The Uromigos Episode 154: ASCO GU 2022 MAGNITUDE Trial of Niraparib and Abiraterone in Metastatic CRPC

Kim Chi, MD, describes the results of this randomized phase 3 study.

Episode Transcript

Brian:

Hey, welcome everyone to our ASCO GU 2022 podcast series. We are here with Kim Chi today to talk about some more PARP inhibition data, which seems to be a theme at this year’s ASCO GU. So, Kim, welcome. If you want to just introduce yourself briefly, then maybe you can launch into just basics of the study design, and we’ll go from there?

Kim Chi:

That’s great. Yeah, thanks for having me on, Brian and Tom. And I’m glad you’re able to highlight our MAGNITUDE study. So, I’m Kim Chi. I’m a medical oncologist at BC Cancer, Vancouver Center, and a Professor of Medicine at University of British Columbia. So, the Phase III MAGNITUDE study is a double-blind randomized placebo-controlled study. But I think what is unique about it is that it has a prospective, a prior approach to looking at the outcomes of both biomarker-positive and the biomarker-negative cohorts.

That is, patients that have metastatic CRPC, they were pre-screened at study entry to determine whether they had a gene defect in a homologous recombination repair. This was based on a nine-gene panel, including BRCA1/2, ATM, and others. And then they were allocated to either a biomarker-positive cohort or a biomarker-negative cohort. Then subsequently randomized to either niraparib, which is a potent PARP1 and 2 inhibitor, in combination with standard-dose abiraterone versus placebo plus abiraterone.

For the biomarker-negative cohort, there was a prespecified futility analysis in the biomarker-negative population, after enrolling about 200 patients and after 125 progression events. For the biomarker-positive cohort there’s … the first look was at the BRCA1/2 subgroup. Because we know that PARP inhibitors have probably the most activity in patients with BRCA1/2 gene alterations. So, this was tested first. And then if … then the overall biomarker-positive cohort was to be tested. So, it’s a little more complicated kind of design. However, it really is testing whether the biomarker makes a difference in testing that clinical utility.

Tom:

Kim, why could you not just do an ITT trial? Include everyone, but the primary endpoint being the biomarker-positive and then the secondary endpoint be the ITT population?

Kim Chi:

That could be done. But if you’re … and what you’re going to do is a secondary analysis, that’ll be retrospective as you look back at that patient population. Because we know that some of those population, they’re not going to be able to be categorized as either biomarker-negative or biomarker-positive, because testing will fail. In the MAGNITUDE trial, patients that were biomarker-positive had to at least have one test, and all patients were tested based on tissue and blood-based assays. While in the biomarker-negative cohort, they had to have both tests done, to show that they were biomarker-negative. Now they may not have had a result for both assays, but at least one of them had to be negative.

Brian:

And can you talk about the panel that was chosen? Because we’ve done a few of these. As Tom mentioned, talked to Pete Nelson on podcast, and it seems like every drug sort of selects different panels. Some more narrow, some more broad. Was there previous data with this particular agent, and that’s why you selected it? And you also made it hierarchical, where you did BRCA first and then only tested the others. So, just talk about that a little bit?

Kim Chi:

Yeah. In part, the biomarker panel was because initially when niraparib was being developed, it was being tied to a liquid biopsy assay. And that was sort of the genes that were validated around that assay, that were also relevant of course for homologous recombination repair. And then the GALAHAD Phase II study, which was actually just recently published in Lancet Oncology, demonstrated the activity with using this gene panel. But of course, I think we all know that different genes have different sensitivities or are associated with different sensitivities to PARP inhibitors.

With BRCA2 being the most frequent gene alteration, but also the most sensitive to PARP inhibitors. While with the other gene alterations there’s kind of a mixed response, particularly with ATM, CDK12. Then once you’re starting to get into the other gene alterations, these start to become quite infrequent and difficult right now to really understand what is the sort of benefit of PARP inhibitors in these gene alterations. There will be a gene-by-gene analysis done and presented in the near future. So, those will be interesting results as well.

Tom:

Kim, I’ve got two questions. I’m going to ask them one at a time, because it irritates Brian when I do both at the same time.

Brian:

It really does.

Tom:

It really does. The first question, you mentioned PARP1 and PARP2 inhibition, we haven’t really talked much before about the differences and the importance of those two separately? Number one. And number two is therefore, are all these PARP inhibitors the same or are they slightly different?

Kim Chi:

Well, niraparib is I think one of the more potent inhibitors of both PARP1 and PARP2. And that their … the importance of those, I think it’s … hasn’t been clinically validated. But beyond the direct enzymatic inhibitor inhibition of PARP, there’s also PARP trapping. And that may be more relevant to its activity. Because there have been some PARP inhibitors that have been potent inhibitors of the enzymatic activity, but less potent in terms of PARP trapping.

So, there is some data suggesting that niraparib and talazoparib are more potent inhibitors of PARP trapping or potent promoters of PARP trapping, while something olaparib is a less potent promoter of PARP trapping. However, that’s preclinically. I don’t know if there’s much of, if we can say that there’s any difference between the activities of the PARP inhibitors at this time.

Tom:

In our previous podcast, Pete talked a lot about this. Not like … a bit about this trapping and seemed to feel that was really important, so the distinction really in PARP1 and PARP2 remains somewhat academic at the moment. The other piece I wanted to ask you about is there’s a confusing issue in the current status, we’ve got a positive trial with olaparib versus placebo for OS in a biomarker-selected population. That has a license in the US. Yeah, I think a broad license in unselected patients, correct me if I’m wrong. But in Europe, as a license which is selected only for a smaller cohort of patients. Is that correct?

Kim Chi:

That’s correct. In the US, it’s for the whole panel, both the cohort one and cohort two …

Tom:

But is BRCA1 …

Kim Chi:

Or A and B in US. BRCA1 and BRCA2, ATM was in the cohort A.

Tom:

How do you …

Kim Chi:

And then in the others, it was a series of other alterations in the cohort B. And so, the US FDA approved all the biomarkers. In Canada, it was just BRCA1/2 and ATM. In Europe, I think it’s just BRCA1 and 2.

Tom:

And so therefore, what’s the current position of niraparib and other PARP inhibitors beyond olaparib? In terms of where they can be used in, from a global perspective?

Kim Chi:

Well, niraparib in … is approved for ovarian and other selected gynae cancers. But for prostate cancer, of course it’s not approved as of yet. And if we look at the MAGNITUDE study, if we were to look at the alterations, I would say for biomarker-negative, that is without an identified homologous recombination repair gene alteration. Given the toxicity of the combination, both for the patient as well as financial toxicity, really, we need to be selecting those patients that will optimally benefit from the combination.

Brian:

So, Kim, let’s dive into the results. And maybe as a prelude to that, talk a little bit about the rationale for combination with Abi. And was it specific or was it just any hormonal agent, maybe just touch on that and then let’s recap the results?

Kim Chi:

Yeah, so there’s both the rationale in terms of bringing PARP inhibition earlier in the disease for the patients that are biomarker-positive, that have an HR gene alteration. And because we know these patients have been associated with poorer outcomes and poorer prognosis, we did a study, a prospective study. And those that were identified as having a BRCA2 alteration actually did five times worse than those without, on first-line hormonal therapy, either abiraterone or enzalutamide for metastatic CRPC. So that the rationale to bring it earlier, of course is one rationale. As you, we all know, the patients with most aggressive disease often don’t go on to get subsequent line of therapy because they have such aggressive disease. So, bringing it early is important.

I think the other rationale to combine it with abiraterone or an inhibitor, that is with PARP inhibition, is because it’s been shown in preclinical models that androgen receptor signaling regulates DNA repair in prostate cancer. And so preclinical models suggests that combined AR targeting and PARP inhibition, in disease that is HRR-intact, may in fact have additive or synergistic activity. So that’s why the MAGNITUDE study was designed a priori, to really test those two groups.

Now in terms of the results in the biomarker-negative population, there was … the futility was met. The hazard ratio for that early futility analysis for the composite progression endpoint was 1.09. And the criterion futility was defined as anything greater than 1.00. This analysis went to the independent data monitoring committee. Then given that added toxicity of the combination, but no added efficacy, it was recommended that enrollment was to be stopped in this cohort.

Tom:

And Kim, that composite endpoint was radiographic or PSA progression?

Kim Chi:

That’s right, or death. And whether it was … that there was about equal number of PSA events and radiographic progression-free survival events. And whether you looked at it either way, the hazard ratio was above 1.00 for both those events.

Tom:

Kim, let’s pivot to the primary endpoint of the biomarker-positive population. Do you want to just talk us through that?

Kim Chi:

Yeah. So first, it was tested in the BRCA1 subgroup. This was about 110 in either arm, of niraparib versus placebo. And so, for the primary endpoint of radiographic progression-free survival or death, it significantly reduced this by 47% compared to control. This was as assessed by a blinded, independent central review. And now since that was positive, then we went to the overall biomarker-positive cohort. Again, there was a significant reduction in the risk of progression or death by 27%, as assessed by independent central review.

I would like to remark though that that BRCA1/2 mutated group, the placebo group, those that just got abiraterone, it was a median progression-free survival of 10.9 months. So, pretty short. With niraparib, it was 16.6 months. Now if we recall abiraterone efficacy in the 302 study, as well as more recently in the ASO study, the median radiographic progression-free survival for an unselected all-comers population was around 16 and a half months. So really, those BRCA1/2 patients that are mutated do considerably worse.

Tom:

Kim, and so it’s prognostic and predictive, what would happen if we looked at all HHR … sorry. HRR biomarker group, but we removed the BRCA1 and BRCA2 patients? Would the results still be significant?

Kim Chi:

Well, we did do a subgroup analysis. There is a consistency of a fact, and really statistically there is no heterogeneity of effect with niraparib plus abiraterone radiographic progression-free survival. However, the point estimate is 0.99. But the confidence intervals stretch quite wide, from 0.68 to 1.45. Yes, if you do look down at the subgroup removing the BRCA1/2-altered, there seems to be, I’d call it a watering down of effect.

But I think we’re going to go … we’re going to see a gene-by-gene analysis in due time, because we know that there are probably some alterations. Like ATM and CDK12, which were included it, that probably may not be as benefiting as much from PARP inhibition as some of these other gene alterations.

But at the same time, I think we need to also look at some of these secondary and other endpoints where it starts to get a little more complicated. Yeah, for the radiographic progression-free survival endpoint, the primary endpoint, we’re seeing a bit of a watering down effect. But we actually don’t see that for some of the secondary and other endpoints. So, for example time to cytotoxic chemotherapy, the hazard ratio was roughly 0.59, 0.58 for both the overall biomarker-positive cohort as a … 1/2 subgroup. And similarly, time to symptomatic progression, the hazard ratio for the overall biomarker-positive cohort was very … was essentially the same as for the BRCA1/2 subgroup.

Brian:

So, Kim, why that is, is because our PFS used PSA as well and that kind of diluted the effect? And then these others …

Kim Chi:

No.

Brian:

Are more real effects, or …

Kim Chi:

No, our PFS for the primary endpoint in the biomarker-positive group was … did not include PSA progression. So, that did not include PSA progression. Yeah. I’m not quite sure why it hasn’t, and maybe that the events haven’t happened? Some of these alterations probably have a better prognosis, such as PALB2 and ATM probably have a better prognosis than BRCA2. So, perhaps we haven’t seen that emerge. But we’re seeing some of these other earlier endpoints emerge perhaps, or these symptomatic endpoints. Hard to say, I’m not quite sure why we’re seeing these other benefits.

I think that also goes if we go look at the objective response rate, and here is very intriguing. In fact, the objective response rate is nearly doubled in the overall biomarker-positive cohort. Even in both partial responses and complete responses, and similarly in the BRCA1/2 subgroup. So, there isn’t a watering down, in fact the objective response rate is even higher when you include the biomarker-positive cohort, compared to the BRCA2s in the BRCA1/2 group.

Tom:

When we look at your overall survival curve, they are kind of on top of each other. It looks like one of my survival curves.

Brian:

Better actually, but …

Tom:

How does that … how do you square that?

Kim Chi:

Oh, we’ve all done those studies where the overall survival curves are on top of each other. But there’s a couple of … number one, it’s immature. There’s, it’s only a medium follow-up of 18 months. And we know the median overall survival for unselected patients is upwards of 30 months in first-line therapy. So, it is immature. And only 46% of the required, death events have occurred for, with the final analysis. So, this is the first interim analysis.

Now, I didn’t mention this at the beginning. But in the baseline characteristics, although they were evenly matched, there was a preponderance numerically with patients that were ECOG performance status 1 and visceral metastases randomized to the niraparib arm. Numerically, there was a bit of diff difference.

And so, in a multivariate analysis, a prespecified multivariate analysis accounting for those baseline prognostic factors, there was actually an overall survival that favored the niraparib arm with a hazard ratio of 0.77. P-value not statistically significant, but less than kind of 0.20, if you want to use that as a trending effect.

I think one of the reasons is because as you … as I mentioned earlier, this was specifically powered in that BRCA1/2 subgroup. A lot of the BRCA1/2 patients didn’t get on until the later part of the trial, and so therefore haven’t had the events, haven’t had the same sort of follow-up. In fact, the follow-up for the BRCA1/2 subgroup is actually less than 18.6 months. Even less than 18.6 months, if you look at that group specifically. So, I think it’s just too early, and time will tell.

Tom:

Kim, survival advantage for olaparib in the biomarker-positive cohort, particularly the BRCA1/BRCA2 was at an FDA and an EMA level label. Why not just, I’d test the patients and wait for the cancer to come back and treat at relapse?

Kim Chi:

Yeah, I mentioned this earlier. It’s because … that that is one option. But I think we know that patients, particularly with BRCA2 can do very poorly and have outright resistance to primary treatment. And I think getting treatment in earlier is always better, so that the patient has the opportunity to get that treatment. And earlier treatment, just by even a … earlier may have significant benefits. So, if we take the PROfound study, where the radiographic progression-free survival difference was only three months and a large proportion of patients crossed over at radiographic progression-free survival, they … there was still an overall survival. So, delaying therapy by just a few months in that population had an impact on overall survival.

Brian:

Was there crossover in this study? Either on, as part of the study or off-study so to speak?

Kim Chi:

Yeah, and there was no crossover as part of the study. But patients could be requested to be unblinded, and the study steering committee felt strongly that this was the ethical thing to do.

Brian:

Mm-hmm (affirmative). Absolutely.

Kim Chi:

As soon as patients had radiographic progression-free survival, the investigators could request unblinding. And then they could go a therapy of their choice.

Brian:

And do you know if some of those placebo patients may have gotten PARP inhibitors subsequently?

Kim Chi:

We do know some got PARP inhibitors, some got platinum. And as we do the next survival analysis, we’ll also look at PFS2 and look at the subsequent therapies they received.

Brian:

So, if overall survival doesn’t mature and become positive, is this enough to impact practice? Where would you go next, what do we do with that rPFS data?

Kim Chi:

Yeah. If overall survival, I think we’ll have to look at why overall survival isn’t impacted. When we can see it impacted for, for example, olaparib in the second-line setting. And is it because of the crossover? I think what would be of interest is PFS2. Because PFS2, not just in prostate cancer, but in other cancers as well, has been a good indicator of overall survival benefit or at least the power of a therapy. So, if PFS2 is improved because we have all sorts of patients crossing over to receive platinum or PARP inhibitor, that may explain that lack of an overall survival advantage. However, that’s conjecture. So, we’ll have to see what happens, and we should see the second and final interim analyses within the next year.

Tom:

Kim, we’ve not had the chance to talk about adverse event profiles. We probably should because it’s important.

Kim Chi:

Oh, for sure.

Tom:

What’s going on with the adverse events? How long are patients on therapy for? How many discontinuing for adverse events, and what are those adverse events?

Kim Chi:

Actually, I don’t, I can’t recall what the median dose or median treatment duration was. That’s just escaped me at this moment. But in fact, there was actually a relatively minor number of patients who discontinued niraparib or placebo for adverse events. It was 11% on the niraparib arm and 5% on the placebo arm. And median relative dose intensity was very high, 99% in the niraparib group.

Now the side effects are what’s been consistent with what’s been reported for niraparib, as well as other PARP inhibitors, including hematologic toxicity as well as nausea. There were more grade 3/4 events, 67% on the niraparib arm and 46% on the placebo arm. Almost all driven by anemia, which of course is manageable with interruptions and of course transfusions and supportive care. Yeah, there are more adverse events. And that’s why I think it comes back to, we really need to identify the patients that are going to benefit fit from these more toxic treatments.

Brian:

And so, Kim, what’s next for this combination then? I mean first of all, congrats on the big study and also for including the biomarker-negative. As you said at the outset, that’s not usually done in very difficult to do, right? It costs you an extra 600 patients, which is quite a big resource. But what’s next, I mean is … again, are you waiting for overall survival? Seeing what happens with maturity, planning other studies in this space?

Tom:

You’ve written down here the MAGNITUDE study supports as a new first-line treatment, niraparib plus … I … do think it is a first-line treatment? Because when you say, “Supports,” it’s not a first-line study? So, it either is, or …

Kim Chi:

Well, I …

Tom:

It’s lying?

Kim Chi:

You can’t say that it is. Because it hasn’t had, received regulatory approval yet.

Tom:

In your opinion, do you think it should get regulatory approval and would you approve in just the BRCA1/BRCA2 population?

Kim Chi:

I’m actually a big believer in … so number one, I think it should get approved. I would support it getting approved, and I would use it. Now, I would definitely use it in the BRCA population. And then, I am a big proponent of having wide criteria for the other gene alterations. Now you could argue CDK12 and maybe ATM, and particularly the other gene alterations, they occur at such low frequency. We really need to be able to study these in real world populations, bring all this data together. So, we can understand, what are the biomarkers? What are the genetic biomarkers at least, that help us select patients that will get benefit from PARP inhibitors?

Now I think, Brian, you were asking what’s next? Well, obviously we want to know what the next interim analysis shows. I want to know; we want to see that gene-by-gene analysis. More detailed health-related quality of life, and I think there’s a whole host of data that we can mine from this. Particularly for example, we didn’t touch on what kind of assays, right? Does it matter whether you do a tissue or a plasma assay, also what kind of gene alterations benefit the most? What about germline alterations, et cetera, et cetera? So, I think we’ve got a wealth of data to mind from, from this study.

In terms of the combination, I think going even earlier is going to be important. Because again, these patients do badly, particularly BRCA2. And in that sense, there’s the AMPLITUDE study. Which is accruing now, which is looking at patients with metastatic castration-sensitive disease that are receiving ADT plus abiraterone and then randomizing those patients to receive niraparib versus placebo. So, I think going even earlier is going to be important.

Tom:

Kim, we know a press release that the olaparib trial in the frontline space is positive too, educated guess would suggest there’re more similarities than differences between the data. Back in the day, Brian got himself involved in a big ugly argument about, sort of there’re differences between axitinib and sunitinib and pazopanib and all these drugs. Is prostate cancer going to end up in the same place? With a series of approved PARP inhibitors not being tested against each other, with slightly different biomarker panels and will there be a big commercial debate? Or do you think the prostate community and the study point of view, actually there are better drugs and better biomarkers, and you’ll find the right place?

Kim Chi:

Yeah, I think the community … the industry is doing what it’s doing. We’ll have a series of PARP inhibitors that will be available to us, and then I think it’s up to us as investigators and claw-through groups to look at other questions. I don’t think we should get in an argument, into which PARP inhibitors better. We’re … we also don’t want to get an argument which androgen receptor inhibitor is better. I think there’s probably subtle differences. But I think it’s good to have choice and it’s good to be able to evaluate these things through, as investigators in cooperative groups.

Brian:

Yeah. Kim, I think that’s right. I think trying to do the drug A versus drug B versus drug C studies doesn’t get you or patients anywhere, I can certainly tell you that from kidney cancer days. And like you said, better biomarkers, refining that list. And that’s from the outside of it that list of which HR mutations and which are more important and less important, I think that’s the part that gets confusing. Is trying to refine, what biomarker are we talking about? Because right now, it’s sort of a pool of markers.

Kim Chi:

And we’re using it as a surrogate to understand homologous recombination repair, whether they’re actually intact or defective, right? Because not all patients with BRCA2 respond, and so we really need to even that. Are there other alternative ways? Can we use other immunohistochemistry assays, signatures, and so on and so forth to really better identify these patients?

Tom:

Kim, we just saw a presentation by Matt Smith who showed data similar to PEACE-1 with a survival advantage for essentially the triplet versus the chemotherapy doublet. Not abiraterone, I agree. Do we know the activity of … is this, these BRCA1/BRCA2 patients, are they more responsive to chemotherapy? Can we be using the triplet instead, is that a poor man’s PARP inhibition? Because obviously it’ll be easier, I suspect to give the triplet to all patients rather than having to measure the biomarker and select some for this. And of course, without a survival signal, I would guess that want to give the chemotherapy triplet. Is that a fair statement?

Kim Chi:

So, the chemo, this … that was a lot of questions as one.

Brian:

Welcome to the podcast, yep.

Tom:

I’m sorry, I apologize.

Kim Chi:

So, I think the doublet versus triplet, it’s a difficult question. Because what we’ve seen is adding an AR pathway inhibitor to ADT-docetaxel improves ADT-docetaxel. But what we haven’t seen, does docetaxel added to ADT-AR-PI improve ADT-AR-PI? And you can … there’s a consistency of effect. Of ADT-docetaxel, overall survival, hazard ratios in the 0.70 range, mid-0.70 range. But if you look at ADT plus AR-PI, the overall survival hazard ratio is actually quite a bit lower. It’s in the 0.60 range, low 0.60 range.

And if you look at the TITAN analysis where we looked at adjusting for crossover, because there’s so many patients that crossed over, the hazard ratio was 0.52 or 0.53 or something. So, I’m not actually quite convinced that the triplet regimen is the new standard. There’s never going to be that study of adding AR-PI to ADT-docetaxel.

So really, we won’t address that question. So maybe in some high-risk populations or very high-risk populations, I think we all would consider doing the triplet regimen. But I think people have also voted with their feet. If we look at use of ADT-docetaxel, even up to 2019, 2020, in our jurisdiction was about 10% of patients were getting ADT-docetaxel. I think there is a … biased against chemotherapy, for a variety of reasons. And there’s also data suggesting that patients aren’t even getting ADT intensification. Instead of pushing the triplet, I would just push any sort of intensification that our patient should be receiving.

Now, you asked about the PARP inhibitors or chemotherapy sensitivity. I think there’s varying data that they may be … they may do worse, but that could just be prognostic. Our data that we have locally suggests that they do about the same with chemotherapy, but I think they have substantially better responses with PARP inhibitors. So, in the castration-sensitive setting, I think we’ll have to wait for the data to see what it shows up as.

Brian:

Kim, this was really great, thanks for your time. Congrats on the data and really doing a big important study. That I think PARP inhibition seems to be the theme of this year’s ASCO GE, and these certainly contribute. Tom, anything else from you?

Tom:

Fantastic, Kim. Thanks so much for today.

Brian:

Yeah.

Kim Chi:

No problem, thank you for having me.