Dr Neal Shore – Carolina Urologic Research Center, Myrtle Beach, USA
Dr Neeraj Agarwal – Huntsman Cancer Istitute, Salt Lake City, Utah, USA
Dr Elena Castro – Spanish National Cancer Research Center, Madrid, Spain
Dr Kim Chi – BC Cancer Research Centre, Vancouver, Canada

NS: Hello, I’m Neal Shore, I’m the Medical Director of Carolina Urologic Research Center in South Carolina in the US. I’m a urologic oncologist. Today’s programme is brought to you by ecancer and we’re really happy to review the most recent cutting edge presentations from ESMO 2020. It was a virtual programme this year but nonetheless there was some really great activity. What we’re going to focus on today is the topic that’s really very important right now, known as homologous recombination repair defects, gene alterations. How do we find these patients? How do we appropriately test for them and, more importantly, what are the data that are coming forward and have been presented most recently here at ESMO 2020? We’re going to review the phase III trial known as the PROfound trial which has had a New England Journal publication as well as a presentation on its benefit for patients who have HRR mutations. We’re also going to review different presentations, abstracts from ESMO 2020, which included topics on testing – testing in North America, testing within Europe. We’ll also look at some other early phase trials combining different PARP inhibitors with other well-known CRPC therapeutics. Of course, we’ll have an interactive discussion with our faculty.

Let’s begin by commenting on this notion of HRR mutations, or what some might call DNA repair mechanism defect alterations. We’re familiar with the notions that they can be within the germline where you would have inherited these from your mother and father and therefore if they are inherited they essentially have ubiquity in all the cells of your body. They can be both monoallelic and biallelic and we’ll talk a little bit more about that. Then you can inherit, or not necessarily inherit, but you can develop alterations in these HRR mutations, homologous recombinant repair mutations, from the tumour tissue itself and we call these somatic alterations. We can test for them both with tissue as well as with liquid based testing. We have renowned experts on this call today, both Kim Chi and Elena Castro have done really super pioneering work in understanding the incidence prevalence of these alterations, both in patients who are newly diagnosed with high risk disease who have family histories, patients who have metastatic disease, both castration sensitive and castration resistant.

So it’s not a question any longer about when do you need to be thinking about this testing or when do you need to be thinking about these therapeutics, these PARP inhibitors which we’re going to describe, the essential comment is you have to be doing this now because this will help optimise your patient care. So let’s begin with you please, Neeraj, can you comment on the presentations that we had at ESMO of the PROfound trial, a global trial that, as we all know, will clearly change the landscape of therapeutic options for mCRPC patients?

NA: Absolutely. Thank you so much, Neal. It’s such a pleasure to be here on the panel with you, world-renowned experts like Dr Chi and Dr Elena Castro. So thank you for the opportunity. I think, no doubt in my mind, one of the most exciting presentations in the ESMO 2020 Scientific Programme meeting was the PROfound trial as presented by Dr Joaquin Mateo. I would just go with the introduction of the PROfound trial first followed by the results. The PROfound trial is important to me because this is the first ever conducted successful positive phase III trial in patients with metastatic castrate resistant prostate cancer where patients were selected based on a biomarker. This was history in the making in prostate cancer. Until then prostate cancer never had a single trial of this magnitude where patients were selected based on a biomarker. So out of 4,500 patients screened originally who were actually pre-screened and screened for this trial, ultimately 400 patients were registered. Approximately 400 patients were registered. There were two cohorts, cohort A and cohort B. Cohort A included 245 patients who had BRCA1, BRCA2 and ATM mutations. Cohort B had twelve other mutations which are non-BRCA1, BRCA2 or ATM mutations, 145 patients in cohort B. The primary endpoint was radiographic progression free survival in cohort A and then, of course, there were several secondary endpoints such as radiographic progression free survival in all patients and how much delay was there to time to pain progression or time to cytotoxic chemotherapy and so on. So I will focus on the primary endpoint and the overall survival data which were presented in ESMO just last week.

Previously we have seen the data on radiographic progression free survival being presented in the ASCO meeting. So regarding the overall survival data, what we saw in this meeting, these were the final overall survival results from the PROfound trial. We saw something really striking – in these patients who had progressive metastatic castration resistant prostate cancer progressing on a novel hormonal therapy with abiraterone or enzalutamide, and many had these progressions even on chemotherapy with docetaxel, when they were randomised to a PARP inhibitor olaparib versus an alternate entity with either abiraterone or enzalutamide there was a very remarkable, significant overall survival benefit with a hazard ratio of 0.69 favouring olaparib. The median overall survival was almost 4.5 months improved from approximately 14.5 months in the control arm to 19 months in the experimental arm with olaparib.

I would like to bring this to your attention that 66% of patients in the trial had crossed over during the conduct of the trial from the control arm to the experiment arm with olaparib. We all know that whenever we see this crossover, the crossover has the potential to attenuate the degree of survival benefit. So what I think was quite interesting is that the effect of crossover was adjusted by a mathematical statistical formula and then what we saw was if you adjust for the crossover of those 66% of patients from novel hormonal therapy to the olaparib arm, the overall survival improves even further, bringing down the hazard ratio to 0.42 favouring olaparib. I think this is very exciting news for our patients.

NS: Thanks Neeraj. That’s a great overview. So, yes, not only was the primary endpoint met for the rPFS but now the secondary endpoint of overall survival was demonstrated and that’s what Joaquin did such a great job of demonstrating. To your point, not only was it met but there was a really high crossover and any time you do a crossover it blunts the delta, so to speak, of those patients who would have never received an active therapy such as olaparib. I think it’s also interesting, as you say, it puts another nail in the coffin, at least the way I think about it, in sequencing the androgen receptor access drugs, especially when you can personalise, as you say. This is a landmark trial in that you now have compendium testing to say, ‘I have an HRR mutation positive finding and here I can benefit clearly from a unique mechanism of action of PARP inhibitor as opposed to sequencing to yet another well-known androgen receptor access drug, whether it’s abiraterone or enzalutamide.’ Kim, what do you think about this? I know you’re a co-author here, you’ve been very involved. How will this change how you manage patients or how our colleagues should think about managing patients going forward, assuming they have accessibility?

KC: I’d like to touch on two points. One is your point about next generation hormone therapy and, as we can see again, multiple studies have shown that switching from one next generation hormone therapy to another really yields minimal benefit and in this case patients are progressing within three months. Really that’s from one scan to the next. So it is not benefitting our patients. The other point to emphasise is this degree of crossover and we have to remember that on average on the control arm the patients were progressing at about three months and therefore crossing over and yet we still saw this huge survival benefit so emphasising that earlier treatment is better. We’ve seen that over and over again in multiple trials that we’ve done in prostate cancer – earlier treatment is better. So we really need to identify these patients and get them the treatment earlier so that they can have the most benefit. Delaying with another next generation hormone therapy, and we could argue about chemotherapy as well, may not be in the best interests of these patients.

NS: Elena, your thoughts on this study?

EC: Yes, I agree with what is being said previously that this is a shifting in the management of prostate cancer entirely and that we should start screening for these alterations as soon as patients become at least metastatic in order to start having a whole picture could be the best treatment for our patients so we can consider a PARP inhibitor, if that is the case, as soon as possible. Because what we have seen is that at least for some of these alterations patients have very poor prognosis, it may not be applicable for all the HRR alterations but at least for some of them. It’s really a chance to change the prognosis of these patients by using a PARP inhibitor as soon as possible. So definitely I would recommend to do this and try to screen all patients as soon as possible.

NS: So let me ask you all a question about cohort A was BRCA2/1 and ATM and we see the most common of the 15 gene panel was BRCA2. Not surprising we’ve seen that in other PARP inhibitor trials where testing has been rigorously performed. You’ll hear some of our colleagues say, ‘Well, most of the rPFS and the OS was really dominated by the presence of the BRCA2.’ I’m curious now if you had access to olaparib, based upon this data, and in the US the FDA has now approved 14 of the 15 genes that were studied. Based upon the trial design, so if a patient is mCRPC and they’re progressing after a novel hormonal agent, typically abiraterone or enzalutamide, they don’t have to have received docetaxel or they can, and you check the patient with germline or somatic or maybe both and they had one of these lesser of the 14 genes, that was in cohort B, how do you think about advising patients? Kim, and then maybe Neeraj and then Elena.

KC: I think that’s a great question, Neal, and it’s on the minds of many of us about how are we going to prioritise these biomarkers for selecting patients for treatment. No doubt, BRCA2 demonstrates a very high response and a large benefit for patients. But when we look at some of the subgroup analyses, some of those other genes, we don’t see as good of a benefit in the population. However, we know from some single patient single gene analysis that there are gene alterations that respond well to PARP inhibitors but the numbers are so small because you have this long tail in the curve of these other alterations. So, for example, RAD51, PALB2, these have been associated with high response rates although the numbers are very small. So at this time it’s difficult to answer that question. I think I’d be less enthusiastic about ATM and CDK12 if I have an alternate treatment to try. So we know that we have docetaxel, cabazitaxel, for instance, and most studies have demonstrated good activity of these agents in patients with homologous recombination repair defects.

So I think if I have an alternate treatment to try with known efficacy, I would probably do that prior in these particular alterations. However, if there was another patient with PALB2 or RAD51 I’d be more inclined to try the BRCA2, there’s not enough data. So what we need to do is continue our studies in this, analysing not just the alterations but other ways of identifying patients with a homologous recombination repair defect.

NS: Great, great. Neeraj?

NA: Yes, I agree with what Kim said, most of it. I always agree with him anyway but I agree. ATM is somewhere where I still am not feeling comfortable. I think we need more data on ATM. But otherwise I would like to point out that patients with individual genes, like how many patients were there in each one of the small cohorts of those twelve genes? They’re so small that it is difficult for us to infer the advantages for each of those groups. So I think I won’t get into subset analysis beyond ATM. ATM was a pre-specified analysis so I think I’ll stick to that but other gene analyses, when I see my patients with PALB2 mutation of RAD51 mutation I’ll tell them, ‘Look, this is a trial which included these patients and you’ve got to try this,’ and what other options we have. I’ll be honest with you, most of my patients are not interested in chemotherapy with docetaxel and cabazitaxel. Alternate NHT, Kim’s study has shown us that alternate NHT doesn’t work very well. So what other options do we have? So I think if I have a phase III trial showing benefit with olaparib in these patients I would try that.

NS: Just for time, let’s switch over to some of the other work that we saw at ESMO. I’d like to get your thoughts first on this, Elena. Let me review some of the things that I was fortunate enough to be part of. We did some retrospective analysis of a large data registry known as the Flatiron Data Registry in the US - a little over 5,000 patients over a six year period. Medical oncology practice: 90% community, 10% academic. What we found during that six year period from 2013-2019 was about 13% of patients were receiving testing for HRR mutations. The majority of it was germline and a very small minority was somatic testing. Now, you can argue why would I be testing, doctor, if I’m in one of those clinics if I didn’t have a clinical trial? Which, of course, we always think that’s a great thing to be doing, we should be enrolling patients in clinical trials whenever possible. All the international guidelines support that. But if I wasn’t part of a clinical trials access and I didn’t have an approved therapy, why would I test? The answer is now because you have an approved indication and actionable findings. But in our study it was only about 13% of these patients were receiving any type of testing.

Then in a second abstract that we presented of that same registry we looked at sequencing regimens and the patients who had had the testing and were HRR mutational positive did particularly poor. The other finding in that second abstract we presented at ESMO was that, no surprise, but the algorithms, the recipes, the cocktails for how our colleagues across the US and, I think, true for the world, are sequencing our patients. It’s all over the place. So now that we have the opportunity to test someone and, quote, personalise their care could theoretically really optimise their delay in progression and their survival.

Similarly there was a nice presentation of the largest five European Union countries by Leith and others which also showed a remarkable paucity of testing. So, Elena, what do we recommend now to our medical oncology, our urologic oncology, radiation oncology colleagues, not just really in the EU but in North America and really globally about testing? Your thoughts on getting maybe a little more granular about germline and somatic testing.

EC: I think this is going to change, as you were saying, now that we have some therapies that we can use when we identify any of these HRR alterations. So, for sure, I’m convinced that the uptake of tumour testing will be increased. With regards to germline testing, I believe most clinics, most hospitals, have access to genetic counsellors and genetic testing but it has not been until recently that the guidelines have been updated to include the recommendation for germline testing for patients with metastatic prostate cancer. So in many countries if this was not included in guidelines there’s no chance of reimbursement. In most healthcare systems in Europe this is one of the things that drives the opportunity of having therapies or counselling reimbursed that will drive these decisions.

The other thing is there is a shortage of counsellors in order to provide counselling and testing for all the patients with prostate cancer, ovarian cancer, pancreatic cancer. So this is something we really need to improve either by acquiring ourselves the needed skills to provide a minimum counselling for our patients in order to recommend this testing. Pre-counselling should always be done before requesting the test but it could be a very short test without the need to refer to a counsellor. Then when the results come back, if needed, if it is a positive test, then we refer the patient to the counsellor. But germline testing, I believe, is something we should be doing for all our patients with advanced disease because around 10% of them may harbour a germline mutation.

For patients with BRCA2 mutations it is clear that it will have an impact on their prognosis. Of course we have all these PARP inhibitors but also we don’t really know which is the best treatment sequence for these patients. I’m sure we will have more studies in the years to come and we will better understand what is the most appropriate sequencing for these patients. But we have to do that for the families, for the healthy relatives at risk of increased different types of cancer.

NS: Yes, I really appreciate your comment. We have to up our game. It’s great if you have a genetic counsellor, certified genetic counsellor who has oncologic training and is not just schooled in reproductive genetic counselling and that’s ideal. But practically speaking for the world we need to up our game and be able to have some understanding of what does a positive HRR mutation mean and other mutations as well, both germline and somatic. Let me ask you, Kim, if a patient is being seen by his or her medical oncologist and a patient has mCRPC, they’re on enzalutamide or abiraterone and they get a germline test and the germline is negative is that it? Do you just stop there? What do you do next or anything?

KC: We know that for patients with metastatic CRPC and having a homologous recombination repair defect about half of them are germline, roughly, as Elena said about 10%, but the other half are somatic. So that’s an important factor. So you’re only picking up half or even less of the patients that could be eligible for a PARP inhibitor as well as understanding their prognosis more. So I would also advocate that all patients get tested, not only just for the germline but also the somatic as well.

There are various ways for testing this. I found your abstracts very interesting, Neal, because of the low testing rate. It’s widely accessible in those centres as well. So only 10% of the patients, or roughly 10%, were getting tested. Just on the germline alone it really should be closer to 100% of people getting tested because of the implications for families. But we also have to emphasise that we need to do the somatic testing as well.

NS: Thank you. This is a sea change. This is a big deal. We have a lot of data over the course of our congresses that we make some incremental advances. This is a big incremental advance, this is a real change.

KC: It’s a huge number of patients as well. This is prostate cancer, not ovarian cancer which is a relatively small number. So we need to work with our hereditary cancer folks and genetic counsellors to get this organised. Our team is starting large class pre-test counselling as well as video counselling so patients can do it from the comforts of their home. So we really need to change how we do things at this point in time.

EC: Yes, sorry, I was going to say that I totally agree. This cannot be an excuse that we don’t have… We have so many resources to counsel these patients, to explain to them why are we doing that and we should use them. Definitely we should do testing in all our patients. I didn’t mean only germline, definitely both – germline and somatic. There are so many ways how we can counsel our patients, that shouldn’t be the reason why not to do it.

NS: So, Neeraj, what are your thoughts about the differentials between tissue testing, somatic versus liquid testing. Sometimes that creates some confusion, I think, for our colleagues who are getting into it. If you’re going to tissue test when do you get the tissue to test it? Is it the primary organ, the prostate? Do you get it when they’re castration sensitive, metastatic? Do you wait for them to be castration resistant? Where are you right now in your thoughts on liquid based testing? Are they the same? Do you get the same concordance of positivity?

NA: Those are great questions, Neal. First of all, I agree with Elena that these are early events, at least that’s what I have seen in my clinic. So any time I see a patient for the first time with advanced prostate cancer, and this is my practice which I have been following for at least the last one year, any time I see a patient with advanced prostate cancer for the first time, when I am ordering CT scans and bone scans and haemoglobin levels I order three more tests at the same time – tissue comprehensive genomic profiling, liquid biopsy, liquid CGP profiling and referring the patient for germline screening, high risk genetic screening. This has become part of my initial work-up for all patients with prostate cancer. I know that many of the patients would have had a prostatectomy 5-7 years ago and the tissue may not be eligible for testing by CGP. Most of the commercial testing platforms don’t want to extract tissue which is more than six years old.

Regarding liquid biopsy we are going to present in the GU ASCO the concordance of hundreds of solid tumour biopsies versus liquid biopsy. I can tell you right now that concordance for the overlapping genes is pretty good. The liquid biopsy platform, it has been fascinating to see the liquid biopsy platform evolving so well and so nicely that they are now competing with solid tumour panels, at least the commercially available ones which are easily available. So that’s my take – every patient should be tested whenever we see them for the first time, not necessarily waiting for the onset of mCRPC. Because sometimes we lose these patients if we wait for the onset of more aggressive disease, or we lose the tissue or tissue is thrown away by some of the local hospitals. So I think test these patients early on as much as possible.

NS: Thank you, I agree. I do exactly what you do, I do exactly that and I’ve been doing it for a couple of years. You have a Eureka moment, you go, ‘Wow, I’ve found something now that is actionable. I can put you into a trial or now that we have approved therapies.’ So I completely agree. You’re right – sometimes the biopsy specimen, the archival tissue, the prostatectomy specimen, you can’t get access to it, it may have been poorly preserved, it may have degraded the tumour tissue. So I tend to have the mind-set of seize the moment while I can. Of course, for all of us and everyone listening around the world, accessibility to doing this and cost are clearly considerations. We need to continue to fight for our patients to be able to have that.

So, with the time remaining, there were some really interesting abstracts of combination concepts. So now that the PARP inhibitors as a class are here and olaparib is really the first in class to have a phase III trial. In the US we also have rucaparib which has received FDA approval. There are two other PARP inhibitors that are being actively trialled across the world – niraparib and talazoparib. There are combination studies combining the PARP inhibitor drugs with checkpoint inhibitors. There was a nice abstract presented with the combination of pembrolizumab, part of the KEYNOTE-365 large phase IIs. There are combinations previously and ongoing with abiraterone, with enzalutamide. So there is combination work being done, the NRG had a nice paper looking with ADT and radiation therapy, patients who had HRR mutations. What are some of your thoughts, let’s maybe go around the room with Neeraj and then Kim and Elena, where you think the future will be going. We typically in prostate cancer always get approvals with the patients we look at and who we accrue are the most advanced, heavily beaten up as the PROfound trial showed. These are patients with extensive amounts of tumour burden on multiple lines of therapy and yet got a great result relative to the terminal nature of their disease. So where are you optimistic, Neeraj? You do a tremendous amount of trials.

NA: I remain very optimistic about the use of PARP inhibitors in patients who are selected for DNA repair gene mutations or defects either in germline or in somatic or in tumour tissue. However, I do not feel as enthusiastic for the use of PARP inhibitors in unselected patients. So far we have a small phase II trial which has shown some encouraging signals but if you look at the pembrolizumab olaparib trial, the KEYLYNK-010 trial, the disease control rate, if you look at the objective response rate in unselected patients it’s still in single digits or around 10%. The original trial by Dr Karzai, Fatima Karzai from NCI, in durvalumab with olaparib, I think it was olaparib the PARP inhibitor, most of the patients who responded were selected for DNA repair gene mutations. But we’ll have the answer very soon. I think that three large phase III trials are going on, all of us are on the panels and steering committees of those trials, and those large trials are recruiting patients with androgen deprivation therapy plus novel hormonal therapy plus or minus PARP inhibitors such as talazoparib, olaparib and niraparib. I think it’s a matter of time in the next months or years, the next one year, we’ll be starting to hear how these agents performed in unselected patients. So I will wait, cautiously waiting and watching.

NS: Yes, absolutely. Kim?

KC: I agree with Neeraj and we often agree together with these things, as Neeraj stated earlier. But I think we’ve seen great benefit in this late stage biomarker positive homologous recombination repair altered patients and we’ll see that move earlier into the disease, into castration sensitive disease and then into even localised disease. So that’s a reasonable way to proceed and I think we’ll see benefits there. For unselected patients the jury is out. There are a lot of hypotheses for the combinations, however, I think it will have to remain to see whether it pans out in phase III trials.

NS: Yes. And Elena, I’m sure you probably also agree, but I want to hear your thoughts on combinations that you’re working on. Maybe if you would just comment for our colleagues listening about any particular adverse event management that as a class that one could expect to see with PARP inhibitors.

EC: Well, with the combination of PARP inhibitors, for instance, with abiraterone we have seen a number of patients with hypertension, manageable but a significant number of patients develop hypertension. There’s also an increased creatinine which is not related to renal failure but is rather associated with the renal transporter. So it’s quite common to see in patients to do with PARP inhibitors a rise in creatinine, it doesn’t really mean that the renal function is being altered. Of course anaemia, anaemia is probably the most common side effect. Thrombocytopenia in some patients. Nausea more than vomiting and in some patients asthenia. But I have to say that in general, in my experience, PARP inhibitors are quite well tolerated.

Once again, patients who receive PARP inhibitors earlier usually have less toxicities and can benefit from these therapies for longer. So, once again, we should try to use them as soon as possible in those patients likely to benefit from them.

NS: Great, thank you. That has been my experience as well. I think we’ve had a great discussion, maybe I’d like to give you all just thirty seconds or so, a wrap-up take-home message for our colleagues as it relates to PARP inhibitors, germline/somatic testing and some of the pearls that we learned from ESMO 2020. Let’s start with you, Neeraj.

NA: We have, for the first time in the history of prostate cancer medicine, a very well tolerated class of drug known as PARP inhibitors, in the form of olaparib and rucaparib, available in the clinic for those patients who have germline mutations or somatic mutations in their DNA repair genes. These agents improve survival, improve response rate, delay chemotherapy drugs. There is no reason for us not to test our patients for these mutations anymore. Every single patient with advanced prostate cancer should be tested for gene mutations in DNA repair genes.

NS: Great, well said. That’s hard to follow that one, Kim.

KC: Test early, test everyone and we should be using PARP inhibitors in those that are homologous recombination repair altered.

NS: Yeah, great. And Elena?

EC: I will add something else, that there is also the IPATential trial that was also presented. We will be testing early and not only for HRR defects but we will start testing for a number of other profiling tumours, for a number of alterations that are finally targets for available therapy. So it’s an exciting new time and we should really, really get ready for it.

NS: Yeah, you’re spot on. I love that you brought that up. We need another hour to talk more about some of these other important studies. The IPAT trial was a great trial in the whole notion of these PTEN loss alterations and treating patients in this other remarkably new class of PI3K/AKT pathway drug targets, ipatasertib, capivasertib, real exciting but you will never know to do it if you don’t do the testing. So thank you for that last comment, Elena, that’s great. And thank you all for your dedication and all the work that you’re doing. Thanks for your impression of some of the work that was presented at ESMO 2020. We thank the folks at ecancer for allowing us to speak with our colleagues throughout the world and wish everybody well and safety and health and happiness during the pandemic. Thanks very much.