Read the Full Video Transcript

Ian Davis: We've got about 10 minutes of questions before we get to the panel questions. Please come to the microphone, and feel free to put questions in online if you wish. Well.

Noel Clarke: Well, thank you to the speakers. Noel Clarke from Manchester. You'll have to forgive me for feeling that I'm living in an Alice in Wonderland world. Johann, you've given us an excellent overview of PARP inhibition and being openly critical of the PROpel trial, as I know you are.

Yet we have two other excellent speakers in Johann and Niven who've told us how difficult it is to get a gene test done and how unreliable it is, except when you're at the Institute of Cancer Research or some other center of excellence.

So given that this is a consensus conference, and the audience wants to know what to do, I'm really not sure anybody's told us what to do, actually. So perhaps we could take a view.

Johann De Bono: Actually, I disagree, Noel. I think what you've heard is that for CRPC, you should pursue a second biopsy if you can. And if you can't, get the archival tissue tested, do genomic analysis. And that could be with a good targeted NGS assay that gives you tumor fraction, allele frequency of the mutation, and preferably a biopsy with high tumor content.

So I think we do have clarity on what we need to do. But yes, as you're rightly saying, there is a significant false-positive and false-negative rate with these assays, which is partly why I'm so concerned with some of the claims made by these trials.

There is, particularly for the false-negative rate for the BRCA2 homedels. This is an issue, and I think we do need better assays, period. But particularly when we don't have OS benefit for the non-HRR patients, I think—

Noel Clarke: Yes, the problem is, Johann, forgive me for cutting across you. But the tests are simply not reliable in the clinic to be able—to enable us to do that. They're running at about 50% to 60% possibility on tissue test. You know this, I know this, and everyone on the panel knows this.

Johann de Bono: That's incorrect, actually. I think the failure rate now in most labs is probably about 15%.

Noel Clarke: With fresh tissue, not with archival tissue, Johann. I completely disagree with you. That's fresh tissue. And mostly, it's not available.

Johann De Bono: It depends on the lab.

Audience Member: Nice talk, Elena. Quick question. Any thoughts on the mechanism of that sort of initial dip in the hemoglobin? That's one question. And the second question—so I wonder if there's any sort of supplementation that could be used to try to prevent that if we kind of know the mechanism.

But the other question is, have people looked at any predisposing factors for these MDSes in PARP-inhibitor-treated patients? Are we looking at CHIP and things of that nature to see if we can identify those at risk?

Elena Castro: So regarding the first question, it's not very clear, but it seems that the inhibition of PARP, particularly PARP2, is actually the one that could be interfering with erythropoiesis and could be causing that.

And I guess what happens is that once we see that and we reduce doses or we treat these patients better, this is how we manage to improve, and they don't present that later.

Audience Member: Can I answer that real quick? Because I've had the experience where people will—you'll interrupt, and then they come back up, and they're good, without a dose reduction, and I didn't really do anything.

Johann De Bono: I agree, Anna.

Audience Member: And I don't know why.

Johann De Bono: In fact, Shahneen and I published the niraparib paper many, many years ago. I had a bit more hair then. But actually, instead of dose reducing, we gave a drug holiday. And I had several patients where you gave two weeks on, two weeks off, and you actually got good disease control without a dose reduction, with a drug holiday.

And I think that's a preferred way forward, actually, than dose reducing. But to go back to your question earlier, I think duration of therapy on PARP inhibition really matters. These are drugs that are causing double-strand DNA breaks in hemopoietic cells, right? Although in a normal cell, you will repair.

And if you look at the ovarian cancer data, there's a lovely paper from Gustave Roussy, the Gynecological Cancer Group. They are claiming that more than 6% of the ovarian cancer patients are now getting MDS or AML after PARP inhibition. And this was something they did not see much before PARP inhibition. So it's a big increase.

And so I think this is something we're going to have to be very careful about, particularly as we introduce also the radionuclides and immunoconjugates with Top1 payloads.

Audience Member: Thank you.

Audience Member: Thank you. Recently, we did a really, really large systematic review and meta-analysis. It seems we only see thromboembolic events for PARP inhibitor mostly for patients with prostate cancer, but not other tumors like breast, ovarian.

But there is actually no explanation under it. So I don't know whether you have a similar observation or an explanation for that.

Elena Castro: There is actually another meta-analysis that suggests that this is seen not only for prostate cancer, but across different tumor types. So there may be some relation with that. The other thing—I don't know whether if we see that more in prostate cancer is also because prostate cancer is a procoagulant situation. Also, it's the androgen deprivation and all the things that we add on top.

I don't know if this could also be contributing to the fact that we see this more frequently in prostate cancer.

Audience Member: Yeah, this is exactly more common in prostate. But, for example, for ovarian, they have more abdominal surgery and platinum-based chemotherapy. But still, it is—we don't see so many thromboembolic events in those patients.

Johann De Bono: I've given a lot of PARP inhibition. I haven't seen a lot of thromboembolism, to be honest, in my patients, more than the usual risk. Prostate cancer patients, 5% to 10% get thromboembolism whether they're on PARP inhibitor or not, so.

Niven Mehra: They're all asymptomatic on imaging. But I've seen a few.

Elena Castro: I've seen some as well. What I don't know is also because when these trials were conducted, we were also in the COVID situation. We had COVID. That would also have contributed to this. And we also have many—I don't know if any of those cases could be explained by that or not. I don't know. But I have seen some.

Audience Member: Thank you. I have a question, especially taking advantage of so many experts on PARP inhibitors, in terms of a new combination, at least in theory, with lutetium. Based on the mechanism, lutetium may induce some damage in the gene repair.

So there has been proposed, actually, PARP inhibitors may enhance, actually in all comers, the treatment. So I don't know whether there's data there or if there are some ongoing. So I was just curious to know from your [INAUDIBLE] side.

Johann De Bono: You should talk to Shahneen Sandhu who's here somewhere and is running a trial. Oh, she's actually—go and talk to her. She's running the trial. But actually, that's what PARP inhibitors were first developed for—as sensitizers to DNA-damaging agents. A work by Ruth Plummer in Newcastle, beautiful work many moons ago.

But that's a different story to synthetic lethality. And I think when you're giving PARP inhibitors in that scenario, you can give them for a few days, three or four days, not continuously. And I think the big worry for us with regard to MDS and leukemia is we are giving a DNA-damaging drug continuously daily, some would argue with abiraterone [INAUDIBLE], for long periods of time.

And as we move to the front-line population, HSPC, this could be years. I have patients on abiraterone for over a decade, several of them. So is it appropriate to give PARP inhibition for long periods of time? I don't think so based on what I know.

Audience Member: But anyway, so it's a reasonable option in the setting of the trial.

Johann de Bono: For sure. Speak to Shahneen, yeah.

Audience Member: Thank you.

Giulio Francolini: Good morning. Giulio Francolini from the University of Florence Radiation Oncology. What about combination with SBRT? We saw that patients harboring HRR mutation may have an increased effect by SBRT in oligometastatic hormone-sensitive scenario. Do you think there is some space to understand how to combine SBRT and PARP inhibition? It's for all the panelists.

Johann De Bono: Not so.

Audience Member: Mine is just—my question is sort of building on what Anna had touched on but wasn't actually answered by the panel. And something I've been thinking about in recent months is the issue of MDS. And in the context of CHIP or clonal hematopoiesis, we know that the incidence in a normal population about age 40 is 2%. And the incidence of that goes up to approximately 10% by the time you get to 70.

And I wonder whether—and we know that we can actually—there are assays that you can do, looking at baseline samples, and then in the context of somebody getting some of these toxicities. I wonder if the panel, one, knew whether there was this data set to try and understand, in fact, is the subset of patients that are getting these sorts of toxicities, do they have a preexisting predisposition that we could potentially identify prior to actually putting patients on?

And if not, could I suggest that we start thinking about building some of these assays into the subsequent studies, particularly in the context of early disease?

Johann De Bono: Agreed, Shahneen. We've done quite a lot of work in CHIP. We don't see a higher frequency of CHIP in prostate cancer subjects, which is why we were looking. But I think this is a really important question for sure.

Elena Castro: So there is a question from one of the panelists, I think, for Dr. Mehra. And regarding increasing tumor mutational burden on repeat testing in plasma, I suspect that could be from more passenger subclonal mutations accumulating over time from multiple sites being sampled rather than true high TMB. Was that looked at?

Niven Mehra: Yeah. Excellent question. So I think this is data we are not sure of—if this is multiple clones which all give a higher mutational burden. But we have some data which is not published yet. But we did whole genome sequencing on 400 genomes.

And we—especially in late mCRPC, where we biopsied in fourth, fifth line, the visceral metastasis, especially the liver—we were seeing actually an enrichment in high TMB in these patients. And when we went back to their original tumors, it was not present. So I'm not sure yet. I think it could be a late event which arises in the end stage in mCRPC.

Audience Member: So I'm not sure if we have time to answer this, but I was going to bring up this issue about the control arms in some of these clinical trials, which have been brought up by two speakers. And there doesn't seem to be a consensus question that has been submitted on whether these are appropriate for our patients. But I would just like to sort of get people's opinion.

Silke and I have actually published on this, that I consider these ARPI switch control arms some potentially unethical, in the sense that they're not transparent, which is the problem. They're not in the title. The ARPI switch implications are not in the consent form, and they're not in the protocol, and they're not justified.

I think it's a big problem. I think there's going to be a day of reckoning from our patients who are going to say to us, why did you allow me to go on a trial when the control arm was substandard? And the only people that are endorsing this control arm, it appears to be, are the FDA, whose mandate is simply to find something which is vaguely applicable to the US population.

It would be far better if the control arms on all these studies were physician's choice, thereby allowing us to do the best for our patients, not the best for the FDA.

Johann De Bono: Anthony, can I take that? Remember, for PROfound, we had crossover. So I think that's maybe a bit of a harsh criticism. And actually, the world has changed since we were designing the PROfound trial quite substantially, i.e. CARD data.

But also we do have data from Alan Bryce and Karim with rucaparib, where actually rucaparib beat docetaxel. So I think your comment is a little bit harsh, but I agree with you, the days of a second hormone therapy are gone.