NS: Hi everyone, I’m Neal Shore. It’s a great pleasure to work today for ecancer and my colleagues, everyone will introduce themselves. I’m the Medical Director of the Carolina Urologic Research Center and the Chief Medical Officer of Urology and Surgical Oncology for GenesisCare. We’re here at ESMO 2022 in Paris and it’s another fantastic event. We’re going to have a really spirited discussion today; I think we’re mostly going to focus on the burgeoning understanding of PARP inhibitors in advanced prostate cancer – the why, the when, the how, how to incorporate into your practice, monotherapy versus combination therapy. So great, and thank you so much for joining us today. Let me start on the far left with our colleague from Italy.
PR: I’m Pasquale Rescigno and I’m a medical oncologist and I’m Director of the Translational and Clinical Trial Group at Candiolo Cancer Centre in Turin.
FS: Hi, I’m Friederike Schlürmann and I’m a medical oncologist in France in the University Hospital of Brest in the western part of this beautiful country.
NC: My name is Noel Clarke, I’m a urological surgeon and Professor of Urological Oncology at the Christie Hospital in Manchester.
NS: Fantastic. So one of the things that’s great about ecancer programmes in general, and I’ve had the privilege to be involved in, is this is multidisciplinary, multi-speciality. We’re learning so much to do better things for our patients with advanced prostate cancer compared to 2003 where it was purely nihilism at its extreme. Here we are in 2022 and we arguably have 12 different novel mechanisms of action, multiple, double-digit, approved therapies that are life-prolonging but we’re going to focus today on PARP inhibition. Noel, it would be great if you could lead us off a little bit on some of the really pioneering work that you’ve done. You did the Study 08, which was a great presentation, a big phase II study at ASCO prior to the pandemic. It’s great that we’re all back in person now seeing each other, the enjoyment of being together, but now talking about a wonderful phase III trial that you were the co-PI on. You published your work in The New England Journal of Medicine evidence and now another presentation here in ESMO on PARP inhibition and the biomarkers and how to think about it. So maybe lead us off.
NC: Okay. So the whole of the combination PARP story owes a lot to Pasquale who did the original TOPARP-A, TOPARP-B study. That led us in to the notion that BRCA was important, that homologous recombinant repair was important and that PARP inhibitors were beneficial in patients who were mutated when it was given as monotherapy in late stage CRPC. There was a body of evidence which showed us that the use of PARP in combination with a novel hormone agent, particularly abiraterone, would open up the androgenic transcription and somehow render a previously resistant cell, or resistant to DRD, HRR inhibition, PARP inhibition, would render it sensitive to that drug approach. So we designed a study which was, again, in late stage heavily pre-treated patients, a large scale phase II study which we called Study 08, which was the one which was presented at ASCO and published subsequently in Lancet.
To our pleasant surprise… the way we designed the study was to take all comers, in other words, not to check for homologous recombinant repair as a randomisation factor. The patients were just treated either with standard abiraterone with olaparib or abiraterone and placebo. Then we followed them, we collected tissue, we collected circulating markers and we analysed those afterwards. So it was a prospective collection but not part of the pre-specified analysis looking at radiological progression free survival.
To our pleasant surprise we found that it was a positive trial. There was an improvement in radiological progression free survival with the combination and that amounted to about 8 months improved survival. When we looked at the characteristics of the HRR mutations what we found was that there was a signal which showed that the combination of abiraterone and olaparib was working in those patients who weren’t HRR mutated using standard methodologies. That led us to go on to design the PROpel study. Again, what we aimed to do was to keep it relatively generic and simple and take a population of patients at first line failure, in other words, primary castrate resistant metastatic prostate cancer. We were careful to look at the cardiovascular profile of these patients, we had some cardiovascular toxicity in Study 08, and we had a stipulation in the protocol that patients weren’t allowed to have any kind of novel hormone agent in the lead-in to having the therapy.
The patients we treated, it was a large-scale randomisation, just under 400 patients per arm, and the primary endpoint was radiological progression free survival. It met its primary endpoint. We also saw improvements in time to second therapy and treatment failure. The overall survival in the first interim analysis, which we presented at GU ASCO and subsequently published, was immature. At this meeting what we have done is present the data from the second interim analysis and we’ve added in, in parts of the presentation that was given today at ESMO but also in one of the poster presentations, the genomic data associated with that. What it showed and has shown so far is that everything seems to be going in the same direction. The radiological progression free is still significant, quite substantially; the secondary endpoints of time to treatment failure were significant; the overall survival, the actuarial curves are still separating but it’s still immature, it’s about 40% mature at the moment.
So our conclusion is, yes, there is a signal. It does seem to be much stronger in BRCA2 type and BRCA mutated patients but it’s still present in the non-HRR mutated population. We had a much more detailed tissue and circulating marker analysis in PROpel than we did in Study 08.
NS: That’s a great overview and so it’s really interesting. Historically we’ve been rather monolithic in prostate cancer compared to other tumour streams – one drug after another after another. So congratulations to you for coming up with Study 08 which led to PROpel. And prior to all of that we had PROfound and, Pasquale, I know you did a lot of work on the TOPARP-A and TOPARP-B trials with Joaquin Mateo. So congratulations on that. Can you just help us understand a little bit about that work and how that work was seminal for going to the PROfound trial and now PROpel.
PR: Absolutely. I think that that work was important because that was the first evidence that a PARP inhibitor was working in selected people. In patients with metastatic CRPC they were progressing on docetaxel and on a hormonal treatment, it could have been abiraterone or enzalutamide. But, more importantly, what really TOPARP-B taught us is that if you want a really good understanding of what HRR metastatic CRPC patients do on olaparib, you really need to test the tissue samples of these patients. All patients that were recruited in that trial were patients that had a fresh biopsy, that was mandatory before starting. We had something like 800+ biopsies and when you interrogate the tissue, I know that it’s not easy to get – you have to ask the patient, you need to find a biopsiable site – but when you do the analysis of the tissue the rate of failure of sequencing is quite small. You have 87%, that was the percentage of success rate within the TOPARP-B.
The kind of information that you can get from the tissue, for example, is extremely important because now we know there has been also a follow-up analysis from the TOPARP-B, first name Suzanne Carreira, now we know that the patients that have the best response on PARP inhibitor as a monotherapy are the patients that have homogeneous loss of BRCA2. That’s something that right now you can know if you have the tissue of that patient and the tissue needs to be a recent one because that reduces the risk of failure when you want to sequence that tissue. Unfortunately you don’t get this kind of information as you are not always able to pick these patients up on liquid biopsy.
So all the studies that have been done on the PROfound data but also on the PROpel one, there was a poster on concordance between tissue and liquid biopsy, what we learned, basically, is that yes there is a concordance, the concordance is high, the concordance is high when you want to define the biomarker positive and also when you want to define the biomarker negative. However, if you look at the agreement between tissue and blood samples, cfDNA, you can pick up, basically, a biomarker positive that sometimes is biomarker negative on blood tests. That’s because you don’t… most likely, [inaudible] sequence loss on blood-based sequencing or because the tumour fraction of your cfDNA is quite low. So what we forget sometimes is that there is a chance that the sequencing fails also on blood when you interrogate the blood because there is not enough tumour fraction in that sample. This happens especially when you want to know this in patients that have been not exposed to a lot of drugs, so in quite naïve patients, even if they are metastatic CRPC. You have a good tumour fraction in patients that have failed more than one drug but still to know about [inaudible] rearrangement, know about homologous loss, that’s not always important.
All these lessons are coming, all these results and data are coming out now and this is something that we have to keep in mind, especially when we look at combination studies and we see differences between different studies. I think that sometimes the difference is not much in how different are the drugs or the PARP inhibitors, that could explain why you have similar studies with different data, but a lot is also how we have defined that biomarker positive population.
NS: So that’s a great review. It gets a little complicated for some of our colleagues when they’re trying to figure out, okay, can I get tissue? Can I get archival tissue? Will I do a metastasis directed biopsy? Do I have the wherewithal to do that? So Friederike I’d like to ask you two questions. Because you also mentioned something else about of course homologous deletions, BRCA2, fantastic – you find that, you get it, expression with tissue, wonderful – but Friedrike for you, you were an investigator in PROpel and you specialise in GU oncology, a busy practice, how are you using these learnings now in terms of the role for testing, both tissue and blood-based ctDNA testing, in your clinic as you’ve been an investigator? Then a second question, if I don’t want to overwhelm you with too many questions, is what is your sense about just the PARP inhibitor landscape, differences amongst PARPs?
FS: One thing that Pasquale has said that it is really difficult to have tissue for some of our patients. This is a real challenge so we all think or hope that the ctDNA will help us one day. We’re working on that right now, even at my centre we try to generate more data about that because I see that patients often are metastatic in a second time and the tissue that we have is too old and patients have bone only metastatic disease so we all know that it’s challenging to do these biopsies. It’s really a challenge and this is really important. We’re learning every day that we have to be organised in a multidisciplinary team to get radiologists, urologists, oncologists to get the best tissue that we can have and the best is always the freshest one, what you said. If we can have prostate that’s perfect but we can’t always have that. If we can have [inaudible] that’s great too. So we are learning what we have to biopsy and that we have to get fresh tissue to have the best result for our patients. So this is a learning curve that we are really working on.
But the thing with PARP inhibition is that we have different studies, that’s what you said, with different results, with different PARP inhibitors and the big question is really is there a difference between the PARP inhibitors that we have today. We have to really think about it because I think they are a little bit different. They have different tropism to bone marrow, for example, which can maybe explain toxicities that we see more with one drug than another drug. So the binding capacities with PARP1 and PARP2 may be different too. So I think there are a lot of things to explore.
From my clinical point of view, it’s a good result, a good thing for patients that even if they don’t have any mutation that they might benefit from a PARP inhibitor because we have a positive study, like you said. It’s a positive study in our PFS and that would help patients where we don’t have a lot of clinical options right now in that setting. So it’s really great news for our patients.
NS: Right. You can’t get the tissue, it’s not accessible, the tumour tissue has degraded. Patients might not want to go for metastasis directed biopsy, it could be bone only which is challenging, it’s accomplishable but challenging. And, as you say, depending upon when you’re getting the blood-based liquid biopsy, as we call it, it just may not be as robust as you want it to be, even though there’s still about 80% concordance with tissue. So that’s the whole concept of your leadership in combining and saying, okay, in an all-comers population we can see right now, today, we can say there has been a rather dramatic rPFS. An 8 month benefit against an active drug, abiraterone – it was not a placebo control and that’s very important for our colleagues. I know we don’t have much more time, but, Noel, what’s your response to contrarians who say, ‘Well, I’ve got to see OS.’ Even though it’s a trend, how do we respond to that?
NC: I think that’s a perfectly reasonable approach in considering a large-scale clinical trial, Neal, and you’ve done a number of these yourself and we’ve all been involved in them. It’s important to give a trial time to mature. So there are two aspects to this. One is will it go into clinical practice, and I hope it will but it’s too early to say at this stage and we need that third interim analysis where we have more events in the control arms. That will tell us with a degree of confidence, I hope, whether or not the combination is effective. What I see so far is very promising and I’m hopeful that it will come into practice. The second aspect to it is the biology question which, as a translational scientist, you know that a trial like this always raises a number of questions. The question for me is what is going on in that non-HRR mutated group of patients? There are elements of HRR mutation, DDR repair, the synergy with different drugs that we simply don’t understand. As you discussed, there are differences in PARPs, there are differences in toxicities and combinations of drugs. I think we need to work that out, particularly when we’re looking at drugs like the amide group of drugs – enzalutamide, apalutamide – that are inducers. So we have to lower the dose of PARP in certain circumstances to accommodate the marrow toxicity. Drug dependency, or the level of dose in a PARP inhibitor is critically important. So the biological questions I find fascinating. Ultimately, will it work for a patient? And I sincerely hope that it will because it will simplify everybody’s task in helping the patients.
NS: No, and thank you for that summary because I think ultimately some of our colleagues might worry about if we’re adding two drugs together is there a toxicity issue. By and large that’s been very nicely answered with PROpel following up and seeing not the same cardiovascular toxicity, there was a small percentage but we did see that in Study 08. This is ultimately our goal is to optimise patient therapy. Once we go from sensitive to resistant then it’s a different phenotype and we can’t always get the tissue, we can’t always get the blood-based sampling and so we clearly see, and Fred Saad gave a great presentation today here at ESMO – his oral presentation will be archival for folks who are listening to this, they can get it on the ESMO website. Andy Armstrong gave really a more detailed analysis of the biomarkers that were looked at, particularly the gene signatures. Your point, there are potentially going to be other gene signatures that we will need to interrogate in the all-comers – RB1 loss, maybe it’s PTEN, other things. I think that’s particularly exciting from a translational standpoint. Maybe just one last comment each and we’ll wrap up. Pasquale?
PR: I agree with Noel that we need to wait for the OS data but that’s something important to remember, that in metastatic CRPC we know that the rPFS is a surrogate endpoint of the OS. So hopefully the OS will go in the right direction and we’ll have a new combination, a combination soon. The other thing about testing, there are people saying, ‘Well, now we know that the combination works for everyone then we don’t need to bother anymore about testing.’ I think actually that’s not true because if you look at the BRCA, how BRCA do on abiraterone alone, you have an rPFS of 8 months. So I definitely want to know who are the BRCA patients because I want to make sure that they don’t have abiraterone alone. So definitely testing is still important. We have to optimise the test, for sure, whatever the test is, but that’s important.
NS: That’s such a great point and then the family implications as well.
PR: Absolutely, absolutely.
NS: A great, great point. Friederike?
FS: Yes, he already said a lot of things. I’m thinking like you, you said it’s so important for the families, it’s not only to know if a patient can get a drug. There is so much more on BRCA1 and 2 that it’s important for our patient and all his family. This is a really big point. On the other hand just one thing, I treat a lot of patients with olaparib and abiraterone and I saw the clinical benefit that they had and I saw a really good tolerance. I think it’s really exciting that we have that study. We still have so many things to understand why it works but I’m pretty enthusiastic about it and I really hope, too, that OS is going to come out positive and that we can give that to our patients.
NC: I think my final comment, Neal, would be in designing studies keep it as simple as you can and keep an open mind.
NS: Well said. That was really one of the brilliant things of PROpel, it was a classic 1:1 randomisation, just a really classic design. So we see obvious, highly statistically significant rPFS benefit, a tolerability profile that’s consistent with the class of agents used and we’ll continue to wait and see on the OS but a rather remarkable rPFS benefit in an all-comers population.
So thank you all very, very much. It’s fantastic. It’s an honour for me to be here with the three of you and to continue to do the great work that you’re doing. For the audience, on behalf of ecancer it’s great to be here at ESMO in Paris. A lot of the work that we’ve talked about you can go on to the ESMO website and you’ll be able to find the presentations. So on behalf of my colleagues and ecancer thank you very much for listening.