem yet. Johann might have some great responders where he’s stopped and rechallenged, but I definitely am not anywhere close to that.

Cases from the Faculty

DR LOVE: So I want to finish out with a couple of cases. Each of you presented 3 cases in your talk. I just want you to comment on one of them. And this is the patient we’ve been referring to, Johann, who had such a great response to carboplatin. Before you mention or comment on the case though, can you just comment about, again, putting aside reimbursement issues and regulatory issues, the use of PARP inhibitors nowadays and, again, rucaparib and olaparib, at least in the United States, in relapsed disease? In what situation do you typically use it? And in particular, if you could, in a patient who had run out of options, would you use it in a patient without any mutation, HR proficient, Johann?

PROF DE BONO: So let me make very clear here from this case. This case was a young man and he had a 4-year remission on platinum despite having previously had abi, enza, doce, cabazitaxel. So even in men who are really heavily pretreated with heavy burden of disease, what we’re seeing here is a complete response to carboplatin single agent, resolution of his bone disease, complete resolution of his liver disease. So if the biology is in favor of sensitivity to treatment, you can really have super response that is just really durable and quite impressive. And I think what is key here for this case for me, because I treated this man with platinum without knowing his genomics, is 2 things. One is that he had a strong family history. And two is the location of his disease as liver metastases that looks like neuroendocrine disease. And when I see that, young man, strong family, or relatively young man in his early 40’s, with that kind of liver miliary metastases, my brain goes straight to thinking this is a BRCA-like tumor. And that’s why I gave him platinum. And if I could have, and this guy wasn’t really eligible for the olaparib trial sadly because of his liver dysfunction, I would have given him PARP. Just to remind everybody, carboplatin is largely excreted by the kidneys and in a man with liver dysfunction, it’s an easy drug to give.

DR LOVE: So I want to close out with this patient. I think, is this the one who was actually on — let me see here. Yeah, he was on the PROpel study. Right. So you actually don’t know what he got because it was blinded. But I thought it would be interesting to talk about this. And we also have a question. Maybe, Johann, you can give some thought to this after we talk about this case, about where lutetium is going to fit in, and what about combining lutetium with PARP? Of course, now approved and so much excitement. But, Fred, maybe you can tell us a little bit about this 61-year-old man. What happened with him?

DR SAAD: So this patient is still on the PROpel study. And to cut to the chase, he is still in complete remission. There’s no evidence of disease that we can detect. And so this patient was diagnosed with aggressive disease. He had gotten chemotherapy and hormonal therapy and then progressed and was sent in for possibility of a study for mCRPC in the first line after having been exposed to docetaxel. So he was put on the PROpel study in February of 2019. Very early on, in the first 3 months, he had fatigue, some digestive issues, and a decline in hemoglobin that went from 13.3 to 10.5. And then the hemoglobin restabilized and went up after several months. And his PSA became undetectable after 6 months. And we had clearly complete remission of all measurable lesions and no change on bone scan. And he’s seen on a monthly basis and it’s almost become a social event to have him come in. He comes in with a bottle of port every couple of months because he goes to Portugal on a regular basis. So my wine cellar is full of port. But this is a patient that obviously is still blinded, but based on what happened in the first 3 months, it’s not something I would routinely see with abiraterone alone. So I strongly suspect that he is on the combination. And it’s been now 3 years and longer that he’s been in complete response in a patient post-docetaxel. So really not the average kind of response that I see in patients who had aggressive disease treated with docetaxel and ADT and then rapidly became mCRPC. So this is — and the tolerability, because it’s an earlier disease setting in terms of bone marrow tolerability, the adverse events that we saw in PROpel with 15% Grade 3 anemia that was easily managed through supportive care. So that was the most common adverse event.

DR LOVE: So in a second, I’m going to ask for a final comment from Johann, specifically about how you think maybe lutetium is going to affect the algorithm of using PARP, what you think about combining lutetium with PARP, what we know about that. But before you do that, audience, we’re just trying another way to figure out whether we’re accomplishing anything here. I want you to, if you could, please answer this question. I want to know how prepared you feel right now to administer a PARP inhibitor. Consider the whole issue of PARP inhibitors in men with prostate cancer. I want to know to what extent you feel prepared to deal with that situation now compared to before you did this webinar. If it’s about the same, you can put a 5. If you feel you’re more confused now, you can put somewhere between 1 and 4. If you feel you know more, you an put from 6 to 10. I’m just kind of curious whether we’ve accomplished anything here in our ongoing effort. But we’ll finish out, Johann, with your thoughts about where lutetium and PARP is going to end.

PROF DE BONO: Well I think first of all, lutetium will be widely used, probably will replace radium. I think radium, in my opinion, is likely to be a drug that’s phased out and replaced by lutetium although we’ll have to see, of course. I think the data combining PARP and lutetium will be very interesting to see. Obviously, we haven’t got much data yet because PARP is already a sensitizer. Although I do worry that the 2 together may be quite bone marrow toxic. Clearly, let’s watch that space and I look forward to seeing that data.

DR LOVE: All right. Well thank you so much, both of you, for working with us tonight. We learned a lot, as always. So much more to think about in terms of this topic, not just in prostate cancer, but as we’ve been alluding to in a number of other cancers. Come on back on next Tuesday. We’re going to start our CML series and explore that with Dr Cortes. Thanks so much to the audience for participating tonight. Be safe, stay well and have a great night. Thanks so much, Johann.

PROF DE BONO: Thank you.

DR LOVE: Thanks, Fred.

DR SAAD: Thank you.

DR LOVE: Take care.