AstraZeneca PLC (AZN) Earnings Call Transcript & Summary

Welcome, ladies and gentlemen, to AstraZeneca's Investor Conference Call ASCO GU 2022. Before I hand over to AstraZeneca, I'd like to read the safe harbor statement. The company intends to utilize the safe harbor provisions of the United States Private Securities Litigation Reform Act of 1995. Participants on this call may make forward-looking statements with respect to the operations and financial performance of AstraZeneca. Although we believe our expectations are based on reasonable assumptions, by their very nature, forward-looking statements involve risks and uncertainties and may be influenced by factors that could cause actual results to differ materially from those expressed or implied by these forward-looking statements. Any forward-looking statements made on this call reflect the knowledge and information available at the time of this call. The company undertakes no obligation to update forward-looking statements. Please also carefully review the forward-looking statements disclaimer in the slide deck that accompanies this conference call and webcast. There will be an opportunity to ask questions after today's presentations. [Operator Instructions] And with that, I will now hand the call over to AstraZeneca.

Thank you, Martin. Hello, and welcome to this Investor Science Conference Call. I'm Tom Waldron from the Investor Relations team at AstraZeneca. We're delighted to be hosting this event to discuss the exciting data that was presented at this year's ASCO GU Cancer Symposium. Could we have Slide 3, please? I'm very pleased to say that we're joined by Professor Fred Saad, the Principal Investigator on the PROpel trial, Professor and Chief of Urology at the University of Montreal Hospital Center, who will give us highlights from his oral presentation last week. And from AstraZeneca, we'll hear from Susan Galbraith, Executive Vice President of Oncology R&D; and Dave Fredrickson, EVP of the Oncology Business Unit. For Q&A, we'll be joined by Sunil Verma, SVP and Global Head of Oncology Medical as well as Andy Barnett, Global Franchise Head of GU and GYN Cancers, DDR and Established Oncology. Slide 4, please. Here's the agenda for today's call. Susan will start with an overview of ASCO GU 2022 and touch on the scientific foundations of PROpel. And Dr. Fred Saad will go through the PROpel data that he presented last week, followed by Dave, who will put PROpel into its commercial context. Finally, Susan will briefly cover some of the other data presented at GU before we move to a Q&A session that will be chaired by Dave. As we move to the next slide, I'll now pass you over to Susan. Susan?

Thank you, Tom. Please go to the next slide. So this slide shows our oncology R&D strategy. On the left-hand side, you can see the drugs that we already have approved, of which Lynparza olaparib is the focus of today's call. And on the right-hand side, you can see that our strategy is to have a diverse pipeline of different mechanisms that have the potential for combination with non-overlapping resistance and the ability to kill cancer cells in different ways. The theme of combination is one that is manifest in the ASCO GU conference. If you can go to the next slide, please. So we had a good presence at ASCO GU, 19 abstracts and 3 oral presentations. Obviously, the PROpel data is the key focus of today's call, and I'll also briefly later go on and describe the high-level results from Lynparza and Imfinzi in the BAYOU Phase II trial in urothelial cancer and Enhertu plus nivolumab also in urothelial cancer that are HER2+. Please go to the next slide. So the PROpel data and the PROpel study design was based on success from Study 08, which is a randomized Phase II study, again, led by Dr. Saad, which showed an improvement with a hazard ratio of 0.65 in rPFS as the primary endpoint looking at the combination of olaparib and abiraterone versus abiraterone alone in this castrate-resistant prostate cancer setting. And again, in that study, we saw a significant rPFS benefit regardless of HRRm status. The pre-clinical science underpinning this is also shown on the right-hand side. We understand from data that's already been published that PARP inhibition involves a reduction in androgen-receptor-dependent transcription. And therefore, PARP inhibition may increase the activity of novel hormonal agents like abiraterone. In addition, we also understand that novel hormonal agents can induce relative deficiency in homologous recombination repair, thereby increasing susceptibility to PARP inhibition. This is the basis for understanding why there might be a benefit more broadly than the HRRm alone population as has been seen in Study 08 and also in PROpel. And with that as background, I would now like to hand over to Dr. Saad to go through the PROpel data.

Thank you very much. So I'm very happy to be able to be here and to share some of my thoughts and to share with you the data that I presented last week. So just very briefly, I am the Head of Urology, the Chairman of Urology at the University of Montreal. But I'm also the Chairman of Genitourinary Oncology, and prostate cancer is my focus, and I -- in dedicated prostate cancer clinics that I've been running for almost 30 years, we now see 30 to 40 metastatic patients per week. And so metastatic CRPC is obviously a huge focus of our clinical and our research in our center. Next slide. So very briefly, I think you're well aware, PROpel is a global randomized Phase III study that built on Study 08 that was mentioned. So these are patients that are first-line mCRPC. And when we speak about first line, they could have received docetaxel in the metastatic hormone-sensitive setting, which is a widely used alternative given the fact that its prolonged survival and was in the same range as NHA in combination with hormonal therapy in patients with mHSPC. We disallowed prior abiraterone for reasons that I think we now believe were the right -- was the right strategy. We insisted an ongoing ADT, and we stratified according to site of metastases, so we allowed patients to come in with visceral metastases. It was really an all-comer mCRPC population. And when you -- when we look at the demographics, it really does represent what we see in the clinic of patients with mCRPC. We also stratified as to whether or not they got taxanes in the mHSPC setting because from experience, if you receive a taxane in the mHSPC setting, you clearly don't do as well when you become castration-resistant because it's an extremely effective drug to delay castration resistance. And there were patients who were randomized 1:1 between olaparib at full dose [indiscernible] on the full dose, which I think is going to be relevant for a discussion later on. And abiraterone full dose compared to placebo plus abiraterone, with almost 400 patients in each group. And the primary endpoint was radiographic progression-free survival or death by investigator assessment, which we now strongly believe is an excellent surrogate, an approvable surrogate for overall survival, given all the work that's been done in this area to try to find a surrogate that can lead us to new therapies without always waiting for overall survival. Obviously, overall survival was an important secondary endpoint and other secondary endpoints are -- and additional endpoints were studied, including time to second progression, time to subsequent therapy, objective response rates. But also importantly, since we took an all-comer population of mCRPC, we have the opportunity to go and look retrospectively to what extent patients harbored mutations versus no mutations in terms of homologous recombinant repair. We looked at health-related quality of life and the safety tolerability, clearly, in this patient population, is essential. So really a very important study that was designed to really address the question of whether or not this is relevant for the first line mCRPC patient. So next slide. So as I said, baseline characteristics, patients came in generally feeling quite well. We did have patients over 20%, almost 25% in the treatment arm that were symptomatic. And when we talk about symptomatic, these patients could have been on narcotics, could have had pain scores that were high. And these patients are often excluded from other studies. We had patients coming in with visceral disease, and we had patients -- 22% of patients that actually got docetaxel in the metastatic hormone-sensitive setting, which, again, reflects real world. Extremely important is this is one of the first, if not, the first study to let us understand the prevalence of mutations in an all-comer population. And about 28% had mutations of any sort, not enriched for any specific mutation and almost 70% harbored no mutation that we were able to detect on tissue and ctDNA. So when tissue is not available, ctDNA allowed us to be able to classify patients. Next slide. So the primary endpoint was, as PI of this study, surpassed any of my expectations. We reduced the risk of progression or death by 34%. In terms of months, it turned out to be an improvement of 8.2 months over pretty much the standard of care around the world of abiraterone. What's amazing is that 8.2 months is exactly the same as the 8.2 months that was observed in the study that I helped to lead also of abiraterone against the placebo. So abiraterone against the placebo improved rPFS by 8.2 months. And here, we were able to further improve another 8.2 months over a very active control. So truly, one of the most outstanding, if not, the most outstanding improvement in rPFS against a truly active controller. Next slide. Surprisingly, again, when we go to independent central review of imaging, this improvement of rPFS turned out to actually be 11.2 months. So clearly, investigators were not overestimating the benefits. If anything, they were clearly underestimating the benefits of the combination therapy. Next slide. When we look at the subgroups, when we dream of results for a clinical trial, these are the kind of things we dream about. Having pretty much every endpoint of subgroups being clearly on the side below 1, which is non-significant. So we are clearly within that range even considering the confidence intervals, whether or not they had visceral disease, whether or not they had received docetaxel prior to starting their therapy for mCRPC and especially important whether or not the harbored HRRm mutations. So with mutations, 0.5 hazard ratio, this is all mutations, not only BRCA and non-HRR 0.76 as the hazard ratio is really outstanding. Next slide. In terms of overall survival, the good thing is patients are not dying any faster than we would have expected, given that we had a control arm that was effective. We didn't expect the curves to separate earlier than this, and the curves are separating in a trend that we think is very encouraging, but still too immature to talk about significance. Next slide. We do have 2 other endpoints though, that in other studies that I've been involved in, have helped us to predict that we are making a significant difference. So patients went on to subsequent therapy much later in the combination arm. This was very significant. And importantly, time to second progression. So we followed patients as they went on to subsequent therapy and had a second progression. And here, you see that the upfront combination has significantly better outcome than the patients who started on abiraterone alone, giving us the impression that you can't catch up if you don't start upfront. And this was clearly shown in another study that I worked on where [ rPFS 2 ] or [ PFS 2 ] ended up being pretty much exactly the advantage of overall survival in the final analysis of the study. So these are extremely encouraging results that help us to make sure that our immature survival results really are going in the right direction, and help us to further our conviction that the rPFS is not a unique finding, but really in line with everything else. Next slide. In terms of safety, this is extremely important. Patients do very well on abiraterone alone. Clearly, there were no added deaths due to adverse events with the combination over abiraterone alone. There were slightly more patients that discontinued olaparib. You have to remember; this is a full dose olaparib. But clearly, some patients will have to stop, but this was not to any significant degree and discontinuation of abiraterone was not affected by the addition of olaparib. Next slide. In terms of cardiac and thromboembolic adverse events, there was a signal that maybe there was a cardiac issue in the small study [ 08 ] of only 140 patients. This difference in cardiac failure or other cardiac events was not at all seen in this study. In terms of arterial thrombotic events, there was no difference. In terms of embolic -- thromboembolic events, in terms of venous events, there was more numerically -- numeric patients having pulmonary embolis. These were, by far, almost all incidental findings on repeat standard-of-care CT scans and for the vast majority of patients did not need to study discontinuation or discontinuation of drug. Next slide. So the adverse events that we expected with drugs like abiraterone or olaparib for what we saw, it was very reassuring that we didn't see more adverse events because of the combination. The rate of Grade 3/4 anemia was actually a bit lower than I would have expected based on what we knew in the past, with 15% of patients getting a Grade 3/4 anemia, which is quite easy to treat and all of the other adverse events that are related to PARP inhibition or abiraterone were very much in the range that we would have expected and quite low, which might reflect a better patient population coming in, in first line rather than waiting for second or third line to introduce these kind of combinations. Next slide. So I'd be very -- looking forward to the discussion. But clearly, olaparib plus abiraterone led to not only a statistically significant improvement, but clearly, clinically significant and meaningful improvement of rPFS. And this is the first of any combination that's led to a radiographic progression-free survival that goes beyond 2 years. This is unheard of in first line mCRPC and this benefit was observed irrespective of HRRm mutational status. Quite reassuringly, secondary and all the exploratory endpoints also support the treatment benefit of our primary endpoint in the overall patient population, and the safety profile was consistent with the safety profile of each individual drug and there didn't seem to be an added effect in terms of adverse events by combining these drugs. And in terms of quality of life, we looked at this. There was no detriment in quality of life by combining these 2 drugs together, allowing most patients to stay on therapy. So as a clinician first before being a researcher, I think this is extremely important data for our patients. And I really believe that this should and will become another therapeutic option in the first line setting for patients with mCRPC. There are complexities with testing. So many, many, many treating physicians are looking for ways to be able to do a better job without having roadblocks that disallow us from doing a better job for patients with first-line mCRPC. So thank you very much.

Thank you, Dr. Saad. Much appreciated. Now, what I'd like to do is talk about, first, the unmet need that really exists in mCRPC and the opportunity to bring this combination to patients around the globe and our view on how we're thinking through that. If you could go to the next slide. Very importantly, and we've talked much about this. We know prostate cancer is the second most common cancer in male patients. You heard Dr. Saad speak about the volume of metastatic patients that he sees even within his institutions. We also know that therapies are limited once patients become mCRPC and that, as I've noted here, they're mostly monotherapy. I think what's very important is the first point that I want to highlight here is that the urgency and the unmet need really does change as you move from left to right on this particular graph. And while there are many men with hormone-sensitive prostate cancer who can and are managed with ADT or indeed with ADT combos. When they become mCRPC, the survival rates plummet rather significantly. The prognosis worsens quite a lot and the need for better treatments is really quite clear. I think also one of the things that's important to note here is, of course, we do see that with treatment intensification strategies that there is an increase in NHA use that's happening upstream of mCRPC, but I think it's equally important to note that there are quite a large number of patients that are not being treated with NHAs in these earlier settings. A number of them being treated with chemotherapies or with -- or excuse me, ADT by itself who ultimately, sadly, flow through into this metastatic side. If we can go -- one other point that I'd like to make on this is that we note here that about 50% are NHA naive at mCRPC and therefore, about 50% are NHA experienced. This is really probably more of a U.S. figure. I think, across the globe, we see that the NHA naive at metastatic is indeed higher in other parts of the globe due to access and other particular issues around cost. Next slide, please. So I won't repeat all of the elements that Dr. Saad went through, but I think that it's very important to underscore that, first and foremost, outcomes are poor. The sense of urgency is high. 30% is the 5-year survival rate for patients with metastatic disease, 3 years is the median OS. And so I do think within this context, it's really critical for us to make sure that we see PROpel as really important for this stage of disease. We build on the success of PROfound, I think that it was very well stated by Dr. Saad as well. Representative of a real-world population, pretty straightforward and simple trial design, having the ability to really demonstrate this in all-comers and to know that, that has been really solidified by being able though to also see retrospectively the benefit within the HRRm and within the HRR-negative population or the wild-type population within this. Radiographic progression-free response, as Dr. Saad said, a clinically relevant endpoint in and of itself also reinforced by important secondary endpoints [ that give ] confidence within this. And we think that, lastly, I'd like to just note that I think that the tolerability at the full dose, together with the maintenance of the quality of life, are really key contributors to the efficacy results that we were able to see within this. We'll go to the next slide. So just to put graphically what Dr. Saad shared in terms of how PROpel has built off of what we were able to accomplish, or the field was able to accomplish in 2012 with abiraterone, you can see here building to now over 2 years of median radiographic progression-free response with the combination, and that's against an active control that is a relevant and innovative therapy today. And I think that as you move into the chart that sits at the bottom, you can also really see here, whether it's the HRRm or the non-HRRm. We know that the median for the combination is above 24 months, and the entire population is above 24 months. So I think that you can look at that and have good confidence here that the non-HRRm population for the first time getting over that 2 years. And we, of course, would expect to see that since we haven't gotten to the median yet for the mutated population that that's going to do even better still in terms of performance. And then last slide, if we could. Just, again, if we put this in a summary and think about how we kind of within the U.S. context, map out how we're thinking about, again, the patients who might be able to benefit from this. We know that about half of patients come into frontline mCRPC, NHA naive, there's about half that have prior NHA use. You can see that then the majority of the NHA naive do go on to receive an NHA within this setting. You do see some rechallenge even happening again within those that had prior NHA use. PROpel did allow for prior NHA use if patients had been 12 months away from that NHA, and so overall, there's 65% of frontline patients more or less that are receiving an NHA today. And so obviously, we're in discussion with health authorities on the exact aspects of how the label will read and we won't know that until those discussions take place. We know that within this, clearly, Lynparza and abiraterone demonstrate a clinical benefit versus abiraterone alone in those patients who are NHA naive. That's very evident. Also for those who've been experienced, we think that it's quite reasonable for Lynparza and abiraterone to be seen as a well-tolerated chemo-free treatment option. And as we think about the extent to which those patients will be receiving the combination, again, it's really kind of within those that are NHA naive and a portion of those that are prior NHA that are receiving an NHA where it might be considered. So with that, I turn it over to Susan.

Thank you, Dave. So if you can go to the next slide, I just want to highlight some of the results -- top line results from 2 other studies that were presented at ASCO GU. ON the left-hand side, you've got the Imfinzi plus Lynparza combination, which was reported from the Phase II BAYOU trial in platinum-ineligible patients with metastatic urothelial carcinoma. This was a randomized study in around 150 patients -- 155 patients. And what we were looking at was the addition of olaparib to a background of durvalumab in this setting. Overall, the hazard ratio was 0.94, but interestingly, in around 20% to 22% of the patients that with HRRm mutation in this patient population. And within that group, the median PFS was different between the 2 groups with a numerical estimate of 5.6 months in the combination arm versus 1.8 months in the durvalumab plus placebo arm with a hazard ratio of 0.18. So I think that data does suggest a role for PARP inhibition in the HRRm subgroup of urothelial carcinoma, which is quite a significant group. On the right-hand side, you have the data of Enhertu in combination with the PD-1 antibody nivolumab in the Phase II U105 trial, again, in metastatic urothelial carcinoma, this time in a group of patients that are HER2+. And again, what you can see is an encouraging response rates, and the spider plot here shows that those responses were deep and durable in this setting. And thus, this is something that's also included in the DESTINY-PanTumor02 study. So with that, I just want to thank you again for your time, and then we'll now move to Q&A, which Dave is going to moderate.

Thank you, Susan, and thanks to all the presenters. We'll move now into the Q&A. One request, please, that I have out of probably more than anything else, respect for Dr. Saad's time, who's with us for the next half hour is, while there is enthusiasm among all of us for the [ DBO4 ] data, which we had announcements on. If we could keep today's conversations, please focused in on the PROpel study in prostate cancer. That would be great. [Operator Instructions] And the first question we have is with Simon Baker.

Dave. Two quick questions, if I may. Firstly, for Dr. Saad. It doesn't look like there are any patients within this setting where you wouldn't use this combination. But I just wanted to get your perspectives on that, if there were any patients where -- where you would possibly be less willing to use this combination? And secondly, a question probably for Susan. Given the safety profile we've seen here, which looks pretty good, what's the scope for adding additional therapies? I was thinking particularly of your ATR inhibitor, ceralasertib. Would it be possible to add that in with -- while maintaining a reasonable tolerability?

So I guess the first question is a very good question. Who would I not treat rather than who would I treat because you're -- I think the perception that this looks like it's benefiting all patients that we put on the trial, I think, it's correct?. I think we didn't see a signal of patients who are not getting benefit. Obviously, in the real world, there are patients, sometimes patient-driven reasons why we wouldn't do a certain combination or intensification of treatment. Some patients are quite frail. Some patients don't have a good bone marrow. Some patients have severe cardiac issues. I would say that probably if patients have a contraindication to abiraterone, that would be one of the major factors, right? If they have severe cardiac failure that precludes them from getting abiraterone, well, obviously, this combination is no longer feasible in those patients. But patients that are in average health or decent bone marrow function, you're perfectly correct. I mean, these patients, on average, don't do very well for very long with metastatic CRPC. The median time to PSA progression is constantly within less than a year and radiographic progression, very much in line with what we saw here around 16 months, between 14 and 18 months on average. So clearly, our biggest opportunity is upfront to make an impact down the road. And add to that the fact that many patients don't go on to chemotherapy or to subsequent lines of therapy in the real world, sometimes our only opportunity is the first line. We'd all like to believe that patients all get 4 or 5 lines of therapy, but that's not the reality in prostate cancer. Less than half in most studies would suggest even less than 40% get anything beyond first-line therapy. And so that is the reason we're working very hard in improving the first-line options, which are -- which is by far our biggest bang for our buck.

Thank you, Dr. Saad. And just to answer the second question. We are looking at the opportunity for ATR inhibitors in a couple of places. One is in patients with ATM mutations and the other is in combination with olaparib in the context of PARP resistance, which we're looking at in -- initially in ovarian cancer. But obviously, this is going to be a very important question within prostate cancer as well. So that is a combination we will absolutely continue to look at. We'll need to see the initial data from those studies in order to work out whether or not we could then add it to this kind of setting.

Thank you, Susan. I'm going to go to a question on the webcast and maybe first pose it to Dr. Saad, but then invite if either Susan or Sunil want to complement to that to please do so. Andrew Berens asks, can you give some color on why PARP inhibitors should complement androgen inhibitors in patients that are HRR negative? What's the physiological rationale for the combination?

Yes. So that's -- I think this is important because this is the basis of Study 08. And clearly, what led to PROpel is that in the lab, we've clearly had indications that our PARP inhibitor has other mechanism of action that are beneficial in prostate cancer, mainly we think, through AR transcription. And on the flip side, the NHT really does seem to cause changes in the cancer in terms of susceptibility to PARP inhibitors even in the absence of an HRR mutation. So that was really the basis and needed to be proven, at least in a Phase II study, and it was in Study 08. And obviously, we needed the Phase III to confirm that this was -- went beyond the lab and a few mice but actually has this impact in humans. And Sunil can build on...

Thank you, Fred. If I can just -- I think what was really promising that we saw in Study 08, as you noted, was a hazard ratio of 0.65 or so. And as you brought forward in the PROpel, we saw confirmation with exact same hazard ratio. So I think the hypothesis that stemmed from Study 08 was really confirmed in this trial, giving us the proof of concept is valid. There is activity that is seen across the spectrum including HRR negative patients. Susan, I would love to get your thoughts on this.

Well, yes. No, I think it's been said. So there's published data on the mechanism here, and I think relative HRR deficiency can be induced by a variety of other settings. We've seen that in multiple settings. Typically, the level of sensitivity and sensitization isn't as great as you get in a BRCA-mutant population, for example. So the data that you're seeing here clinically is very consistent with those pre-clinical observations, not just, in fact, that you see this effect, but the size of the effect is not expected to be as big in the HRR negative group even when you've got this relative sensitization. So I think all of the data in totality are coherent.

And it clearly, it seems to be -- dosing seems to be important for those that don't harbor mutations because this synergy needs, I think, even more intensified treatment than in patients who are on the other extreme, BRCA-mutated patients where maybe sprinkling a little bit of PARP inhibition is enough to at least have a [ sub ] effect. And so clearly, I think this is what we're seeing here.

I mean, I think to that point, in summary, it's important to note that the first element around synergy is a reason to believe why in prostate cancer, we're seeing activity in an all-comers population, whereas in other tumor types, we have not. The second point around the dose intensity may prove relevant if we look at why all PARPs aren't able to demonstrate this [ type of ] benefit even within prostate cancer. And I think that's an important point. Next, going on to James Gordon at JPMorgan.

James Gordon, JPMorgan. I'll resist asking a deeper question then. So I will ask about prostate cancer. So the PROpel data look really good, but only addressing half of the castration patients and a big patent expiry in [ 2018 ]. So when do you think you might -- the [indiscernible] follow-on part approved in prostate cancer? And is the plan to be a lot broader with that and also to go earlier? When will we get more visibility on sort of the longer-term plan for prostate cancer [indiscernible]?

Susan?

Yes, yes. James. And I appreciate your restraint from [ DBO4 ]. So it's a great question. I do think that the data that we've got -- that we've seen today gives us the opportunity to think about other settings where we might want to use PARP inhibition. Obviously, the data on Part 1 is selective, we hope to update you on how Phase 1 is going at a congress later this year. And when we're ready to disclose what we're doing with that in the CDP, I'd be happy to discuss that. But it's a little early to lay out the entire CDP, clinical development plan that is today, so bear with us.

Great. Thank you, Susan. Next, we'll go to Luisa Hector at Berenberg. Luisa, please.

Thank you, Dave. So I just wonder if you can comment on how that population of prior NHA treated, I realize it's a bit skewed to the U.S., but how that might change in the next 3 years? Could we see more patients arriving with prior NHA? And any color you can give us from PROpel between those 2 cohorts since you did have some prior NHA, I'm not sure if you can really comment on any differences between the efficacy for naive versus prior treated?

Great. So maybe why don't I propose, Dr. Saad, that if you can start first with your perspective on changing treatment landscape upstream of the front line mCRPC. And then also, I think the second question on what we know about prior NHA within PROpel as fairly straightforward. And then Sunil, again, if you want to build on the answers.

Yes. So I can speak to the practice in Canada because we've done some real-world evidence in Canada. And I can speak to my individual center where we're seeing patients on a daily basis coming in with mCRPC naive of NHA. So in the real world, and I think sometimes we get the impression that it's all patients getting NHA when they're hormone-sensitive metastatic. The reality is it's far from that number. Even within Canada where access is free and especially in many other countries, and I would include the States in that. So right now, there's an issue that there are a lot of patients that are progressing on ADT with metastatic hormone-sensitive disease have never got an NHA. We've got patients that are doing well on ADT for metastatic hormone-sensitive disease, where there's no indication to layer in an NHA. So there are a lot of patients right now sitting around, and this entity won't disappear anywhere in the near future. I don't predict that there's going to be a huge shift before 4 or 5 years around the world. And as I said, there are many countries where NHA is just too expensive to introduce in the hormone-sensitive setting, and so many are opting for chemotherapy, which is 6 cycles, generic docetaxel and as effective in terms of improving overall survival when it's introduced in the hormone sensitive. So -- and we included those patients. So I think you're perfectly correct. If cost issue, access issue goes down tremendously, the percent of patients treated in the hormone-sensitive setting will go up, but it's going to still take several years before there's a huge shift in what we're seeing today of patients walking in with mCRPC naive of NHA. Obviously, those that are already NHA exposed wouldn't be candidates for this protocol. We have to accept it. But we -- but clearly, there are a lot of patients that will benefit within the States and especially around the world.

Great. Dr. Saad, if I can just bring and highlight just 2 [ bells ] on your [ excellent ] summary. One is the consideration of treatment decision is really driven by the type of specialists that the patients are seeing. And also, I think, as you mentioned, the regional variations. And Dr. Saad himself as a urologist, and -- but we do know that medical oncologists play an important role also in the care for patients with metastatic castration-resistant prostate cancer. And there are more likely to, I think, consider combinational therapies. And in the earlier hormone-sensitive, there's probably a higher [indiscernible] of specialists such as urologists and radiation oncologists treating these patients [ wherein dose ] intensification or treatment intensification is not as generally considered. And then the final comment on regional variations is certainly, we do note that the prior NHA use varies as low as potentially 10% in certain parts of the world, including certain parts of Europe, up to 50% to 60%, as we have noted here in North America. So there are some regional variations to consider. But I think the type of specialist also plays a big factor here.

Great. Thank you very much. Next, we'll go on to Peter Welford at Jefferies.

I just wanted to follow on actually on the [indiscernible] comment that was made just with regards to -- if we think about the community setting in particular, I wonder if you can give us some idea of what proportion of prostate cancer patients are treated in a community setting still, particularly when we get to the metastatic castration resistance stage? Because I imagine it's quite a high proportion initially. And if you just talk about as well that in this setting, perhaps it's a question for Dr. Saad. How often currently in prostate cancer do you either biopsy or in the case of prostatectomy then do a genetic scan or map of the tumor currently in this cancer setting? Or is it still relatively immature given obviously the right targeted therapies that are really in this indication?

Right. So -- So you're perfectly right. In terms of patients treated in the community, I think there are a lot of patients. And this, I think, was what Sunil was alluding to is that in the community, a lot of patients are being treated for hormone-sensitive, even metastatic disease. They're not routinely sent because patients do well. You're giving ADT. If PSA goes down, patient feels well. There's no sense of urgency. This changes when they become mCRPC. Many community practitioners start to feel uncomfortable once they face a patient with metastatic CRPC where the complications, morbidity and especially mortality sets in. So there's more of a likelihood they will be referred or kept in the community by people that are more comfortable in treating where there's a sense of urgency that sets in when they become castration resistant. And so I think even though there is less of an urgency to do intensification in the hormone-sensitive setting, I think the intensification of the castration-resistant is clearly there and people are actually waiting for ways to do better than what we're doing now. In terms of biopsy and doing genomic testing, it is dismal around the world. I think many, if not most, academic centers in the States are doing active testing are doing an excellent job, but it's definitely not the real world. When I speak to people all around the world, and even within Canada, we have a very hard time doing genomic testing, even though we have agents in the post NHA setting approved, the biggest issue is we're unable to do routine testing to give access to these therapies to our patients that are mCRPC and failing NHA, where there's a real sense of urgency, we're unable to treat them because of their limitations getting access and the complexities of doing testing. So I think this is one of the reasons people are very enthusiastic in having an opportunity to do better without the hurdles of testing that even if you are able to get testing, it's not the easiest thing. And if you limit it to tissue, at least 30% are non-informative. So there are a lot of issues around testing and even with the results of testing, how to move forward. So this is unfortunately still the reality in a country like Canada, we're having a hard time getting testing, and we realize it's less than 10% of patients that should be tested that are being tested in Canada. I'm speaking for a country where we have a public health care system that I think is excellent, but in areas of testing, it's still a huge problem.

Andy, I think it'd be useful also. Could you just share some of what we see within our analytics as it relates to patients being treated in the community, particularly in the mCRPC setting? And then maybe also we can share what we're seeing around testing rates as well just from our work that we're doing within PROfound.

Yes. Thanks, Dave. When you think about metastatic CRPC, we focus on the U.S. Only about 20% of patients are managed by urologists, about 80% are managed by oncologists. It's definitely a referral point when the disease becomes severe and reaches a metastatic CRPC state. That varies somewhat around the world, but I think it's -- you can generalize it to say that once it's in this state, it's the larger community practices of urology or oncologists that are typically managing the disease in the U.S. And around the world, urologists may play a more significant role in managing prostate cancer overall. So it does vary from country to country, but that gives you a sense of the perspective of how we're thinking about building our teams to interact with these people. I think it's also worth just flagging because another question came up about prior NHA-treated patients in PROpel. We did allow prior NHA but needed to be a washout. And as a result, we didn't recruit many patients into the trial. It does, however, raise the question, what does a physician do once the patient reaches this state that has progressed on an NHA? There's an absence of treatment options. And as a result, I think, where funding permits, we may -- it's reasonable to assume we may see some spontaneous use.

Thanks, Andy. Much appreciated. Now let's move on to Seamus Fernandez at Guggenheim. Seamus, please go ahead.

So 2 questions. Can you guys help us understand the mix of BRCA versus the overall HRR patient subsets? I know that was something of a question as it relates to the discussion there. And maybe as a follow-on to that, can you just help us understand the correlation of ctDNA versus tumor -- specific tumor testing of the HRR status? This was, I think, a little bit of a confusing controversy on our side at ASCO GU. I'd love to just kind of get the physicians thought on this largely because when I looked at the control arm between the 2 studies, MAGNITUDE and PROpel, the control arm performed almost identically in the HRR patient population. So just trying to understand what is the controversy here? And why was the community so, I guess, seemingly been out of shape over what was discussed. And then just as a quick follow-up. Your team -- so Dave, the team is talking quite a lot about differences between the U.S. and international kind of treatment standards. Can you guys just update us on where you stand with regard to a U.S. filing? Is there something unique or different required to file in the U.S.? And are there issues as it relates to being able to file in the U.S. that perhaps we don't quite understand yet?

Perfect. Susan, would you like to start first with -- I guess there's a couple of questions that are in here, right? There's a mix of BRCA versus non-BRCA but HRRm, there's ctDNA and tumor, and then there's a filing question.

Sure. Thanks for the question, Seamus. So just in terms of the context. Obviously, the proportion of HRRm patients in PROpel was in line with what we would expect in the real world about -- which is at 25% to 30%, and that's exactly what we saw. Within that, based on real-world data, you would expect BRCA mutant to form about 1/3 of the HRR mutant patient population, which is in line with what we saw. So again, I think when you're thinking about the effect size that we saw in the HRRm, right, you can do the math and work out that whilst BRCA mutant is likely to be an important component, it's unlikely to drive the totality of the effect that we saw in that group. So that's the first point that I would make. The second thing in terms of testing, just for clarification, I'm not going to go into the controversy, maybe Sunil or Dr. Saad would like to do that. But just for clarity about what we did, 98% of the patients that were enrolled in PROpel provided both tumor tissue and plasma ctDNA. Of the tissue samples, around 70% are valuable, again, entirely in line and consistent with prior data. So only 30% of the non-HRRm would be classified as such on the basis of ctDNA alone. Based on concordance data, again, which are published, you would expect only 6% false negative rates on ctDNA. And because it's only 30% of patients, you therefore expect a total number of maximum of around about 2% of the non-HRRm likely false negative, which actually represents 12 patients with the 552 in the non-HRRm group. So we do not believe that this could possibly be driving the effect size seen in the non-HRRm population. I hope that clarifies that discussion. But Dr. Saad or Sunil, if you want to make any comments on the controversy question.

Yes. Maybe I can take a stab at it. So it was a really a spirited discussion at ASCO GU. So there were 2 elements I think that led to a bit of confusion and Susan, you walked us through it. So one was, how is the testing done? Because I think the abstract just looked at the ctDNA and there was an assumption it was just ctDNA only. But it was, as Susan just walked us through, it was tumor tissue and ctDNA. So I think that was one element of the controversy. And the second element was related to frequency of the BRCA mutated positive patients. In the table that was presented where we discussed at the numbers which were quoted between 12 to 20, it's actually, as you would expect, around the 1/3 of these patients, and we would see that number. And then the third controversy was really driven by different prognosis assumptions. And really the -- as you noted, Seamus, our outcome in the HRRm group was exactly the same of 13.9 months or so. There was no different. It was an accurate number that was included in one of the slides that led to the controversy, but it was exactly the same number as we've seen in the other study. Fred, any other comments?

No, I think it's just important to emphasize that the testing was exactly the same as in the other study that was presented after me. So we always try to do tissue testing. The reality is in every study that's done. For many reasons, tissue is noninformative, either too old, not cellular -- not enough cellularity, for a lot of reasons, about 30%. And now the standard of care is when tissue is unavailable or uninformative to go with ctDNA, and that's what we've done in this study. So the combination of both allowed us to have a status for 98% of patients and as was emphasized, clearly, the potential of some patients being -- not diagnosed as being mutated in the non-mutated subgroup is minimal and is impossible to be. And this is with no enriching at all. I mean, when you're doing a study where you're actively looking for mutated patients, you can artificially inflate the percentage of patients with BRCA mutations, which was not done here, and I think really reflects the benefit in the real world.

Thank you, Dr. Saad. I'm going to now move to a question on the webcast from Sarita Kapila from Morgan Stanley, and this will be, Susan, a question for you. Is there scope to combine next-gen ADC [indiscernible] with PARP inhibitors? Is there a synergy between [indiscernible] warhead and Lynparza?

Thanks, Dave. Just before I go into that, let me answer Seamus's last question about the U.S. filing. We're obviously working with regulatory authorities around the world, and we do intend to file globally. So I don't think there's a particular challenge there. So to move on to this question, absolutely. Thank you for the question about the [indiscernible], which is our first of our internal antibody-drug conjugates with our own proprietary [indiscernible] warhead and proprietary linker. It's targeting B7-H4, and that is a target that's overexpressed in breast cancers, in ovarian cancers and in cholangiocarcinomas as well. So is there potential for synergy for that? Yes, the risk in terms of the [indiscernible] warhead and Lynparza, we've seen that pre-clinically, And we've looked at that actually also with [indiscernible] pre-clinically. So there is rationale for that combination. Obviously, one of the things we need to see is the clinical data in combination and see if there's any modification of dose that's required for that combination. That's something that we'll be pursuing in the Phase I study of the [indiscernible] selective molecule.

Thank you, Susan. Okay. We've got a couple of questions still in the queue, and we'll endeavor to give short answers so we can get through them all and finish the top of the hour. But first, Tim Anderson, Wolfe Research.

I'm just trying to figure out how PROpel plays out across the mutant versus wild-type segments. You saw a bigger rPFS benefit in mutants versus wild-type driven by better control arm performance in the wild type. So I'm just trying to figure out if there's going to be pushback in U.S. and Europe on using this regimen in wild-type patients? If there is, it seems like that's a problem because as you pointed out, you don't do genomic testing routinely at the moment to identify who's mutant versus wild-type. And then just your overall level of confidence that you'll be able to show a formal OS benefit at some point in wild-type patients?

So maybe -- please go ahead, Dr. Saad.

No, no. I assume -- I think it's addressed to me. So I just maybe want to put -- take a small moment to address the overall survival, and we'd all love to see overall survival. I'm extremely hopeful. But I just want to put things into perspective. Overall survival was shown in abiraterone after a third cut of the data with over 1,000 patients against a pure placebo. It took 1,700 patients for enzalutamide to show an overall survival against the pure placebo. So the bar in an all-comer population is extremely high in an 800-patient study to be looking at overall survival, especially if we're going to look at a subgroup of a very, very small cohort of patients of 800 patients. Almost -- every study in first line mCRPC, even in second line, is over 1,000 patients for the studies to [indiscernible] to approvals of agents, always against a placebo or a treatment that clearly is not [indiscernible]. So that's just maybe a little bit of some thought in terms of expectations. So I think in the non-HRR mutated patients, 0.76 for me is very clinically significant and we're seeing it clearly, even with the confidence intervals on the left side of the one line, and we can look at subgroups where there wasn't any [indiscernible] of HRR mutations. Patients post-chemotherapy, there was not more patients with HRR mutations. But clearly, we're doing much better than what we would have expected if we were giving abiraterone alone. In patients that had visceral disease, we're clearly doing much better than we would have expected with a monotherapeutic approach. So I think there are many signals that are telling us that this is not only benefiting patients with mutations. And in the clinic, that's on top of the issues of testing. So I'm very convinced that this is beneficial. And I'm -- should we have done a 2,000-patient study so that it's powered for survival. I think the whole field is hoping that we don't always have to do [indiscernible] studies to end up with key values that are significant, especially that we did an interaction test to make sure there wasn't one subgroup leading the positive results in another subgroup. And if we even look at other studies that were presented like [ Aerosense ] where I helped to lead, you're seeing, even with overall survival, not every subgroup has an overall survival advantage. So I would just -- I'd like to caution the insistence on seeing overall survival in every individual subgroup. I think that bar is going to be extremely high.

Thanks, Dr. Saad. I mean, I think -- Tim, just maybe me a couple of thoughts on the other aspect around your question of will we get pushback across globe, and I think that [indiscernible].

Sorry, if you could just answer that from a Europe perspective as well as the U.S. perspective.

Yes, sure. So I mean, I think that there's a regulatory question, there's a reimbursement question and then there's a physician individual prescriber component to that. I think in terms of the first component, to the points that Dr. Saad was making, we designed an ITT study. We met the primary endpoint. We can see also with the HRR wild type and HRRm subsets that we've got clear benefit within those, and we think that we've got a package for regulators across the globe that's pretty darn compelling, especially in the context of the unmet need that exists that we've been talking about. I think then from an individual physician choice perspective, this is an important part of the education that we're going to need to do. I think that this is an important part of our medical education and our promotional efforts because there are some things and aspects to this that are not necessarily intuitive and in line with other things that we've seen in other tumor types. But I think that this is, again, where the narrative that we've been trying to articulate over this hour is going to be one that we're going to have to really make sure that we bring to the practicing oncologists. And I think on the reimbursement side, I mean, I -- there's no question that this will be a discussion that we're going to have to have with various authorities across the globe in terms of how that might affect the value portion of it. But again, I think we come back to the data points that we've got being strongly supportive of benefit in the ITT and in the individual populations that we've looked at. Next, let's go to Richard -- Richard Parkes on the webcast asks, the discrepancy between outcomes of the MAGNITUDE and PROpel and HRRm patients has raised a debate amongst clinicians. Can you discuss your perspectives on the likely contributors? Maybe first, Susan, I go to you, and then Dr. Saad, I can ask you to offer any other commentary.

Right. Okay. So just in terms of the context in terms of the design of the study, obviously, as we discussed, PROpel was an all-comers first-line castrate-resistant prostate cancer. And you saw the distribution of HRRm as expected in the real world, whereas MAGNITUDE enriched for the HRRm subgroups and particularly within that for BRCA mutant. So I think as has already been discussed, within the HRRm group, the performance of the control arm was exactly in line across the 2 studies. I think one of the things that is likely explaining the difference between the 2 is the dose intensity. As Dr. Saad mentioned in his presentation, we went with full dose of both olaparib and abiraterone. And we know that in both prostate cancer and in other settings like ovarian cancer and other settings that dose matters in terms of PARP inhibition and that we have seen decreases in response rate and PFS when dose has to be compromised. With olaparib, we believe we've got a best-in-class tolerability profile, which has enabled us to go at the full dose of olaparib whereas in the MAGNITUDE study, the dose of [ niraparib ] was 200 milligrams, not the 300-milligram dose. And we also know that from the niraparib submission to the European Medicines Agency that in an ovarian cancer setting from the PRIMA study, when they modified the dose, they had a reduction in mean PFS and there's an exposure response relationship, which is described in that. So I think, as Dr. Saad has said, there are 2 things that are happening that are different. One is that dose intensity matters in both the BRCA mutant and HRR mutant groups as well as in the wild-type group. And therefore, you have a better effect size seen in the PROpel study in each of the groups compared with what has been seen in the MAGNITUDE study. And I think that erosion of the effect size potentially based on dose intensity may well explain the difference between seeing an effect in the HRR wild-type group in the PROpel study and not seeing it in the MAGNITUDE study. But I'll hand over to Dr. Saad for any additional comments that you want to make.

Yes. I mean, obviously, there's always a lot of debate when we take people by surprise. I think it's been a long time we haven't had a study that people were surprised to see the results and needed some time to discuss and looking for reasons to try to explain the discrepancy. I don't think there's so much a discrepancy as just questioning whether, as was mentioned, whether the drugs are equally effective, whether dose makes a difference. And I would just like to maybe come back to just a few numbers. When we talk about -- when MAGNITUDE presented their HRR-mutated population, their result had a hazard ratio of 0.73. Our non-HRR-mutated patients has their hazard ratio of 0.76. Looking specifically at BRCA-mutated patients in MAGNITUDE, it was 0.53, whereas all our HRR-mutated patients was 0.5. So clearly, if we eventually will do the analysis of BRCA alone, it will be way better than 0.53. So I mean there is something going on, whether it's as we think the dosing, the timing and maybe the drug efficacy in itself, there's something clearly going on because our non-HRR-mutated patient is pretty much in line with their mutated patients. And when we -- I don't want to be doing review in MAGNITUDE because [indiscernible] I've looked at the data very closely. You look at patients post-chemotherapy, it's really almost not existing [indiscernible]. If you look at the non-BRCA patients, it's actually 0.99 in terms of hazard ratio. So everything is being driven by the BRCA patient, which is clearly not the situation in PROpel, as was mentioned. 10% of the patients can't be driving everything in the whole [indiscernible] group and about 1/3 of the patients in the mutated group can't be driving all of this beneficial. So I think the discrepancies can be explained quite easily.

Thank you very much, Dr. Saad. I'm going to turn now to Emmanuel Papadakis at Deutsche Bank.

So maybe a couple to follow-ups, and I'll ask in a slightly different way [indiscernible] desire or need to do biomarker testing? Or do you think folks will still want to test and then make an informed decision about what they do in that wild-type subgroup? Second question is, do you think this result changes the proportion of abiraterone use in first line relative to enzalutamide? And any thoughts you're willing to give on the [indiscernible], which presumably will inform that debate? And then just a final one on BAYOU, implications for the DuO-O Phase III [indiscernible] expecting next year, we very clearly saw the benefit concentrated in the HRRm subgroup. Is there any reason why that does not imply -- we're only going to see something similar in DuO-O in the next year as well?

So I'll try to remember the questions. But the first question was do I think it should be accepted in an all-comer population? My simple answer would be yes. I'm very -- I hope that will be the situation given all the reasons, not only the efficacy, but also the complexity of insisting on biomarker testing. Do I think we should still be doing biomarker testing when it's possible? Absolutely. I think it's an added element, an added urgency in some situations that might be on the fence. It's almost like testing for the risk of an osteoporotic fracture. You don't always need to do a bone marrow density. You can be just looking at the age of the patient, the frailty, all the rest and realize they're at high risk. So I think testing might be important when patients really look relatively low risk, even though it doesn't really exist in mCRPC. But there are situations post-chemotherapy, patients that are young, you saw the hazard ratios on which side they were on, patients with visceral disease. There are multitudes of patients that there's an urgency to treat better than what we're doing right now with abiraterone alone. Now whether this is going to shift abiraterone -- abiraterone, as far as I know, it's still the #1 drug being used in mCRPC. Obviously, with [indiscernible] if that turns out to be as effective, then you'll be able to have the option. But I wouldn't interchange enzalutamide with abiraterone in combination with olaparib. That, I don't think, should be done. We have the data that you can use full dose olaparib with abiraterone. I wouldn't be ready to say that any combination could work because there are situations like there might lead to more toxicities. So I don't think the intent here is to take over abiraterone in mCRPC, but I think it clearly is an option for all patients, which is not true for enzalutamide. There are some patients who were uncomfortable using enzalutamide in mCRPC for multiple reasons that I probably don't have time to go into. In terms of your last question, I don't remember anymore.

Well, it's okay. It was for Susan, Dr. Saad, so I think we'll -- I'll have her take it. So Susan, I think the first question around BAYOU. And then I also think that it would be useful to come back to what was part and parcel to Emmanuel's first question, which I kind of hear as, is there a risk of an outcome that will be similar to the PAOLA outcome as it relates to a subset of patients in a companion diagnostic.

Okay. So let's answer the second question first. I think we're in a different situation than we were with PAOLA because we clearly have an effect size that's clinically meaningful both in terms of the hazard ratio and the median improvement in the HRR wild-type. And we've also got a second randomized Phase II study, which also shows the same effect and the same effect size, order of magnitude. So I don't think it's going to be the same situation, but obviously, discussions with regulators are ongoing, and they will look at the totality of the data. In terms of BAYOU, in terms of the data, I think situation in bladder cancer is different from the situation in ovarian cancer. There's a much higher proportion of patients with homologous recombination deficiency in ovarian cancer relative to bladder cancer. And of course, [indiscernible] also includes a VEGF containing regimen, and we know that there's interaction between VEGF agents and immunotherapy as well as VEGF agents with agents like [ olaparib ] within that context. So I don't think you can cross read from BAYOU straight into the DuO-O and DUO-E settings, to be honest. And obviously, the total number of patients that we have that were within the HRRm subgroup was relatively small in terms of understanding the contribution of durva to olaparib within that setting. It was asking the question, the contribution of olaparib added to durva, which isn't quite the same thing. So I think there's more to learn in terms of that interaction, and I look forward to the data from DuO-O and DuO-E to answer that question. I think it's an important question.

Thank you, Susan, very much. All right. I think we have time for one more question. I appreciate everybody staying a little bit longer than our planned top of the hour. I'll go over to Michael [indiscernible], please.

My question is for Dr. Saad. If I understand correctly, Dr. Saad, you've stopped just short of suggesting that this regimen should be standard of care in PROpel patients. And I'm wondering as we analyze the potential of Lynparza in the setting, what's the main reason we might want to temper our enthusiasm?

Just -- being a standard of care, a standard of care? Absolutely. I think it is -- it will become the standard of care. I always temper, I mean, I'm a clinician who sees patients coming in that are 92 years old. To say that I will give a drug or a combination of drugs to every patient who walks in the door is not what I do in my practice. It's a case-by-case discussion. There are some patients where I'm worried about adverse events. We don't give chemotherapy. I mean chemotherapy is a disaster, less than half the patients die of prostate cancer without ever getting a chemotherapy, and we know it is a standard of care. And here, I think it's a much lower bar for most patients to be getting olaparib. But I always caution to say this is the standard of care because it would sound like anything else is unacceptable, unethical. And I developed guidelines. And in prostate cancer, it's not like leukemia in a 7-year-old patient where you cannot diverge by one -- patients have an opportunity. So maybe I'm being just conscious of everything I say, but being a standard of care? Absolutely. Now, will it be used in every patient? Absolutely not. There are patients where we are going to be concerned about. There are patients that we don't even treat with first line mCRPC because we're concerned of certain, but the vast majority clearly are being treated. And I treat 90-year-old patients with abiraterone who come in the door that are at home, very, very -- I wouldn't give them chemo, but I'm very comfortable with AbbVie. But it will be a case by case of saying, a 90-year-old patient who's frail, will I give every 90-year-old patient the combination? It's going to be a case-by-case discussion. So I'm just being -- I guess I'm just being honest in what I do in practice, and I think most oncologists would not use a one-size-fits-all approach for every patient that comes in the door.

Thank you, Dr. Saad. Much appreciated. So with that, we'll bring the conversation to a close. I just want to reiterate and highlight that I hope you can hear from the commentary from Dr. Saad, how important these data are, how high the unmet need is globally for patients who can benefit from this combination and why also we have a huge degree of confidence in just the credibility and the strength of these PROpel data. And obviously, we will take that with pace to health authorities across the globe and appropriate educational efforts are already underway because there's a lot that needs to be worked through as this extraordinary data set is being digested by clinicians across the board. So with that, thank you again to all my panelists. Thanks to those who called in, and we wish you a good afternoon or evening, wherever you are. Thank you very much.

Thanks.