mCRPC and nmCRPC: Combinations and latest data, what does COVID mean for early treatment?

Bookmark and Share
Published: 19 Feb 2021
Views: 3408
Dr Eleni Efstathiou, Prof Stefan Oudard Prof Boris Hadaschik and Prof Gerhardt Attard

Dr Eleni Efstathiou (MD Anderson Cancer Center, Houston, USA), Prof Stefan Oudard (Georges Pompidou Hospital, Paris, France), Prof Gerhardt Attard (University College London, London, UK) and Prof Boris Hadaschik (University of Duisburg-Essen, Essen, Germany) discuss combinations and latest data regarding non-metastatic castration-resistant prostate cancer and what does COVID-19 mean for early treatment?

They initially discuss treatment paradigms, novel drugs and new, verified treatment regimens and advances announced at the ASCO GU 2021 for nmCRPC. They then, move on to further discuss different combination therapies for patients with metastatic and non-metastatic castration-resistant prostate cancer. Within the discussion, they talk about the data from several like the ARAMIS trial, final results from ACIS and TheraP ANZUP. Prof Attard talks about PI3K/AKT pathway biomarkers analysis from the phase III IPATential150 trial of ipatasertib plus abiraterone in metastatic castration-resistant prostate cancer and Prof Oudard and Attard also discuss molecular determinants associated with long-term response to apalutamide (APA) in nonmetastatic castration-resistant prostate cancer. Dr Efstathiou and Prof Hadaschik talk about what the future looks like for these patients in terms of new treatment options.

In the end, Prof Oudard and Hadaschik also discuss how the COVID-19 pandemic has impacted early treatment of these patients and what are the consequences of it. They explain the role of androgen deprivation therapy on the clinical course of COVID-19 infection in men with prostate cancer and other studies that were presented at ASCO GU 2021 regarding it.

This programme has been supported by an unrestricted educational grant from Janssen.

2021 Genitourinary Cancers Symposium

mCRPC and nmCRPC: Combinations and latest data, what does COVID mean for early treatment?

Dr Eleni Efstathiou – MD Anderson Cancer Center, Houston, USA
Prof Stefan Oudard – Georges Pompidou Hospital, Paris, France
Prof Gerhardt Attard – University College London, London, UK
Prof Boris Hadaschik – University of Duisburg-Essen, Essen, Germany

EE: Good day. I am Eleni Efstathiou and thank you for joining us in this ecancer review of the data that was presented during this last ASCO with regards to advances and prospects in the therapy of patients with non-metastatic and metastatic castration resistant prostate cancer. We have the pleasure and honour to have in our company three very esteemed colleagues from Europe. I am Eleni Efstathiou, also from Europe and the United States and I would like to turn to Stefan Oudard, a person I would like to call a friend apart from an esteemed colleague, to have his introduction. Then to Dr Boris Hadaschik, also a urologist from Germany, and finally Dr Attard – as I see you, there is no rhyme in how, as I see you on the screen to present yourself before we embark on our discussion.

SO: Thank you Eleni. So my name is Stefan Oudard, I’m a medical doctor. I work in Georges Pompidou Hospital in Paris, France, and I’m really happy to be part of this discussion on the ASCO GU meeting which unfortunately was not a face-to-face meeting but nevertheless nice and interesting things to look at.

EE: Boris?

BH: Hello Stefan, hello Eleni, thank you very much. My name is Boris Hadaschik, I’m a urologist based in Essen in Germany and I’m very happy to provide a urologist or surgeon’s perspective to our discussion.

EE: And Gerhardt? We go way back with Gerhardt. We started together, I think, right?

GA: Yes, many, many years. At least fifteen, I think, Eleni, since we were…

EE: Thirteen, thirteen, don’t give us more. We don’t want to be old here.

GA: We were both awarded the Young Investigator Award from the Prostate Cancer Foundation in 2008. Always good to discuss the abstracts, thank you for inviting me. My name is Gerhardt Attard, I’m a medical oncologist in University College London and the third medical oncologist on the panel. We are outnumbering urologists, we need to address that for the next discussion.

EE: Absolutely. And we’ve turned the tables – it wasn’t like that. So I think we need to find a balance again, Boris, so I’ll be turning to you a lot for your thoughts because this is not about who is going to take over, this is about all of us sharing the knowledge and making the most for our patients. So with that in mind, indeed as Stefan pointed out, we’ve been really doing all this from the confines of our home, our office. We’re able to more interact now with our own practice, apart from Zoom meetings like this. So sometimes this leads to a little bit of an isolation that is concerning. The whole idea of these big meetings was to come together to change ideas to get the real vibe of what we think about what was presented. Now if we follow it through Twitter you see, those of us who do, that people walk away with their own impression. There’s not the opportunity for a lot of dialogue. So we’re trying to fix that a little bit through these meetings, I believe.
With that in mind, let’s get started with the non-metastatic data, and when I say non-metastatic, turning also to Boris Hadaschik who is a person who has been all about novel imaging, apart from his other work, you didn’t mention that, Boris. I want to stress that we are going by non-metastatic with the conventional term, by conventional imaging, not by enhanced and novel imaging modalities. So not a lot of data presented with regard to non-metastatic, Boris. I think the main presentation that had to do with clinical assignment of treatment paradigms was the analysis of the crossover patients in the ARAMIS trial, meaning the patients who upon unblinding were offered darolutamide when they were on placebo. The comparison, even though I believe it was both of those outcomes, as compared to those patients who were early on. We know well that in the SPARTAN trial this was done a priori set thanks to the PFS2 exploratory endpoint but I’d like your thoughts on this ARAMIS presentation because it’s an issue of reproducibility and reinforcement of what we already know.

BH: So if we remember M0 CRPC is a niche and we have three big phase III trials with the primary endpoint of metastasis free survival. Then later on, last year, we got the presentations of overall survival benefit and all three trials showed really remarkable efficacy of treating early, so before there are metastases on conventional imaging. There was an improvement of roughly two years with all of the three drugs in MFS. So the overall survival data are a little bit less mature but they also point out a big 30% difference in overall survival between men receiving treatment early and men receiving treatment only at the time of metastasis, so mCRPC. In the SPARTAN trial, which I was involved in, we did an analysis to account for crossover because in SPARTAN, which is an older trial than ARAMIS, the time there were patients on placebo at the time of final analysis was more than two years. So we saw an improvement in the hazard ratio when we accounted for crossover.

Now the data presented here by Neal Shore on ARAMIS did not show a big improvement if you account for crossover but this is, in my opinion, based on a lot shorter time that men had in the crossover portion. So the remarkable overall survival difference of darolutamide of 30% remained stable despite accounting for crossover. So I think this is reassuring and interestingly the hazard ratio of 0.69 in ARAMIS compares well to the crossover adjusted hazard ratio of 0.69 in SPARTAN. So I think all three drugs are good and they just reinforce that early treatment in this patient population makes sense because we know and we have to remember that all of these men have a short PSA doubling time. If you use other imaging, for example, we know these men have much more disease than this M0 on conventional imaging.

EE: Thank you for that very, very detailed explanation and reinforcement of the fact that the data is reproducible. We’re not going to handhold anyone to decide which one is their choice, a lot of it is based on your experience with patients, of course. As you very clearly said, Boris, we are all involved in different trials and have our own experience with using these datasets. I don’t know if, Stefan, you wanted to add something with regards to the nmCRPC space? It’s still a difficult space and a lot of people are still asking, now that we’re using more and more enhanced imaging or if we have a local disease that is recurring and we’ve exhausted radiation, for instance, what do we do? Should we still rely on these datasets that came from these three trials or should we look further? Do you have anything to add to that?

SO: Well, if you look at the SPARTAN study or these three studies are fine, are great, and outstanding in terms of outcome regarding MFS. In the SPARTAN study we had an update regarding the exploratory analysis regarding the biomarker analysis for those patients. They tried to find out, to separate long-term responders to early progressors to find out if you have any different signature for those patients. They performed signature characterising into three general mechanics classes: either immune regulation proliferation or hormone dependency. They found out for those patients with long-term responder that they have increased immune activity, they have decreased vascularisation of the tumour and also proliferation of the tumour cell capacity at baseline which is associated with a good outcome. In the contrary, for early progressors it seemed that they have a kind of basal component of the tumour as well as neuroendocrine-like tumours as well. So maybe so far it’s not possible in daily practice. We could have some information regarding those patients who benefit the most from this new hormonal therapy.

EE: Thank you Stefan for the comment regarding the effort of reaching that Holy Grail of being able to identify early on who are going to be the big winners with specific treatments. I’m going to hold that thought because we’re going to dedicate a little bit of a bigger part of the discussion led, I would say, by Gerhardt regarding this effort to refine and identify biomarkers that will lead us to that precision that we really, really need as the next step of our development. That’s where we’re lacking in prostate cancer. With that in mind, I want to move over to mCRPC and just a brief comment from Boris, sorry to go back to you on that one. We saw one big trial, it wasn’t small, it was I think 200 patients from the Asian population which was addressing the obvious; something that we are looking at like, ‘Is this still a question?’ We know that in Asia and other parts of the world even, I believe, in South America, there is still that view of life that we should use first generation antiandrogens before we move on to second generation enhanced androgen signalling inhibitors. With that in mind, we see patients being treated with bicalutamide still early on and then moving on to flutamide. Full disclosure – I haven’t used bicalutamide, apart from that first induction period when we use an LHRH agonist, for years so I’m the least specialised to speak on that. Boris, your thoughts a little bit on that trial and whether you think the answer is here and we should no longer challenge the tumour with these approaches.

BH: So my disclosure is that I sometimes do use bicalutamide in the context of salvage radiotherapy because of the Shipley data but I have not used flutamide for years. But, as you correctly pointed out, there are regions in the world where it is still used and it is much cheaper than the newer generation drugs. So I think there is a reason so it’s important that now from Japan we have this phase IV study called AFTERCAB where they compared enzalutamide to flutamide in men with asymptomatic or mildly symptomatic castration resistant disease. There is a clear improvement in outcome and PSA progression rates with enzalutamide over flutamide so I think we can really put this question to rest. If you have the newer drugs available and they are reimbursed there is no reason anymore to change between first generation antiandrogens and I don’t see any space anymore for flutamide.

EE: When everybody asks us, no, but we’ll do it, we’ll just show this trial and we’ll be done with it. But let’s move to the big deal of this ASCO meeting with regard to clinical data presented. We usually look to phase III trials to get a reaffirmation of what we believe would be the future. I’m going to actually just say right off the bat that all three of us medical oncologists were part of this phase III trial that was presented, we were PIs, we’re even listed in the co-authorship. I’m referring to the ACIS trial. This is a long-anticipated trial, I remember initiating this trial in my institution in 2015. The data took much, much longer to mature than we expected – six years –and this is usually a good thing. It means that our patients are responding and potentially to both arms and it’s a good thing overall for the space, for mCRPC. Never would we expect back in 2007 that we would run an mCRPC trial that would take more than six years to mature. This was a trial in the space of mCRPC where patients were offered ADT plus abiraterone with or without the addition of apalutamide. The primary endpoint was just rPFS as defined by the PI, the investigator who was treating. Secondary endpoint was overall survival. So I’m going to turn to Gerhardt to get his thoughts then hear from Stefan in his experience and finally I’d like a comment from Boris on the results as the independent reviewer who was not part of the trial. Gentlemen.

GA: Yes, thank you. You introduced this well, Eleni. The primary analysis which had reported about eighteen months ago, I think, was to define rPFS but the IPMC recommended not reporting at that time. So this is the final analysis which, as you said, is about two years later than we expected the first report. So it’s the first report of the trial but it includes both the primary analysis which, actually, was positive and this final analysis. The primary endpoint, rPFS, a very significant benefit for the combination of apalutamide plus abiraterone versus abiraterone alone which proves the hypothesis that dual AR targeting is more effective than single AR targeting.

The question now stands that will subsequent treatment with an AR targeting drug like enzalutamide or apalutamide recover or salvage the benefit and result in an overall stable overall survival. The trial has not addressed that. OS was a secondary endpoint. There is a numerical benefit in terms of survival for the combination versus abiraterone alone but that is not statistically significant and the trial wasn’t designed to answer that question. So the jury is out.
The second question that now arises, and that you asked me about earlier, is in the HSPC setting. Repeatedly we’ve now seen that a more effective treatment when started at the start of ADT has a greater magnitude of benefit in terms of overall survival. The data would support that that is maintained even when that effective treatment is used later. So, earlier treatment with docetaxel, with the AR-targeting next generation drugs, improves survival that is not salvaged by later treatment in CRPC.

The only trial that will address that or that has recruited patients to address that is the STAMPEDE trial. There was a comparison of 2,000 men randomised between abiraterone/enzalutamide plus ADT versus ADT alone. That trial has a primary endpoint of OS. That trial should report probably in the next twelve months; it’s event driven and was powered to hit the events I guess in the next six months or so. But just a footnote, I was explaining earlier I’m very involved in the STAMPEDE trial and, in fact, led that comparison. When we designed it we aimed to show a difference versus abiraterone alone in an indirect meta-analysis because patients are not concurrently randomised to abiraterone/ADT or enzalutamide/ADT versus abiraterone/enzalutamide plus ADT. This is an academic trial and you have to allocate resource in the best way possible. The statistical analysis plan was powered to detect the difference in the benefit for survival with abiraterone alone it was about 0.75 hazard ratio. There was a hazard ratio of 0.63 showing a superiority in an indirect comparison in an unselected population. It’s going to be statistically challenging.
So, to be honest, I’m still mulling over how best we use or interpret the data from ACIS. I guess it certainly increased the relevance of that question and over the next six months we need to think really carefully how we’ll interpret that data now in the context of this positive result in ACIS.

EE: Thank you, I’m smiling listening to you because we’ve had over the years, all of us, had these discussions from those first presentations combining abiraterone with enzalutamide back from now 2012. I’m going to turn to Stefan with that question because you really pointed it out, saying that even numerically we see that benefit in the combination suggesting that we might reach a point where we are able to define which group has this very extreme androgen signalling addiction where one is enough versus the group where you need to combine versus a group that it doesn’t really work that are known as primary resistant. That’s going to be our next topic of discussion, based on this initial SPARTAN dataset that Stefan mentioned. With that in mind, and taking into consideration what people in the community are saying: ‘Hey you guys, you’re going to ask me to do three drugs, ADT, apalutamide, abiraterone or enzalutamide or darolutamide or what have you. What’s the safety there and what’s the proof that I’m actually serving the therapeutic index, meaning efficacy over safety?’ Stefan, your thoughts? Will this transcend to real practice eventually?

SO: I agree, I agree with Gerhardt about the fact that the primary analysis was blinded because rPFS was six months at the beginning and then moved to 7.4 months at the end. So maybe for the OS, that maybe will move as well with mature data, this is my first point. Anyway, the study is positive according to the primary endpoint so this is a good point. The second point, as you said, about toxicity. I don’t know in real life what is going to happen to the patients with maybe more comorbidity and so on regarding combination. The third point is about should we go for combination or sequencing? This is really the point because we need to look at PFS2 to see whether or not with a combination you don’t induce complete resistance to further treatment so far. This is not well understood so far so we need to look at that. So the question is there. If you look at the Alliance study from Michael Morris, the PFS was obtained – 4.5 months, no benefit in OS. So what should be concluded in the end? Should we go for combination right away or should we try to sequence as well on specific patient populations? This may lead to introduce about which patients could benefit the most – maybe patients with, I don’t know, liver mets or luminal versus basal subtype. We need to see which patients could benefit the most from that.

EE: Thank you Stefan and, indeed, the Alliance trial was positive for rPFS, negative for OS but OS was the primary endpoint there so that’s the problem. And here go my peacocks again, apologies for that. But, Boris, one quick thought on whether you think this will be, under circumstances, a practice-changing study and what would convince you further?

BH: I think at this point in time it’s not a practice-changing study because since 2015 things have changed. Especially now in COVID, first line treatment in hormone sensitive metastatic disease is already an androgen receptor signalling targeted agent. So we don’t give right now docetaxel in first line; we tend to use one of the new drugs. The hazard ratio, the updated one with apalutamide for all comers, is pretty convincing and so that makes things for then first line mCRPC challenging. ACIS was first line mCRPC, now things have changed because we deal with a different patient population. So I’m waiting for the data that Gerhardt is going to present so this would make a difference for me in the hormone sensitive setting. But until now it doesn’t change anything because we are not able to stratify men at this point in time in a fully convincing and practice changing way. But in Germany biomarkers are a little bit more difficult than imaging so I’m looking forward to the results of you because I know you look more at biomarkers than we do.

EE: As you said it very nicely, STAMPEDE is going to come again and save the day. You guys have been doing it since 2015 right off the bat from the first data that confirmed CHAARTED and since that time you’ve led the way and thank you all for that. I want to move on, in the interests of time and our viewers, to one little comment. I’ll take it back it to Stefan. Stefan, you mentioned a little bit about sequencing and we thought we’d put it a little bit to sleep after we looked at this great FLAC data suggesting that you should not sequence androgen signalling inhibitors in the setting of mCRPC and that you should transition to another mechanism of action, a.k.a. cabazitaxel. A wonderful study, again disclosure – I was part of it, but we saw data presented that suggests that their efforts to look at novel androgen signalling inhibitors yet. There were two abstracts presented; I apologise, I’ll have to look at my notes first because a name has not been assigned, it’s JT0918 of the new antiandrogen. There were two abstracts, one was a phase I that was done in Asia and then a phase II US experience where they are transitioning from prior abiraterone or enzalutamide treatment to this new agent and showing some initial data. My question to you is not to comment on the data but now we have four novel androgen signalling inhibitors and the space is getting crowded. I’m not the right person to comment because I said the same thing for darolutamide and I was a fool – darolutamide is alive, kicking and doing very well. What do you think? Is there more space for further refinement in the androgen signalling treatment space?

SO: Maybe I would say yes and no. Yes because regarding side effects maybe we could find a new drug which does not cross the blood brain barrier. As we know, maybe darolutamide is greatest, in fact, in not inducing these kinds of neurologic side effects compared to abiraterone or apalutamide. Maybe if we can find new drugs without any problems regarding the neurological problems so far it could be great and maybe it would be also greater at the androgen receptor level in terms of efficacy so far. So that could be also good points. But, nevertheless, it’s going to be difficult for the laboratory to go ahead and to put the drug as a good comparator later on because, as you said, the space is already completely crowded so far. So yes.

EE: It was more of a overview question because this specific agent, by the way, let’s comment on the fact that it looks like it doesn’t pass the brain barrier, it does have an adverse effect with regards to concomitant treatment with lipid lowering agents, with toxicity, liver function test elevation and the like. So we’ll look for more data on that one.
But I wanted to press the point about combinations in mCRPC that are still being pursued and especially with regards to that Holy Grail of finding the sweet spot where we can use immunotherapy to treat prostate cancer. We’re still failing to help a big percentage of our patients, even though we see those magic responses in the space of tumours with MSI or high TMB. So your comments, Stefan, since I’ve got you here, on what was presented and what the future holds from what we saw in GU ASCO?

SO: I think that regarding immunotherapy there are two studies so far – the KEYNOTE study and CheckMate study – looking at either pembrolizumab or nivolumab with docetaxel so far. But we don’t have yet the biomarker to select those patients because, as you said, it’s an elderly patient population with comorbidity and so on so we don’t want to add too much toxicity. In these two studies, if you look into the detail, you will see that we had some deaths due to the immunotherapy – three in the first one and two in the other one; so five patients died from the immunotherapy in this clinical trial. So immunotherapy to chemotherapy will add anything at the end because if you look at the data it doesn’t seem to be that really it’s outstanding in terms of efficacy, PSA response and so on. But I think that we do not select well those patients. You said that we should maybe select our patients based on the MSI status but, as you know, the MSI status is only 3% in prostate cancer so it’s going to be tough to select the right patients for those clinical trials. So now I think I’m not completely convinced about the fact that to combine and to see whether or not it’s better. But so far there are a lot of phase III trials ongoing and I’m not sure that the results will be positive data.

EE: True, true. It was really confusing to me to see so many efforts but a lot to expect in the future. I want to turn back to Boris because we said a phase III trial that was presented during this trial, ACIS, really the high point of the presentation with regard to clinical datasets. But, Boris, you mentioned earlier, before we started this meeting, that it would be a miss to not go back to the big expectations, the expectations on the therapeutic PSMA Lutetium. Would you like to comment on what was presented during this GU ASCO and are we closer? I know you guys have it and I turn to you because I know in Germany you have access but the rest of us don’t unless it’s within a trial. So what does the data tell us and how soon will we have something that’s in our clinic?

BH: Yes, I do share Stefan’s scepticism with regards to IO at the current time and I think with Lutetium PSMA we have this radioligand therapy that is really promising and that is used in third line systemic treatment in Germany on a routine basis because we have reimbursement. Now, at this year’s ASCO GU we have an update from Michael Hofman of the TheraP study, a study of men with mCRPC post-docetaxel. These men were randomised in a phase II setting to compare cabazitaxel versus Lutetium PSMA radioligand therapy. There was a good patient selection and that is important because all these men have had to be PSMA positive and, at the same time, they are not allowed for FDG mismatch. So if they are heterogeneous tumours that have high FDG signal but low PSMA signal these men were excluded. If you have this careful patient selection the data that was presented on Lutetium PSMA was really nice, showing a doubling in PSA reduction and progression free survival as compared to cabazitaxel which is a strong comparator. So I’m really happy but certainly we have to acknowledge that the OS data is not mature yet. This will probably be reported next year and the exciting theme of this year for Lutetium PSMA will be the VISION trial, a phase III trial where Lutetium PSMA serves as an add-on to ongoing therapy. This is powered for overall survival and I’m expecting it for ASCO. So I think this will broaden our armamentarium and I’m a strong believer that this therapy will have its place. I don’t know whether it’s third line mCRPC or in the future a little bit earlier. We have to wait for that but it’s definitely promising and it’s easy to handle if you have good nuclear medicine colleagues.

EE: We now know that you’re a little bit partial towards Luteium PSMA, we’ve got it. But, honestly, we’re all super-excited and you made it even more on the edge, waiting for the data. I’m going to turn now, because you did mention earlier, Boris, that it’s going to be important to get biomarkers and biomarkers are not just molecular assays, they could be more. They could be like what you described that Dr Hofman did. But I want to turn to Gerhardt because I know his life’s work has been pursuing biomarkers and get, Gerhardt, your thoughts on the dataset from nmCRPC to CRPC that was presented during GU ASCO and how it can further our efforts to refine biomarkers that are going to be predictive of outcome and maximisation of that therapeutic index across the board and where you think we’re going. Where is it going to transcend to reality that’s practical? Where is that? A lot of us don’t have access to these very, very, let’s say complex assays. What do you think and what do you expect?

GA: Molecular biomarkers, and as you said, Eleni, the focus is on prediction in CRPC. So should I use drug A instead of drug B? Or patient selection, and most of the idea for PARP is a treatment selection rather than A versus B. So as a temporary meeting, I guess for the past decade, multiplex tracks on biomarkers. We’re discussing probably four here, one that Stefan has already mentioned presented by Felix Feng on the SPARTAN trial. Then one from the IPATential trial presented by Johann de Bono and then two from the olaparib trial.

What comes to mind, the lesson to learn, is no biomarker is perfect and it’s a balance between ease of use, ease of reporting, cost and maximising the target population. There’s a number of settings of being in where we’ve discussed biomarkers and we’ve been really purist and we’re pushing down the target population to 1%. Then we think about how on earth are we going to run that trial, how are we going to fund it. So behind the scenes there is this balance on the spiralling upward cost as purists try to choose the perfect biomarker versus maximising efficacy or observing enough efficacy to have a positive trial.

Briefly on the SPARTAN data that Stefan has already talked about, identifying this hormone insensitive population. We’ve seen similar data from the TITAN trial using whole transcriptome, and again similar data in ACIS using whole transcriptome. They’re all in subgroups of these large trials so the analyses have primarily been in subgroups of the trials but they’re building a story. Increasingly, putting all these trials together, although none in isolation is to date clinically changing, the case will get there that we will start to introduce subgrouping of patients based on their transcriptome profile and probably this hormone insensitive group. So that’s the first piece which is the one furthest away but I really see the story developing.

The next is IPATential which, as I said earlier, is a beautiful human experiment, a beautiful trial, because it has proven what was previously observed in preclinical models that co-targeting of AR and PI3K/AKT is more effective than targeting of AR alone. That was done in a xenograft model or a genetically modified model. To take that into humans and prove or observe the same effect is a massive leap.

But the question is, what is the best biomarker and clearly you can see that balance I referred to earlier between a very pure biomarker that confirms activation of the pathway and maximum efficacy versus one that can be implemented. Showing there is a genomic change is possibly more robust than a protein change but you lose a lot of patients because of the challenges of performing next generation sequencing because not all modes of loss of PTEN occur through a genomic event. So in this reanalysis where there were multiple cut points for the immunohistochemistry test to define PTEN loss and there’s the use of NGS, basically the stricter you are, the better the hazard ratio, the greater the benefit for combining ipatasertib with abiraterone. It’s certainly food for thought.

These are all post hoc and they’re all exploratory and these meet the second challenge – no biomarker is perfect and how do we use data on an imperfect biomarker in our clinical practice? We see a positive result and we saw a positive result with the first report of IPATential and then we see subsequent reports where the biomarker is refined at a post hoc and exploratory, how do we use that data? It reminds me through what we’re going through or have been through with radiotherapy to the primary tumour which was based on bone scans and CT scans, low versus high volume. Probably an imperfect biomarker but we’re stuck with it because that’s what was predefined and now we’re stuck with that here and how do we break that mould, I guess, whilst keeping within statistical rigour?

Finally, again in this vein, the analysis of PROfound exploratory and looking at specific response rates by gene type where again the greatest benefit is for the BRCA groups. The benefit in ATM remains limited and hypothesis generation for other gene subtypes. I think we will get to the point in a few years’ time, once multiple trials of PARP inhibitors report, where we can then start combining these datasets and understand what the true response rate in a reasonably large group of PALB2 mutants, for example, is to PARP inhibition. I hope we will be able to do that.

The final abstract relates to ctDNA and whether we use tissue or plasma DNA. There is some discordance. So patients who were positive on tissue are not positive in plasma and that’s probably primarily to tumour shedding into blood or metastatic volume. So if there’s not enough tumour in plasma then one is not going to detect the aberration because there’s not enough of a tumour fraction to hit that cut-off. But overall I see high concordance. Personally, if I detect an aberration in plasma, especially BRCA which is what PARP inhibitors are licensed for currently in Europe, I would offer that patient a PARP inhibitor based on a BRCA mutation solely in plasma, not in tissue.

EE: Excellent overview. I’d like to add a couple of comments on your thoughts. First of all, we have this effort that you mentioned, this effort to increase the statistical significance of our findings. Sometimes it might end up leaving out some patients that would benefit. I had this experience through my recent presentation in the ASCO meeting where I was looking for that signature in the diagnostic biopsies and I realised that the more stringent you become, yes, you increase the specificity of your findings but you lose sensitivity, by definition. So we might end up not giving the opportunity to patients who would benefit. That is a very fine line that we’ll need in the future to address.
One more comment that was brought up by Boris, and I believe Stefan, is also the access to these specific assays. So, yes, immunohistochemistry is faulty but we have relied on it for so many other solid tumours, we should be able to deliver something to the community before we are able to move to what is a highly, highly sophisticated assay for now.
Final comment on this, you mentioned at the beginning the great value of actually having concordance between experimental conditions and the delivery of the human result of the clinical experiment. A big shout-out, I would say, is deserved to Charles Sawyers who did that first work and over and over we’ve seen him deliver across the board in oncology. One of my favourite, of course, scientists, I have to say.

So we don’t have a lot of time but I want to hear, back to Boris and Stefan, your thoughts on what was presented with regard to the value of androgen deprivation in the era of COVID and protection from infection or gravity of infection. There were even some efforts to provide ADT to men who were not patients to protect them in the beginning through the story of TMPRSS2 and the AFFINITY with the invasion of COVID. So, your thoughts especially in view that we’re getting great data now supporting that maybe even one shot of the Pfizer vaccine could be protective as high as 92%. Do you think there’s value to think that our patients with ADT alone are protected?

BH: From a urology perspective we were all excited with the Italian data last year that suggested that ADT might be protective. With the data now at ASCO which has limitations but the men on ADT cannot feel safe. So the normal advice to behave safe has to stay there. So what we learned in these COVID times are a couple of things: for localised disease it’s uncritical to delay curative intended radiotherapy or surgery for a couple of months. The other part right now is that we really have to encourage people to get vaccinated. There’s a lot of discussion in Germany right now with the different agents and my recommendation is whatever is offered to you get it because it’s really a huge improvement against not being vaccinated. So in Germany cancer patients get prioritised a little bit, they get vaccinated earlier, so this is the chance for our patients. But with regards to ADT it does not change any of my recommendations; if a man needs ADT he gets it and if he doesn’t need it he doesn’t get it.

EE: Yes, less is more there. Stefan, your thoughts?

SO: Yes, we had this previous article showing that ADT could prevent the appearance or the severe COVID-19 symptoms for the patients. So far in these three abstracts they are all negative regarding the activity of TMPRSS2 and inhibition with ADT. So we need to take into account these three abstracts that they had some guesses so far based on comorbidity and mitigation as well because we know that for various people, for people having high blood pressure and so on, and diabetes maybe, we need to see if the group are well balanced in terms of characteristics so far. So we need to see. Maybe the answer is not yet completely done. We need also to look at the testosterone level of those patients as well. But there is a rationale, to my mind, to maybe continue to go further because if we can block the TMP pathway really it’s useful for prostate cancer as well as for COVID-19. So the hypothesis is really interesting to look at.

EE: Absolutely. So an exciting year ahead. Hopefully we’re going to be able to interact in person by at least ESMO. I’m just hopeful, you know, even though sometimes it becomes a little bit more dire than what we would want. Thank you so much for sharing your thoughts. Thank you all for watching us and to more data that is very exciting and will change the lives of our patients. Thank you very much.