Silke Gillessen: Okay. We are going to move to the last discussion for today and we speak about metastatic castration-resistant prostate cancer. Again, I want to just remind you that a lot of also these questions we have addressed in our manuscript from APCCC 2021. I think this goes really close to your question, [inaudible 00:00:26]. This is, do you recommend even docetaxel re-challenge anytime in the treatment sequence in the majority of patients who have received docetaxel in the hormone-sensitive setting and progress to castration-resistant disease? You see if someone has really progressed within 12 months after start. There is only 14% who would do that. We just chose the 36 months. The patients are progressing within three years. There is consensus to use docetaxel again in the course of disease. So, I think the skepticism that you have against docetaxel doesn't seem to be the case for the majority of the panel. We can discuss that afterwards.

Then Maha, that is maybe interesting for you because you said actually a bit the opposite in your talk. There was the question, do you recommend a direct switch to another AR pathway inhibitor, Abi, Apa, Daro, or Enza in the majority of patients who have received one line of AR pathway inhibitor and then progressed? Here was also a clear consensus to not do this. I don't know. Can I ask someone to comment? Maybe Heather again. Can you?

Heather Chang: Well, I think what this seems is it reflects and in my practice is that if there's another option, and based on Maha's discussion and the CARD data switching the mechanism of action if that doesn't look like there's efficacy, and the same might be true for the docetaxel question. If there was benefit and then durability of that for some period of time, then it makes sense to go back. But if it's really a limited benefit, then the wisdom of going back to that would be in question. That's my practice though.

Silke Gillessen: Kim, you have done a lot of work on that effort. Dissect this one.

Kim Chi: I'd like to comment on the first one about using docetaxel within 12 months at any time in that patient's treatment sequence. For sure, I'd rather use lutetium-PSMA-617, olaparib if he was eligible for it, even radium-223. But after those, if he was still a candidate, I would still try docetaxel. We don't have a lot of data on predictors of benefit. These are patients that are going to do badly, so if I have an option for another treatment, I'm going to try it.

Johann de Bono: Why not cabazitaxel?

Maha Hussain: Yeah, exactly.

Kim Chi: As well. I'm going to try all those things, but this was at any time in that patient's treatment sequence. That's why I answered yes to that question, I will consider it in that patient at one time.

Johann de Bono: I would like to say one other thing. We have clear evidence, going back to what Anwar asked earlier, that prior Abi/Enza decreases taxane benefit, the benefit from taxane at most is very modest from Taxotere. It's going to be even more modest post Abi/Enza. I think I would actually probably not give docetaxel again, but I would give cabazitaxel primarily because actually, cabazi was made specifically to work when doce fails. And in the lab, actually, the [inaudible 00:03:50] of taxane resistance actually [inaudible 00:03:52] upregulation and actually, cabazi works against MDR high tumors. I don't know what. Maha.

Maha Hussain: Yeah. No, I fully agree, Johann. I think the issue comes up is the time from end of therapy to failure. People who are progressing at a shorter interval, you know that their disease is bad. There's only so much bone marrow you can damage. So, if you're going to use something, you'd rather jump into something that has a better chance of working potentially and the data is certainly there. This is slightly different than the AR story because again, a lot of times, we're picking those patients with a whiff of rising PSA especially. Actually, Kim Chi's and his team's data, I think is very powerful. Honestly, if you have room, we have wiggle room and you want to save the patient from getting into something that's too myelosuppressive, too toxic or whatever, I think trying Enza has a potential chance and there is no real harm in doing that. It's really balancing risk and benefit and better assessment of the disease so that you can decide, is it really something that I have to act on right now or can I wait until tomorrow, kind of thing.

Silke Gillessen: But can I just-

Kim Chi: Patient selection is key though.

Silke Gillessen: ... go back?

Maha Hussain: That's what I'm saying, yeah.

Silke Gillessen: Yeah. Can I go back to Johann because now we have the triplets upfront in the hormone-sensitive setting. So, the patients are not progressing on docetaxel because they only get six cycles. So, in reality, you don't know if they responded because it could just be the ADT and it's only six cycles. It's a bit a different story so you would still not consider the re-challenge?

Maha Hussain: No, that's not what I said. I said-

Silke Gillessen: No, no, no. Sorry, this one's for Johann.

Maha Hussain: Oh, I'm sorry. Yes, sorry.

Silke Gillessen: That's for Johann, sorry.

Johann de Bono: Well, I think we don't have data to say anything firm on that, but I worry that these patients will miss out on cabazi because actually, if you wait long enough, you'll never get it. Actually, I would argue that if you're going to give another taxane, cabazi is clearly the better taxane, and that 40% of cabazi becomes doce anyways, so why not give cabazi? I think it's the better drug.

Silke Gillessen: Clear statement.

Eric Small: Okay. Can I make a quick comment?

Silke Gillessen: Yeah.

Eric Small: For both of these sets of questions, whenever there is progression, I would want to put in the algorithm, do a biopsy if at all possible, because if you do discover 17% of patients one of the small [inaudible 00:06:21] variants, then you need to be thinking about platinum.

Aurelius Omlin: Very good point. We didn't ask it this time. I don't think we have a question about biopsy at progression, but we've discussed them previously. Now we have a number of questions that go into the sequencing question depending on what patients have received in the hormone-sensitive state. The first one is probably a scenario that we will see less and less, so patients who received ADT alone. This is not surprising, already a consensus 2017 for an AR pathway inhibitor. Very few, 3%, would go for the combination with PARP and another small percentage would go directly to docetaxel. The picture changes a little bit in patients, and you probably all have seen these patients who progress quite quickly and rapidly on ADT alone. We used six months. It's an arbitrary number. Here, the picture changes and it's about half-half or an AR pathway inhibitor or for chemotherapy. The same 3% would go for the AR plus PARP combination. Any comments from the panel? Someone? Kim, yes.

Kim Chi: Well, we did a trial. It was published in Annals of Oncology I think last year, where we randomized patients like this progressing rapidly on first-line or ADT to CRPC and we randomized them between cabazitaxel and AR pathway inhibitor Abi/Enza. Really, the response rates were the same. Although by the study protocol, chemotherapy was superior, really, when you look at the response rates and so on, they were more or less similar. I think you have your choice, but I tend to go with an AR pathway inhibitor because it's less toxic and prostate cancer is driven by this. We see pretty good responses and we can always use chemotherapy.

Emmanuel Antonarakis: Just a quick point. I would always check testosterone. I have four or five patients a year who don't get testosterone suppressed on ADT alone. If that was the case, I would use abiraterone as a next step.

Silke Gillessen: Good point.

Aurelius Omlin: And Maha.

Maha Hussain: I think this is a space where better understanding of the biology and the genomics is critical. We have a PCF-funded project looking exactly at situation like this where patients progress in a short period of time on frontline management for castration-resistant disease. I do think we need to understand more and more the biology. Otherwise, we're not going to have better answers.

Johann de Bono: Can I plug our data here?

Kim Chi: [inaudible 00:08:59] DNA repair alterations and it was very enriched for patients with BRCA2 and homologous for combination repair alterations.

Johann de Bono: But I'd like to plug the microbiome data with Andrea Alimonti here because we now have very strong evidence that actually after castration, the microbiome changes and we've published in Science the work that Andrea and I led that actually, there is a complete change in the microbiome with castration. It's very producible that leads to an accumulation of microbiota that make androgens. Actually, going back to, I think what Kim just said, was it Kim, I think it was Kim, it's clearly the case that these people often have androgens, and maybe not the typical DHT testosterone androgens that microbiotas are making.

Silke Gillessen: Oh, that is obviously very interesting preliminary data. We hope we find a yogurt sooner or later. Yes. We go now to probably the situation we see much more now. Again, mCRPC without any alterations, but who had ADT plus an AR pathway inhibitor for hormone-sensitive disease. Here, again, the panel was very clear to give docetaxel. So that was the consensus. [inaudible 00:10:13] again, I think the panelists believe in the activity of docetaxel. Here, the same question. Maha, as you said, so this is again the same question, but progressing within six months on this doublet and here even more people would give chemotherapy. So here, Eric, we have the option of giving a platinum-based regimen as well. Obviously, we don't have the data, but I think it's very clear and very clear consensus on both of these questions.

Johann de Bono: Can I just say one thing and that's the abiraterone prednisone to abiraterone dexamethasone switch, which actually we've studied primarily in CRPC, I know, but in someone who's progressing just with a rising PSA and they're feeling fine and not a lot of disease burden, I would recommend consideration if they're on abiraterone, Abi/pred, Abi/dex switch, which a lot of people still don't seem to know about. And it's low dose dex, 0.5 milligrams, which is less likely to activate GR and promiscuous GR mutations.

Silke Gillessen: And you would do that in the situation just someone progressing or also in the patients who have only six months responded to a double?

Johann de Bono: I think the answer is it depends. You're the clinician. You're at the end of the bed of the patient or at the end of the clinic room. If the patient is fine, a rising PSA, symptomatic, not a lot of disease, even if it's six months, I would still consider that.

Kim Chi: I think that's a good point because many patients are progressing with a low PSA, slowly rising PSA even within six months and there's no reason to rush into chemo. I think we have to be cautious as we take these results and then apply it to our patients that we're going to see tomorrow.

Silke Gillessen: Maha, do you want to-

Howard Scher: [inaudible 00:11:58]. Sorry.

Maha Hussain: Oh. Sorry, Howard. Go ahead. Go ahead.

Howard Scher: Ladies first.

Maha Hussain: Ladies first. Thank you. So, I taught you something, that's great.

Howard Scher: It took a while. It definitely took a while.

Maha Hussain: I'm joking. I'm joking. No, I think this is where clinical judgment has to come in. This is not about avoiding chemo. The reality of it is chemo is life-prolonging and the sicker the patient, the harder it is to give them the chemo. So, somebody who goes on a double therapy and progresses within five minutes, to me, you really need to change therapy significantly. I have used dexamethasone or increased, if the patient are on abi, the prednisone to BID in situations where somebody's been on treatment for a good period of time, they are stable, it's only a whiff of PSA or slow trends, not a rapid doubling time and especially imaging, because I don't really just go by PSA. I definitely check the imaging. If there is no significant imaging progression, you have a window. But the reality of it, honestly, from my experience, again, that's an NF1 experience as in myself experience, it's not that promising.

Howard Scher: No, but a lot of the time, you could basically do that trial while you're doing this restaging on the patient.

Maha Hussain: That's exactly right, yeah.

Howard Scher: I've seen some responses that have been surprisingly very, very long. So, Johann, that was a key observation and it really makes a difference for some patients because you still have the docetaxel.

Maha Hussain: Yeah.

Silke Gillessen: Okay. Just to maybe clarify, we didn't write it really here, but progression is not PSA only meant in that question as well.

Aurelius Omlin: Now, Matthew, a question to you. In patients who received the triplet and progressed, so they have received ADT and AR pathway inhibitor and docetaxel, what's the next step? There's still a proportion of the panel who would go for the alternate AR pathway inhibitor, the same 3%, I guess, who would go for AR plus PARP. About a third who would use a taxane, most likely probably cabazitaxel depending on the interval, and then quite a significant proportion would go directly to lutetium-PSMA and a small one to radium. Any comment from Matthew?

Matthew Smith: Yeah. I like options three and four. They're evidence-based, particularly if we're talking about cabazitaxel and the lutetium-PSMA would comply, would be consistent with the patient who would've been enrolled on VISION, not necessarily treated with a triplet, but they would've received those lines of therapy. I'm not at all enthusiastic about the AR pathway switch, particularly in that setting. I like options three and four and I'm glad to see that the majority of folks voted for those.

Maha Hussain: I just want to make a comment. I absolutely agree with Matt, but I think one thing we are forgetting and I think this is important, this is really a window for opportunity for patients to go on clinical trials. I think when available, that should be part of the consideration. Obviously, appropriate trials for the patient, but that should be really always considered.

Silke Gillessen: Very good point. Okay. Here is the question about the debate. So now, mCRPC patients who are about to start an AR pathway inhibitor, do you recommend the combination with the PARP inhibitor as first-line therapy? You see here when there is a pathogenic BRCA1/2 alteration. It's half-half saying yes or no. When there is a DNA repair gene alteration, but not BRCA, there is a consensus not to give the combination. I think that's very clear. I don't know. Noel, do you want to comment?

Noel Clarke: Well, only to say I'm a little surprised that people have decided to jump when we're not quite ready yet. I don't think the information's there to follow this recommendation.

Silke Gillessen: To be honest, it's not a recommendation. It's half-half, and then the other one, it's no.

Noel Clarke: Well, yeah. I would say follow the no. That's the right course.

Silke Gillessen: So it's not going to be your recommendation.

Aurelius Omlin: Okay. A follow-up question on that. In patients who started ADT plus and AR pathway inhibitor plus minus docetaxel, we kept it flexible here for mHSPC and now, progressed. If they are BRCA1/2 positive, germline and or somatic, what would be your recommended option? Is it chemotherapy? Would you switch to an alternate AR pathway inhibitor and combine with PARP? Would you continue with the AR pathway inhibitor that they are on and add the PARP inhibitor or add the PARP inhibitor alone? That's option four. That's what the majority chose. It looks a little bit different in patients who have other DNA repair damage alterations. 56% would switch to chemotherapy and 28% would use a PARP inhibitor in these patients. Emmanuel, you are nodding.

Emmanuel Antonarakis: Well, I think in the BRCA1/2 subset, especially BRCA2, that was almost a consensus for option four. With the BRCA2 patient, I would want to use the PARP inhibitor as soon as I possibly could. With the FDA and EMA label saying they have to have progressed after one prior AR pathway therapy, then he would be eligible. Then Maha's slide about ATM and CDK12 and sequencing, I think was important in the non-BRCA patients, especially ATM. It didn't specifically say there, but if I knew this patient had an ATM mutation, the post hoc analysis that Maha showed does suggest that this should be reserved for after a taxane. I have to admit even without that data, with the responses in ATM and CDK12 being more modest, in my opinion, I would go for a chemotherapy first, which I think has better evidence based on CARD and other studies and then save the PARP inhibitor for a third line.

Silke Gillessen: Oh, yeah. Andes.

Andes: Andes [inaudible 00:18:21], Sweden. This is related about to what Anwar discussed about when to understand when we have a progression to initiate the new treatment. But I'm also thinking about what we heard at the [inaudible 00:18:37] meeting with the PRECISE study. Adding another drug instead of stopping the first one, layering instead of shifting, I think that could be beneficial for many patients. Is there anyone in the panel who would like to comment on why we don't have that many studies on layering instead of changing treatment?

Johann de Bono: I think we need trial data before we can recommend that, in my opinion. They are robust particularly for OS for the subgroup of interests. I think there's a theoretical risk based on [inaudible 00:19:17]'s data and other's data and [inaudible 00:19:20]'s data most of all is that the AR blockade downregulate in [inaudible 00:19:23] and the data from [inaudible 00:19:25]'s lab and other labs that if you don't regulate Q70, DNA-PK can cause resistance to PARP inhibition. So, I don't know. The answer is I really don't know about this. PROfound, obviously, was a single agent. I think we'll have to see more of what the other trials are going to show in that population. My understanding is that in PROpel and MAGNITUDE, those populations were small, smallish. Is that right, gentlemen? Well, certainly there is none in PROpel, right? Second AR antagonist?

Kim Chi: There was very few in MAGNITUDE.

Johann de Bono: Yeah.

Anwar Padhani:  It's like one way of layering. If this show oligoprogressive disease, then one way of layering it would be metastasis-directed therapy in the castration-resistant state. There's been at least 12 papers in the last year looking at that possibility, but that's a physical form of layering, so you can continue the backbone, but you can kill the clonal cells that are coming through.

Johann de Bono: The other big concern is that ends as a potent inducer of CYP3A4 and can really decrease exposure to many drugs, including some of the PARP inhibitors. So, if it’s Enza particularly, just be careful because it may decrease PARP inhibitor exposure.

Kim Chi: I guess there's also a theoretical that continuing the abiraterone prednisone would worsen things if you have an L7 or 2H mutation, which can occur in 10%, 15% of patients. Unless you're characterizing that patient, it could be harmful. And as well, we see AR rearrangements emerge as well so it could be just useless.

Andes: Yeah, because we all know that there will be new metastasis. Some metastasis will disappear, there will be new metastasis. That's a reason to add something instead of replacing a treatment.

Maha Hussain: But I think the point here is that the new metastasis occurred despite the treatment you have the patient on. So, it would seem to be more rational to actually expose the cancer to a different drug as opposed to keeping the same drug that it already developed resistance to. Not to mention, in the setting of prostate cancer, you have a window in terms of organ functionality, specifically the bone marrow, so you don't really want to take chances. Could you possibly exactly do what Anwar suggested? You have only one area that is worse and the PSA is not that high and the patient is not so symptomatic. You can spot weld. I call it spot welding, but basically focal radiation to one area. But I think you'd have to use judgment as to what would be the best package for the patient.

Silke Gillessen: And we discuss that tomorrow. Nick, last statement to this one.

Nick James: Yeah. I think a very interesting voting in that we've very clearly seen that even with a panel with a whole lot of North Americans and nobody recommends an Abi/Enza switch or equivalent. It very consistently heavily voted against in all settings, PARP, chemo, the whole lot, and yet we commonly see it as the control arm in trials like VISION. The question for the whole panel, just to be a bit provocative, do you think ethics committees should refuse to accept that as an acceptable control arm in trials in the future? I'll just lay the cards on the table. I think they should not approve that as a controlled arm in a licensing trial in the future.

Matthew Smith: I agree with you. I think it's a bad tradition to allow the losing arm of randomized control trials to be the control arm of the next trial. When the CARD study clearly showed superiority of cabazitaxel to the switch, I took that to mean that most future trials would have to go up against the winner, but that's not at all what we've seen.

Maha Hussain: I actually agree with you, but remember, it's a low bar. You want to get an indication, and so you're basically comparing to zero and so you should be better than zero for sure. But in all fairness too, I think a lot of these trials are done internationally and what might be available in one country may not be available in another country. So this is where potentially allowing some wiggle room in the control arm.

Matthew Smith: Then I think the regulators should be honest though and just make them placebo-controlled trials and dispense with the ridiculousness of doing the AR switch, which has been shown over and over again to be ineffective.

Nick James: I think also some trials have actively excluded active therapies. Cabazitaxel or radium, these are drugs with proven survival gains that have been excluded from some trials, which I think makes the thing doubly question.

Maha Hussain: I have to say, this is the beauty of having the, I guess what we would call the cooperative groups in the US, the similar would be in Europe, like EORTC and all of that because there is different biases. There wouldn't be as much bias. You're comparing one to the other and you have no investment in either of those things. You're just looking for the better drug. So, there's a lot of business model issues that I think affect a lot of this trial design.

Aurelius Omlin: Oliver.

Silke Gillessen: Can I just-

Aurelius Omlin: I'm sorry.

Silke Gillessen: Maybe we can note there is just two editorials, one in Annals, one in JCO saying exactly that, why are we choosing these control arms. It's really worth reading them even if [inaudible 00:24:39] from one of them, but I think it's very short and it's really worth looking at it. Oliver, I know you are sometimes arguing for the second one.

Oliver Sartor: No, no, no, no, no. I think it's best to change. I think I did a little count and I think it was about six trials from now using that alternative AR as the control arm. When this wave comes through, we're going to have a whole lot of new positives to look at and I think they're going to go away. I think there's an evolution in our thinking and one of the evolutions is that chemotherapy can probably be administered to more patients and chemotherapy adverse patients are going to have to face the data now. There's a lot of convenience and oral to oral and it happens in the US because it's easy, but I think the data's so overwhelming now that it's going to disappear soon. It's a problem about to go away, bottom line.

Aurelius Omlin: We have additional support from these votings.

Silke Gillessen: Yes.

Oliver Sartor: Yeah. It'll take the additional data. Nothing changes overnight. This convenience with PROfound study, that was setting a stage, if you will, and the VISION study perhaps set a trail, but the AR wasn't mandated. Nevertheless, as we go forward, we really do have better alternatives and they'll be there.

Silke Gillessen: Thanks, Oliver.

Aurelius Omlin: Michael.

Silke Gillessen: Michael, last [inaudible 00:26:10].

Aurelius Omlin: Also to this point, I guess.

Michael Morris: The other thing about the control arm is that what is frequently substituted for. We don't like the second line AR-directed therapy, and also inappropriately rigid, it has to be chemotherapy. If we have control arms that are only for fit patients with very aggressive disease for whom that control arm is appropriate, you get a readout that's similarly really limited in terms of the actual efficacy of the drug in a broader population. Truly, you'd want control arms which are basically dealer's choice, because then you're getting the heterogeneity of the population and the heterogeneity as well of the all-available real-world choices that you have. But I don't think that you could say, well, the PROfound style thing is unethical and the only ethical thing is docetaxel because that's equally restrictive in terms of the readout.

Silke Gillessen: Fair. Yeah. We'll select patients obviously for these trials. It's going to be complicated, but we have more options and that's the good thing. Let's go to the last set of questions or the results. This was, again, a consensus. In the majority of patients with symptomatic mCRPC who meet criteria for both treatment with radium or with lutetium, the panelists had a consensus to recommend lutetium first and not radium. On the other hand, when we asked if someone would recommend radium after lutetium in the majority of symptomatic patients, you see here again, it's kind of half-half saying yes or no. Neal, do you want to comment on that?

Neal Shore: Yeah. I appreciate why this may be a feeling that there's not much else to offer the patient, but when I look at this, I think to myself, especially based upon the last set of questions that had the use of radium in low single percentages, I just think that it's missing the window of opportunity where there's a real op. It's fascinating to me. We see a lot of sequencing of AR-targeted agents. We see patients not getting combination therapy in mCSPC and you have level one evidence for patients who meet criteria for radium prior to chemotherapy, and then yet you see the sequencing that was our response. You see patients getting combination AR targeted therapy with PARP inhibitors, which I think when they were asked these questions, the data was really and probably not even that well-known to many of the respondents. I'm fascinated by it.

You have a level one evidence of a life-prolonging agent that's remarkably well tolerated. But I understand it here. I'm just shocked, frankly, that it's not used. My feeling is in the US, it is used a lot more than probably outside the US in pre-chemotherapy.

Silke Gillessen: Interesting.

Aurelius Omlin: Any more comments? No. Then we have two last questions and I think Michael, it would be nice, Hofman, if you can comment after, because we had a lot of questions on lutetium-PSMA in the October meeting that was just published 10 days ago and we missed some of the questions, which we now voted on this time. This is regarding the PSMA PET threshold that you use for selecting patients for lutetium-PSMA treatment. We have the one from TheraP, which is stricter than the one that was used in VISION. It did not reach a consensus. A quarter would go for the more TheraP criteria and 74% would go more with VISION. Maybe I show the last for tonight.

What do you do to identify PSMA-negative sites of disease? If you select patients for lutetium-PSMA treatment, do you correlate PSMA PET with an FDG PET? Do you correlate the PSMA PET with a contrast-enhanced CT? That's what the majority of the voters voted for. Or do you use an FDG PET selectively if the correlations provide equivocal results? That's 17% and 4% would not use or don't think it's relevant to identify PSMA-negative sites of disease. Michael, yes.

Michael Hofman: They're slightly complicated questions there. There's probably no right answer, which is why there's no consensus. In Australia or at least in our center, we're still using the TheraP-like criteria. That's probably an institutional and personal bias having done the TheraP study and having some familiarity with, in part it's because we don't have unlimited resources. Even in our center, there is a short waitlist to get lutetium-PSMA, but in many places, it's just not accessible at the moment or you need to travel far to get it. I think in that setting, we ought to select patients that are more likely to benefit rather than perhaps a 20% response. We see a clear biomarker that PSMA is a predictive biomarker. Response is going from 20%, which is still reasonable, but all the way up to 90% if you have really high PSMA expression.

FDG. Probably also people that have answered SUVmax 20 and using a TheraP-like selection maybe are using a bit more FDG. We have a lot of experience with FDG. I think it's extremely useful. It also depends on health economics. Are you paying for the treatment or is the government reimbursing it? I think if you need to pay for it out of your pocket, you might not want to pay for it if you have a lower response rate. But if you're in that higher response rate, you really want to access the treatment. So, I'm very keen on better selection and really personalizing care as much as we can.

Aurelius Omlin: Matthew showed that threshold of PSMA expression SUV. Can we use the same threshold if we use other tracers for the diagnostic imaging, like not a Gallium-68, but an 18-fluoride PSMA PET?

Michael Hofman: This is a really good question. If we could add another question onto this round of APCCC voting,-

Aurelius Omlin: Let's do it.

Michael Hofman: ... it would probably be on this question because it's highly relevant because the FDA indication for PSMA-617 seems to imply that you should use a Gallium PSMA-11 scan, whereas DCFPyL is FDA approved and widely available. In our center, we use both DCFPyL and Gallium PSMA-11 to select patients for lutetium therapy. I think both of those tracers are bioequivalent and you can pretty much interchange them. In some centers, Gallium PSMA is simply not available and to think that you're going to travel far to get a Gallium PSMA scan because the FDA implies that you should, I think is a little bit mad. I think it would be good to have a consensus guideline. We can make it a little bit trickier by throwing in PSMA-1007 as well, which is more accessible in some European countries. I think for the purpose of selection for lutetium-PSMA therapy, any of those tracers are suitable and we don't want to restrict access to PSMA-617 therapy because you don't have one of these PET tracers over another.

Silke Gillessen: Michael, if I understand you-

Matthew Smith: I agree with you, Michael. I agree with you, but I think you're being generous in your regulatory comments. The FDA didn't imply that. They explicitly state that in the label in the United States, so it's more than an implication and it is-

Michael Hofman: They don't use the word Gallium-68. They say PSMA-11. They leave out the Gallium-68.

Matthew Smith: You're right. You're right.

Michael Hofman: It opens a room for a little bit of interpretation.

Matthew Smith: But they do specify Gallium, yeah. That would-

Michael Hofman: No, no, they don't specify Gallium.

Matthew Smith: I disagree.

Michael Hofman: It specified PSMA-11, but no Gallium in the wording.

Matthew Smith: Okay. Oliver.

Silke Gillessen: Check again and we'll let you know tomorrow, but Michael, just to clarify-

Matthew Smith: I share your concern, Michael, is what I'm going to say.

Neal Shore: Yeah, but in the PSMAfore and the PSMAddition sponsored by AAA and Novartis, they absolutely specify only the Gallium LOCAMETZ scan. Full stop.

Michael Hofman: Well, that's okay. Clinical trial design, you can do whatever you want, but I think for a clinical rollout, we don't want to have that restriction.

Silke Gillessen: Just to clarify for everyone who doesn't have the Gallium, so you were saying that probably the SUV, what you had in TheraP, you could use for all the PSMA tracers. Did I understand that correctly?

Michael Hofman: Yes. I think it's a reasonable threshold for all those tracers.

Silke Gillessen: Okay. Thanks. Oliver, last comment.

Oliver Sartor: Yeah. This is a really interesting question. Maybe just to give a little bit of context, you realize that the phase III trial was designed many years ago and it takes a long time to implement and execute a phase III to get the results and go for regulatory approval. One of the real important goals of the whole VISION trial was to get to a regulatory approval in a global manner and to be able to move the needle from the patients who need therapy. It was absolutely designed from the very beginning for regulatory approval throughout the world. We had to choose something really simple. The idea of doing two PET scans would've been extremely inhibitory. We also felt that we needed to be able to have this really simple criteria that anybody, anywhere, Peoria, Illinois, or wherever could be able to choose the patients.

Oliver Sartor: In our initial cut, the idea of coming against the liver and having a single metastatic lesion, and then having this exclusion of patients who might have PSMA-negative chosen on cross sectional CT contrast-enhanced imaging, that's what won the day. In the end, it did work. By the way, I might mention that there's a UCLA data set that would indicate that in fact is the patients who do not meet the VISION criteria who get treated with PSMA-617 lutetium do very, very poorly. So the biomarker did work. On the other hand, it's incredibly important to realize there was no attempt to understand optimal imaging and how it might play out. It was done for regulatory approval. It was done to be simple and easy and implementable everywhere. So that's how we ended up. We don't claim to be optimal. VISION was never designed to be optimal because we didn't know what optimal was at the time, and of course, it's all perspective.

Silke Gillessen: Makes a lot of sense just to say that, I guess in a lot of countries, the PSMA PET is not done with contrast enhancement. I think everyone should be very clear also about that because there is discussions from the radiology department versus the nuclear medicine department. I think this is a very important point to make clear that in VISION, it was with contrast media. Thank you very much for this clarification. I guess we are all done. We see each other tomorrow morning again. I hope the sun is still shining so you can enjoy the sun. See you tomorrow.

Aurelius Omlin: The panel, please come to the speaker's dinner.

Silke Gillessen: Oh, yeah. Thanks to the panel. Yeah, again.