PARPi Maintenance in HRD+ Ovarian Cancer: The PRIMA Trial

Bradley J. Monk, MD, FACOG, FACS:Let’s get into the new ESMO data. We had 3 phase III PARP inhibitor trials at the European Society of Medical Oncology Congress in Barcelona, Spain. And they were the first 3 presentations in the Presidential Symposium, which, as at ASCO [the American Society for Clinical Oncology Annual Meeting], is the plenary. The first 3 main trials were PARP inhibitors. I’m going to describe to you the 1 that I was involved in as the last and senior author. It’s called PRIMA. PRIMA, also called GOG-3012, was a randomized trial of 487 patients, 2 arms. It took the highest-risk patients, newly diagnosed advanced ovarian cancer, but patients who had incomplete debulking or patients who needed neoadjuvant chemotherapy and randomized them 2-to-1 to receive niraparib initially at 300 mg once daily. Ultimately, over time, we said let’s start our less than 77 kg or less than 150 mm³baseline platelet count patients at 200 mg once daily and randomized them to maintenance, 3 years of maintenance niraparib or placebo. And the primary end point of that trial was HRD. Again, this is becoming an HRD.

Thomas J. Herzog, MD:And a hierarchical at the end, where the step down was the intent to treat after that.

Bradley J. Monk, MD, FACOG, FACS:And it met its primary end point.

Thomas J. Herzog, MD:It did.

Bradley J. Monk, MD, FACOG, FACS:And simultaneously published inTheNew England Journal of Medicine.

Thomas J. Herzog, MD:Right.

Bradley J. Monk, MD, FACOG, FACS:Again, the point of that was that in the HRD-positive beyondBRCAgroup, there’s an opportunity for PARP maintenance. Because we already had PARP maintenance in theBRCA-mutated patients, with olaparib, in a study called SOLO1. But we want to do beyondBRCA. I have a little scorecard here, a little paper that talks about the hazard ratio. The hazard ratio for theBRCApatients in PRIMA was pretty close to SOLO1, so that’s easy.

Thomas J. Herzog, MD:It was 0.4 versus 0.3, yes.

Bradley J. Monk, MD, FACOG, FACS:Yes, but they were higher-risk patients, and if you look at the high-risk patients in SOLO1, it was actually 0.37 and 0.44.

Thomas J. Herzog, MD:Yes. I think that’s 1 thing we need to point out from the get-go. We’re going to start throwing hazard ratios and confidence intervals around here, but it is important to understand that these are distinctly different trials based on a number of factors. The inclusion criteria, who was eligible, and histology differed a little among the 3 and SOLO1. Obviously SOLO1 was just germline with a couple of somatic mutations thrown in. And if we look at the statistical-analysis plan, it was somewhat different for all of these. The primary end point differed among these trials, and the patient population in terms of number of stage III, stage IV, as you said, was interesting. To be stage III here, you had to have residual disease. As you can see by the control arm, it’s a pretty high-risk population.

Bradley J. Monk, MD, FACOG, FACS:We’re sports fanatics. He always says these guys have to work on their technique. I learned a new technique at ESMO. The technique is, let’s take an exploratory end point, make a definitive conclusion, and then cross-trial compare.

Thomas J. Herzog, MD:On an unplanned analysis.

Bradley J. Monk, MD, FACOG, FACS:And then cross-trial compare.

Thomas J. Herzog, MD:That’s perfect.

Bradley J. Monk, MD, FACOG, FACS:That’s not a good technique.

Thomas J. Herzog, MD:No.

Bradley J. Monk, MD, FACOG, FACS:That’s OK, but that’s what we need to be very sensitive about. And the take-home message is that PARPs work in frontline maintenance beyond BRCA, period.

Thomas J. Herzog, MD:Yes. All 3 of these trials hit their primary end point. We’ll get into that, but all 3 of them hit their primary end point. Obviously, we’re probably not going to get to the Presidential address unless you hit your primary end point.

Bradley J. Monk, MD, FACOG, FACS:That’s exactly right.

Thomas J. Herzog, MD:So they were all successful trials.

Bradley J. Monk, MD, FACOG, FACS:In PRIMA, if you were HRD-positive and you had maintenance treatment, you had an 11-month improvement in progression-free survival [PFS] from 18.2 to 19.6. And I’d take theBRCAs out of that. So that’s HRD non-BRCA, 11-month improvement in progression-free survival. Do you like that? Will you use niraparib in the HRD-positive patient population for maintenance?

Thomas J. Herzog, MD:Yes. I think the data are very compelling, I really do. It’s a simple intervention, and I thought they were very impressive data. I don’t know how you interpret the HRP, the proficient, but those who wereBRCAwild type and did not demonstrate any HRD, I call them HR proficient or homologous recombination proficient.

Bradley J. Monk, MD, FACOG, FACS:HRD-negative and HRD proficient are the same, sorry.

Thomas J. Herzog, MD:Yes, exactly. Just to be clear on that.

Bradley J. Monk, MD, FACOG, FACS:They didn’t do well. There’s less than a 3-month improvement in progression-free survival.

Thomas J. Herzog, MD:But they hit their hazard ratio and it didn’t cross 1. At least there’s a signal there, because you can look at the mean. And that’s just the morphology of the curves. They come together, but you can still have some benefit. The point being that, again, we don’t have this completely figured out because there shouldn’t be any benefit in that group if we’re accurately measuring the homologous recombination deficiency. There are still some patients who benefit, which gives me real confidence that in those who do have HRD, we’re really making a difference. That’s my point. It’s not that I want you to go out and treat everybody who’s HR proficient, because I agree that median is in the gray area of whether that’s going to carry water. But it really builds great confidence toward those who have homologous recombination deficiency.

Bradley J. Monk, MD, FACOG, FACS:One of the things that I was struck by is that it was a high-risk population, but if you were HRD-negative, the median PFS was 5.4 months. So not only are PARP inhibitors not going to work, but nothing is going to work because, remember, HRD predicts chemotherapy sensitivity. We used to think that the biology of the tumor was secondary to the size of the knife, that if you just try harder and cut all this out, it’s going to make it better. Listen, if you’re HRD-negative, that is the high unmet medical need now in ovarian cancer. Because even if you have a large amount of residual disease and you have the right molecular signature biology, you’ll respond to chemotherapy and a PARP.

Thomas J. Herzog, MD:One of the things I found most interesting in SOLO1, which looked at olaparib for those with mutation, was how poorly the control group did, right?

Bradley J. Monk, MD, FACOG, FACS:Yes, right.

Thomas J. Herzog, MD:These patients had to have a mutation. I think our bias was these patients are highly sensitive to DNA-damaging agents. Even if you were in the chemotherapy arm with no PARP inhibition, you’re in the control arm there, so you would have a longer PFS. And the PFS in that group was 13.8 months.

Bradley J. Monk, MD, FACOG, FACS:Around 14 months, yes.

Thomas J. Herzog, MD:Yes, 13.8 months in that group, I think many people thought that the control arm would outperform, and it didn’t. And if you look at the curves, the drop-off in that first 6 months means if they’re dropping off then, what happens? They’re platinum resistant and our options are very few. To be able to extend that and push that further I think is important, and I think you’ll be able to do that with these PARP inhibitors.

Bradley J. Monk, MD, FACOG, FACS:Last year at ESMO, SOLO1 taught us that inBRCA-mutated patients, there is a big benefit, but that if you did nothing, they did poorly. At this ESMO, beyondBRCAin the HRD-positive setting, it’s not quite as big a benefit but still clinically important that the HRD-negative patients do poorly.

Transcript edited for clarity.