Role of PARP Inhibitors in Advanced Ovarian Cancer

Transcript:

Bradley J. Monk, MD, FACOG, FACS: HRD [homologous recombination deficiency] is the inability to functionally repair double-stranded DNA breaks. What happens is when there’s a double-stranded break, you line it up to the sister chromatid, the homologue, and you recombine it. If you can’t do that homologous recombination, you’re deficient. Tell us what the relationship is to the inability to repair double-stranded brakes, HRD, and PARP sensitivity.

Robert L. Coleman, MD, FACOG, FACS: The idea is that you have 2 strands of DNA that are aligned with the appropriate nucleotides. You have ATGC, they must go this way, and when we lose this information in one of the strands, PARP gets recruited and develops a scaffold by which these nucleotides can be repaired. As we do this, if we miss this, we know what this is, and we know what must be here, it’s easy. That’s what PARP does, it aligns that, and when you lose both, you have no template. As you mentioned with this invasion into the sister chromatid, we can reestablish one of these strands, we can do both at the same time, but one, and then that will provide the opportunity for the other. This is important because when we measure whether a cell has the capability of performing homologous recombination, it gives us the opportunity to leverage PARP to induce the probability of this happening in the tumor, and knowing that the normal cell can fix it, and the tumor cell can’t. That is the selective sensitivity we see by combining a biomarker such as homologous recombination deficiency with a drug that works in that setting, which is the PARP inhibitor.

That’s the concept, and all the data that we’ll talk about here have supported that. That’s why we’ve moved forward with that. Now, as you mentioned, I think it’s important that outside of the BRCA mutation itself, the HRD scores that we rely on represent a spectrum. You mentioned that 16% cutoff for Foundation Medicine, a test that’s available in the United States; the Myriad MyChoice, which uses a score of 42%, represents a cut where we think we have the lowest number of false positives and false negatives. It is a spectrum. We’ve seen this even exercised in some of our clinical trials where we’ve used different cut points to define what is an HRD test positive and test negative tumor. It’s important because I think now we get that information with our result, not only as a qualitative end point of positive or negative, but we also get the actual score so we can make a decision.

Bradley J. Monk, MD, FACOG, FACS: If you have HRD and PARP inhibition together, they’re synthetically lethal, but individually they’re not; that’s how we always say that. The other thing you say, and you said it better, is that the mutations are there in the genomic instability score, or HRD, but that’s not the whole story, you need to look at the effect, the genomic scarring. Let’s talk about the 3 agents; they’re not the same dose or indication. Niraparib is once daily and it’s individualized, rucaparib has some LFTs [liver function test changes], and olaparib is metabolized in the liver.

It’s a frontline world, and we’ll get into recurrence, but let me summarize the 2 studies that I was first and last author on, and then maybe you can talk about olaparib in their first-line studies. Niraparib was approved 3 years ago, almost to the day, in all comers in newly diagnosed advanced ovarian cancer that responds to chemotherapy, with an individualized dose. If you weigh less than 77 kg, or have less than a 100,000 platelet count, you would get 200 mg, otherwise 300 mg. That was for all comers, but it was clear that it worked the best in BRCA [mutations], and pretty good in HRD, which is non-BRCA half the time; but it got a broad label. Then we did a confirmatory trial, you and I, which was another PARP inhibitor, rucaparib, which was published and presented last year on June 6th at ASCO [American Society of Clinical Oncology annual meeting], which was complementary. It proved the concept that these PARP inhibitors—now with rucaparib, which has more dosing flexibility, 600, 500, 400, 300 mg—could be impactful in BRCA and in HRD, but still had a small benefit. That’s rudimentary. But olaparib is quite frankly, the first in class and has a breast, pancreas, and prostate indication. Tell us about the olaparib development plan. I love this story.

Robert L. Coleman, MD, FACOG, FACS: A lot of what you just mentioned about the frontline setting came from a very similar experience we had using the same design in platinum-sensitive recurrent disease. We had patients who responded to a platinum-based combination, and had evidence of that response, and then were randomized to placebo versus PARP with those 3 drugs, olaparib, rucaparib, and niraparib. All of them showed almost to the same magnitude, to the point estimate of that effect of delaying progression or death in patients who had recurrent disease. We applied that signal strongly in the frontline setting. The first [trial], as you mentioned, was SOLO1, which brought us olaparib in patients whose tumors carried a BRCA mutation and demonstrated a profound effect reproducing what we saw in the recurrent setting. It had a hazard ratio of 0.3, a very strong effect on treatment. What’s been neat about that story is that since that was first presented and published shortly thereafter in 2018, we have seen the development of long-term outcomes with these patient populations. We talk about cure in ovarian cancer being difficult to do in the recurrent setting. Well, in the frontline setting where we have the greatest opportunity for that, we now have 7-year data of nonprogressive events. The curves are flat, showing about a 2-and-a-half-fold increase in the probability for nonprogression at that point in time. We’re starting to bump up against what we might be considered a real cure in ovarian cancer. It’s something we haven’t had a an opportunity to talk about in the past, so this has been exciting. This has now translated into suggestion of an overall survival effect that is looking very promising for us.

Of course, now we’re applying this active therapy in another cohort of patients. These are what you mentioned already, the patient population who have been treated with induction chemotherapy, who had a response, and then were randomized to placebo versus olaparib. In PAOLA-1, we had the opportunity to take the other group of patients that were started on chemotherapy plus bevacizumab and do the same thing, and we got the same result, that it improved it over bevacizumab. Now we’ve tried to do this cross-trial comparison and ask, is it better than olaparib alone? It’s very hard to sort that out. We don’t think that BRCA status, mutation or nonmutation, is a biomarker for bevacizumab effect. If there was a treatment effect there, we would expect it to be only additive. There’s a suggestion that might be the case, but those patient populations have been defined differently because we have patients who went on to bevacizumab for specific reasons. The good news is that now these 3 drugs, olaparib, niraparib, and rucaparib, all basically show proof of principle that indication of patients who have a biomarker aberration. In PAOLA-1, it’s now linked to the indication, with HRD test positive, that these PARP inhibitors seem to extend our survival end points, most notably progression-free survival in annotated populations. It was a strong signal.

Bradley J. Monk, MD, FACOG, FACS: A strong signal and an exciting patient-changing result, a survival benefit for both olaparib frontline trials, SOLO1 and PAOLA-1. In BRCA you can do without bevacizumab all 3 PARP inhibitors: olaparib, rucaparib, niraparib. If it’s HRD non-bevacizumab, now you’ve got 2: rucaparib, ATHENA [trial], and niraparib…. In HRD though, it’s olaparib and bevacizumab if you want the combo.

We didn’t talk about adverse events for PARP. It’s 3-fold and it’s a class effect, because they’re chronic, it’s fatigue; because they are oral, it’s GI [gastrointestinal]; and because they’re cytotoxic, it’s marrow suppression. We must, particularly with niraparib, monitor the counts every week, and then ultimately be mindful of fatigue and the GI adverse reactions. Niraparib is once daily, so there is an opportunity to give it at night. All these can be given with a small meal.

Robert L. Coleman, MD, FACOG, FACS: This is a great time to talk about it because a lot of the recurrent spaces had been annotated by these PARP inhibitors. Because of the movement into the frontline setting, most of our patients who are best candidates for getting it, are already receiving it, so I think this is an appropriate place to talk about that. The one adverse effect that doesn’t get a lot of attention is insomnia. This is an interesting story; I was giving a lecture, and I had a PhD student in the audience. I said we’ve seen insomnia, but I don’t have a good explanation for it. This guy came up and he goes, I’ve been studying circadian rhythms and the impact of PARP inhibitors on metabolic processes that follow circadian rhythms. He said they saw that PARP1 can cause a disruption of circadian rhythm for certain metabolic processes. I was like, “Wow, OK, that makes sense.” It’s something that people recognize. It doesn’t get a lot of attention because people will combine it with…and fatigue. It’s just something also to keep an eye out for.

Bradley J. Monk, MD, FACOG, FACS: Niraparib inhibits catecholamine reuptake, so there is some low level of hypertension, which can cause some jitteriness and that sort of thing.

Robert L. Coleman, MD, FACOG, FACS: Exactly.

Transcript edited for clarity.