 What percentage of patients with advanced ovarian cancer get neoadjuvant chemotherapy? I’m going to see how many patients there are who have an opportunity for surgery because she said there are some who don’t. Give me a number of neoadjuvant chemotherapy cases; what do you think?

 Let’s say 50%. If you get neoadjuvant chemotherapy, you’re not eligible for secondary cytoreduction, period.

 We have 3 studies that show no benefit in that because you wouldn’t have been eligible for DESKTOP and SOC-1, and GOG-0213 said it didn’t work. What percentage are R0 [negative resection margin] out of the 50%?

If we’re talking interval debulking….

That’s right. I was going to say you’re excluding those outright, right? Primary debulking, R0, 50%.

 You’re now down to 25% who are at least eligible.

These are smart people; are these fair numbers?

 I think so. Right out of the gate, 75% of the patients don’t talk about it because they weren’t…primary debulking, and they didn’t have an R0. Then what’s the percentage of patients who are going to have a platinum-resistant recurrence because you have to have platinum-sensitive recurrence? Let’s say 20%. Now you go from 25% to 20%, and then what percentage of patients are going to have these isolated recurrences that you can resect?

Probably about a quarter or a third of patients.

 Yes, so let’s say a third. A third of 20% is about 5% to 10%. That’s OK; I’m not going to fight you.

That’s OK. I’m refereeing because you told me to. That’s OK because we’re looking for the people who shouldn’t have surgery, right?

That’s a good algorithm to go through, in my opinion.

 Thank you for that. What happens, I’m afraid, is that, as Dr Pinkal Desai, [MD, MPH,] who was Dr Randall’s and our mentor, said, “When you’re a hammer, all the world is a nail. And when you’re a surgeon, you’re looking for someone to operate on.”

He also said, “A good surgeon knows when to operate, and a great one knows when not to.”

 Thank you. That’s good wisdom; that’s fantastic wisdom.

Videos

Neoadjuvant Therapy Versus Surgery for Ovarian Cancer

 Matt and Shannon, I want to pivot here to the other most important abstract. I would say at ASCO [the American Society for Clinical Oncology annual meeting], the No. 1 and No. 2, I’m not sure one is better than the other, but that SOLO-2 overall survival [OS]; and maybe it’s not the most important because it’s not going to change practice. Let’s get into the secondary debulking conversation.

 that many of our listeners may have participated in called GOG-0213. It showed that a second operation wasn’t helpful. Tell us about that study, Matt.

We talked a little about GOG-0213 just a bit ago when we were talking about chemotherapy selection. This was a bit complicated because it was both a chemotherapy question and a surgical question that was asked. Part of the reason Rob Coleman, [MD,] has gray hair is mostly this trial.

There were 485 patients randomized; 67% of them had complete gross resection. Follow-up was good for the trial. Most of these patients had bevacizumab, which is the main difference from the other trial; 84% of the patients had bevacizumab. The hazard ratio favored the group for which you didn’t operate: 1.3. It was a year longer if you didn’t operate on them, from an OS standpoint. We scratched our heads after this trial came out.

There was definitely some patient selection that ended up being part of the randomization. The surgical morbidity was about what we expected. We anxiously awaited DESKTOP 3, and Shannon’s going to tell us about that.

 It’s interesting. I always say it takes 2 studies to convince anyone of anything. I don’t know why you guys have adopted the LACC trial with 1 study. The LACC trial was a cervical cancer trail that showed that if you have radical hysterectomy, open, you live longer than if it was minimally invasive. GOG-0213 didn’t change practice; it changed opinions a bit. The second, confirmatory trial was supposed to be DESKTOP 3, and that was presented at ASCO. Did it confirm or refute GOG-0213? Shannon, tell us.

It’s a little bit of both, right? It was a nicely designed study, and to Matt’s point, they were a bit stricter around the selection of the patients. They used their AGO [German gynecological oncology group] model that they had previously validated in other studies that includes looking at ECOG performance status, were they able to get to no gross residual disease after surgery the first time? Did they have ascites at recurrence?

They randomized patients based on if they met these criteria to either secondary cytoreduction or going directly to chemotherapy. They all got platinum-based chemotherapy. There was no mandate to include any other targeted therapies. In this study, about 22% to 23% of patients on each arm received bevacizumab. There was only a smaller population that had the chemotherapy with that drug, which we know can be a great equalizer between groups.

What they found was a significant difference in progression-free survival, which was one of the co-primary end points. We all thought, “That’s great; the tumor’s been resected, so of course it’s going to take longer for you to see something and progress. What we really want to see is overall survival.” That’s what they presented at ASCO this year, and they demonstrated a statistically significant and clinically significant difference in overall survival, with a reduction in the risk of death of about 25%, a hazard ratio of 0.75. That yielded about a 7-month difference in overall survival.

When they teased out the patient populations, that’s where things got interesting. The patients for whom the surgeon was able to achieve no gross residual did by far the best. The median was somewhere around 62 months for those patients.

GOG-0213 showed that too, that could be biology.

That’s right, but when you look at the population that got surgery and did have residual disease, they did worse than the chemotherapy alone arm. This told me that we need to do a better job of knowing how to select these patients. We all, the surgeons here, we all have patients in our practice for whom we did secondary cytoreduction who lived another 6 or 7 years, and we high-five ourselves, but you’re right, it’s the biology. We need to do a better job of selecting them.

For me, this says that there is a population that we should not operate on. And there’s a population that can potentially benefit from surgery. Where does bevacizumab add into that? Does that negate the surgery, and if you’re going to give bevacizumab, then it doesn’t matter? We’re still trying to understand it.

The other abstract that was presented at ASCO, the SOC-1 trial, which was a study that had a similar population; they did a bit different selection. They had a score that they utilized called the iMODEL that incorporates things like stage. Were they able to get R0 [negative resection margin] after their primary cytoreduction? What is the CA-125 [cancer antigen 125] level? Ascites?

In that group, they did not demonstrate a benefit in overall survival. We saw progression-free survival benefits for the patient population that got surgery, but they did not demonstrate a statistically significant difference in overall survival in that group. It’s still early, but it’s again a bit confusing.

When you look at the curves, it’s the same thing. The population that can get that R0, benefits. The population that gets surgery but does not get to no gross residual disease looks worse than the population that gets chemotherapy.

 In GOG-0213, the no surgery did better than the R0.

New Data for Secondary Surgery in Ovarian Cancer

 We’ve had some conversations about frontline PARP inhibitors. There’s no question that PARP inhibitors were first brought to the clinic in later lines of therapy in treatment, maintenance, and recurrent disease. Michael, tell us about those indications and options.

It’s a fascinating story for which I won’t got into great detail because it’s fairly complex. As you and this whole group know, the development of PARP inhibitors was almost abandoned. A lot of that happened at the time when Study 19 came across; it showed a wonderful PFS [progression-free survival], but it did not show an OS [overall survival]. At that point the FDA was still looking for OS differences, and that led to some discouragement among the companies. There was a little respite there where there wasn’t a lot of development.

To their credit, they got back on the ball, and the first approvals were in treatment. We need to remember that there are treatment approvals. The Bella Kaufman, [MD,] study, which was basically a basket study, showed a 34% response rate in patients who had recurrent disease, 

-positive, and that led to the first approval for olaparib, followed very closely by Study 10 and ARIEL2. Let me define, SAVOR2

 was 3 lines or greater, fourth line, if you will. ARIEL2 and Study 10 led to the approval for rucaparib in the treatment of 

-positive ovarian cancer recurrent tumors, 2 lines or greater. That wasn’t by accident. These companies are smart.

Finally, the third indication we have for treatment is based on QUADRA, which is a heavily pretreated patient population, 3 lines or greater. The average was 5 lines in that study. That’s an HRD [homologous recombination deficient] population for niraparib.

There were 3 indications for treatment. While that was being developed, we then had all of the maintenance trials, which was what opened the field up. The lead trial was SOLO-2; we’ve already talked about SOLO-1. It’s similar to SOLO-1 with a 

-positive patient population, platinum-sensitive. They had to respond, PR [partial response] or CR [complete response], randomized to olaparib versus placebo. It was a strongly positive trial. In fact, we just heard at the most recent meeting some of the long-term OS data from this study.

After that, NOVA and ARIEL3, right behind it, presented their data. Again, that broke the whole field open because NOVA, much like PRIMA, used all-comers and showed that niraparib in this randomized phase 3 trial, showed a significant PFS prolongation. ARIEL3 is similar to NOVA, showing that rucaparib is also active.

All 3 of those randomized phase 3 trials were published in high-impact journals and led to FDA approvals for the use of these PARP inhibitors in maintenance for recurrence after patients have responded to platinum. I have to emphasize that, because the platinum responses are used as a poor-man’s biomarker for PARP sensitivity.

 To your point, they all had biomarker indications.

 Olaparib, germline; rucaparib, germline/somatic; niraparib, HRD; and then in the maintenance setting for recurrent, all had platinum responses. That’s innovative. I’m going to go a little out of order here, Leslie, for continuity. At ASCO [the American Society for Clinical Oncology annual meeting], there was an overall survival advantage for SOLO-2, 12.9 months. If you have a 

 mutation, you get chemotherapy and then respond, and get olaparib or placebo, and you live 12.9 months longer. What does that mean to you?

It’s not going to change my decision to use the PARP inhibitor. The PFS benefit was enough for me. An OS end point is difficult in ovarian cancer because of all the post-progression therapy that patients can receive. OS is difficult to show.

If I remember that analysis correctly from ASCO, the OS benefit was not totally statistically significant. It’s almost not fair to say that because GOG-0213 pushed, and we’re talking about the upper bound of the confidence interval around the hazard ratio. GOG-0213 pushed that almost to 1 as well. As Matt eloquently described, we do consider that to have an OS benefit. To me, that OS is nice. OS will happen when we start sequencing and using multiple active therapies in a row. These trials aren’t there yet. These are single maintenance and one-off trials.

 These are insightful comments. Michael, what do you think about that? Technically, it wasn’t statistically significant, but they said, “We don’t really care, it’s good enough.”

A hazard ratio of 0.053 is close. Like many things, there’s good news and bad news. The bad news here is in the sense that I wasn’t surprised. I don’t know if anybody was that surprised. The PFS difference was huge. I don’t think it’s going to change practice, but I will say it’s reassuring to see it. I would never quibble about that small difference; it’s real, it’s good, it’s great.

PARP Inhibitors in Recurrent Ovarian Cancer

 entitled, “Practical Implications for New Treatment Strategies in Advanced Ovarian Cancer.” I’d like to focus on recurrent ovarian cancer, although there’s great hope for cure in newly diagnosed advanced ovarian cancer. We have new FDA approvals in the frontline setting with niraparib in all-comers as per April 29th approval of PRIMA, and as per the FDA approval of PAOLA-1 on May 8th with olaparib and bevacizumab in the HRD [homologous recombination deficient] cohorts. It’s difficult to say but still true that most of our patients develop recurrence.

The first agent approved in recurrent ovarian cancer was bevacizumab. This was November 14, 2014: AURELIA. This was a small step forward in platinum-resistant recurrent ovarian cancer in second and third line with a hazard ratio of 0.38 and only a 3-month improvement in progression-free survival [PFS].

Then in December of 2016, we got a platinum-sensitive option, which was a little better. It had 2 chemotherapy backbones: carboplatin-gemcitabine a la OCEANS and carboplatin-paclitaxel a la GOG-0213. OCEANS showed a bit better improvement in progression-free survival with a 4-month improvement; GOG-0213 had a 5-month improvement in survival. That was good, but it’s definitely not enough. Most of us prefer bevacizumab in the frontline setting. Leslie, I’m going to ask you, what percentage of patients with ovarian cancer do you think get bevacizumab at some point?

Hopefully most. It being frontline, platinum-sensitive, platinum-resistant, and being an active drug, I would hope that most of them, if not all of them, receive bevacizumab at some point.

In the platinum-resistant setting, Shannon, all else being equal, what’s your preferred chemotherapy backbone of those 3 AURELIA options?

I tend to lean toward the weekly paclitaxel. We saw in the laboratory that there’s something about the way those 2 drugs interact that we get even more bonus on antiangiogenic activity. The patients tolerate it well too.

The 2 approvals, Matt, in the United States, as you know are carboplatin-gemcitabine, carboplatin-paclitaxel in the sensitive space. There’s a lot of discussion about carboplatin-PLD [pegylated liposomal doxorubicin] as a third option. What’s your go-to regimen in the platinum-sensitive space if bevacizumab is a part of that?

We’re looking at patient factors to help drive that, the advantage of the PLD [pegylated liposomal doxorubicin] regimen is clearly the Q4 every 4-week dosing schedule, so sometimes that’s driving it. Looking at the quality of the level of the data from GOG-0213, you have an OS [overall survival] benefit with giving Taxol [paclitaxel] again, so carboplatin/Taxol [paclitaxel] plus bevacizumab. That’s my go-to regimen when I want that additional bar beyond PFS. That’s strong; it’s right at the physical cutoff, but it is positive with OS.

Shannon said it nicely that there is some special interaction with paclitaxel and bevacizumab, whether it’s cervical cancer, platinum-resistant or platinum-sensitive ovarian cancer.

You hate to make neuropathy worse, but there are a lot of tricks that we have to try to minimize that, alternate taxanes and things. We can get patients through this. They’re just not so excited about losing their hair again, so it can be a lengthy discussion.

Expert Approaches for Treating Recurrent Ovarian Cancer

 In the last few minutes in our conversation about frontline ovarian cancer, let’s go ahead and talk about combinations. I only want to spend about 30 seconds per study, and this will be brief. Matt, I want to take you to the first one, which has already been published. It is a combination with chemotherapy, but it’s not labeled. Tell us about VELIA.

VELIA was a very large study, 1000-plus patients studied, looking at the idea of using veliparib, which is a PARP inhibitor that you can combine with chemotherapy without too much toxicity. It’s lower dose, can be combined with carboplatin-paclitaxel, and then given as a maintenance therapy.

Cutting to the chase, the take-home message is that it didn’t seem to matter if you combined it with chemotherapy. It didn’t seem to help that much more than what we already have. Do we need a fifth PARP inhibitor? I’m not sure. I will say that it will attempt the FDA approval, but I’m not sure there’s a lot of interest beyond what we already have.

 Another combination obviously is PARP/IO immunotherapy. There’s a study called ATHENA, which adds rucaparib, a PARP inhibitor we haven’t talked about much. It adds rucaparib to nivolumab, with 400 patients in each arm. It takes the PRIMA indication of single-agent PARP inhibitor, in this case rucaparib, and adding an IO to it, niraparib, which will be interesting. The other part of that study though is indeed the PRIMA study, where placebo was compared to niraparib.

There will be 2 readouts from ATHENA; one PRIMA-like with rucaparib versus placebo, and that will come first. The other one, which will be beyond, or sort of ATHENA2, will take the single-agent PARP inhibitor with nivolumab, so I’m excited about that. Then the obvious question, Shannon, is triplet therapy. Maybe we should add bevacizumab, IO, and a PARP inhibitor. There are 3 studies looking at that.

Yes there are, and they’re pretty similar. The first study is taking an upfront population. They get a cycle of chemotherapy while they’re obtaining stratification factors, and then they’re randomized to 3 arms. There’s a chemotherapy alone arm, which can include bevacizumab. Then there’s the chemotherapy followed by a PARP inhibitor maintenance with niraparib, and again, bevacizumab can be included on that.

Finally, there’s a chemotherapy with dostarlimab, a checkpoint inhibitor, with or without, and then including the niraparib. There are 3 arms, so we’re going to get an idea of PARP inhibitor, PARP/IO, and then bevacizumab is included per investigator choice throughout those 3.

There are 2 trials looking at combinations with olaparib. DUO-O is looking at the same population’s response either pre- or post-surgery. They get randomized to chemotherapy with or without durvalumab, and then with or without olaparib as a maintenance. Similarly, ENGOT-OV46 is looking at a similar population with pembrolizumab instead of durvalumab. Importantly, DUO-O allows for bevacizumab, and then ENGOT-OV46 is simply the olaparib/pembrolizumab question.

 It’s interesting. Matt, to your point with VELIA, do we need a fifth PARP inhibitor because talazoparib is approved in breast cancer? You can see that there are 6 checkpoint inhibitors approved. Dostarlimab, which you talked about, will be number 7, and not only is dostarlimab integrated into this triplet opportunity, but so are durvalumab and pembrolizumab. It’s a competitive landscape. The real opportunity here, I can’t remember which of you said it, is in the HRP [homologous recombination proficient] population.

We have a 0.43 hazard ratio in PAOLA-1 in the HRD [homologous recombination deficient] subset with a doubling of the PFS [progression-free survival]. You don’t even get the doublet bevacizumab/olaparib in HRP. Even in PRIMA where HRP is labeled, it doesn’t work well. I’m excited about that opportunity.

I’m going to ask each of you for a closing comment: what you’re excited about moving forward as this rolls out, as we complete our conversation about frontline ovarian cancer. Leslie, why don’t you go ahead and tell us what you’re excited about as a last comment in our frontline discussion.

I’m excited to have options. I’m excited to have PARP inhibitors outside of 

 for sure. Trying to decide who is getting bevacizumab, I’m biased toward bevacizumab, I use a lot of it as well, so trying to decide who’s going to be best for that combination moving forward, is where I’m still working.

 Dr Birrer, as we finish this conversation about frontline, newly diagnosed ovarian cancer, what are your closing comments?

I share what Leslie said. It’s an exciting time for our patients. Having survived the IP [intraperitoneal] versus IV [intravenous] debate, we’re in a much better period, so it’s exciting. We need to settle this issue as to whether IO works, in particular if it works up front. By testing it up front, we can avoid that issue about when the tumors are all beat up and the immune system is beat up. We need to settle that, and I remain optimistic that it may not be a single agent like JAVELIN 100, but it may be a combination, antiangiogenic plus IO.

 As you know, JAVELIN 100 was just avelumab and chemotherapy. To your point, we didn’t review that negative study, which was presented at the SGO [Society of Gynecologic Oncology annual meeting] by Jonathan Ledermann BSc, MD, FRCP. It didn’t seem to benefit. If anything, it may have been a little hurtful, although you can’t make that conclusion.

Two things. One is that we learned a few things during COVID-19 [coronavirus disease 2019]. We’re much more aggressive using neoadjuvant chemotherapy. I don’t think we’re hurting our patients. The value of the knife is going down. We just reviewed our fellowship data, we’re doing much less radical surgery than we used to, and our patients are doing better, which is quite exciting. I’d like to be put out of a job as an oncologist, but we need to learn how to use these new tools the best. With all these targeted therapies, I hope they’re helping our patients.

 To your point, historically, only about half of gynecologic oncologists give systemic therapy. Who gives the treatments? The medical oncology community. I’m honored that the medical oncology community is engaged in these conversations. Quite frankly, I think they do a very nice job. Moving forward, we have to teach our fellows, as you just outlined, beyond surgery because the surgery isn’t changing. The systemic therapy is. Shannon, you have the last word.

In this group, it’s hard because I agree with everybody. I’m excited. We have many options that we’ve gotten over the last few years. It’s thrilling, but it’s like drinking through a firehose at times trying to make those decisions about which things we want to give what patient. I’m excited about the results that are going to read out over the next few years from these combination trials, and I remain hopeful that we’ll continue to tease out the HRD test to help it better select our patients.

 I want to thank my panel. I want to thank Dr Randall, Dr Birrer, Dr Westin, and Dr Powell. On behalf of 

, thank you so much for being with us, and so long for now.

Emerging Combinations in the Frontline Setting of Ovarian Cancer

 Shannon, you and others are integrated in measuring the patient experience. There was a nice SGO [Society of Gynecologic Oncology] paper…and it talked about what adverse effects patients care about. Let’s ask the patient. Let’s also ask the patient how they determine what the best decision is. Go ahead, tell us about that.

Yes, this was an interesting study. They dug in deep with about 34 patients. They did in-person interviews using a visual acuity scale to assess how patients make decisions around whether they want maintenance in ovarian cancer. Specifically, what were their perceptions of adverse events, and what were the adverse events they were willing to tolerate and for what benefit?

What they found was that a lot of the perception of adverse events was dependent on what they’d already had. This is a population in which the bulk of them were recurrent. They had a recurrence, so they had been treated in the past with drugs; some of them had had PARP inhibitors and some of them had had bevacizumab. That seemed to color their responses and their results. What they found is that they were most bothered by GI [gastrointestinal]-related adverse effects. Bowel perforation was far and away their perception of the worst adverse event, severe nausea and vomiting. That was then followed somewhat by fatigue, and then counts were definitely something that was less important to this group of patients.

They used the presence of these adverse events to make decisions, but the No. 1 thing that helped them make a decision was what their physician was recommending, and are they saying it’s going to help them? Are they saying it’s going to help them live longer?

They would weigh those adverse events with the idea, “My doctor says that I’m going to have a benefit from this.” They were even able to nail down, they were willing to accept about a 20% to 40% risk of an adverse event if there was going to be a positive, if they were going to get a benefit. These types of studies can be helpful when we have a number of different options in front of us to understand what’s important to the patient.

 As you know, I’m a big advocate of HPV [human papillomavirus] immunization, and we’ve looked at that. The No. 1 factor in a young individual, 9 to 12 years of age, getting Gardasil 9 is the strength of the recommendation to the parent. It’s the same thing here. If you say, “There’s this medication niraparib and PRIMA, I’m not sure. There’s olaparib, bevacizumab, PAOLA-1, I’m not sure,” then they’re not getting it.

Patient Preference for Treatment in Ovarian Cancer

 To finish up on PAOLA-1 as well, there are multiple things that we can do to help patients do better. We talked about treatment. The other thing that we can do to help patients do better is the quality of the surgery. There’s a paper that talked about the surgical subsets by [Christoph] Grimm, [MD]. Leslie, tell us about that paper.

It is another retrospective hypothesis-generating analysis. Grimm looked at the subsets of surgical outcome: neoadjuvant with interval debulking versus primary debulking surgery, with no gross residual versus residual disease. It’s interesting; this analysis almost looks like it’s all prognostic. The reason I say it almost looks all prognostic is that they didn’t show any benefit favoring any surgical group, except for the stage III patients who had upfront surgery with no gross residual disease, and it then showed potentially more benefit of the combination in that group.

As a surgeon, I was happy to see that because I’m starting to feel like there’s no reason to operate on my patients anymore. I was happy to see that.

In the other groups, there was no differential benefit. In the control arm, the median PFS [progression-free survival] for placebo group was all over the place, suggesting that these surgical groups are still mostly prognostic and not predictive of treatment effect. Interestingly, there was a similar study done, reported by Cara Matthews, [MD,] from SOLO-1, that showed that none of these subgroups mattered. The first time that we had ever seen those surgical factors not be prognostic was in SOLO-1.

 The take-home message, whether it’s SOLO-1 or PAOLA-1, is that no matter what the surgical outcome is, the targeted therapy helps.

 It helps the most in the primary debulking R0 [negative resection margin], right? Even if you have interval debulking, it still helps in the best prognosis, which is complete resection.

Another way to twist that is simply to say, as our drugs get better, they’re having a bigger impact across all patients. Maybe the surgery portion of this has less of an impact.

 Rob Coleman, [MD,] and I have a little talk that we’ve written and that we continue to evolve, and the title of that talk is, “Targeted Therapy Cuts Deeper Than the Knife.”

This challenged me because, as Shannon said earlier, our pivot point is usually, am I going to give bevacizumab even before, in cycle 1, I’m trying to decide if I am giving bevacizumab. Now that this cohort group looks like it’s a favorable population that was helped most from this dual-targeted therapy addition, I’m scratching my head, saying, “Is it everybody who needs this? Is it just those who had our previous need for bevacizumab only, is that my pivot point?” It adds more confusion the more we learn about how these trials inform us.

 Bevacizumab helps everyone. It helps reduce ascites. It probably helps prolong survival in stage IV patients. The hazard ratios in the smaller volume disease patients are not different or dissimilar than the large volume disease. It’s just that the survival and the ascites move on to a greater benefit, but it still has a PFS advantage.

PAOLA-1 Study Surgical Subset for Ovarian Cancer

 First Leslie, and then I’m going to come back to Matt because I saved all the hard questions for Matt. The NCCN [National Comprehensive Cancer Network] guidelines were updated on March 11th before the April 29 PRIMA and before the May 8th approval. Are they still relevant today? What do you think of the NCCN guidelines? They don’t align with FDA approval, but how could they have because they didn’t know what it was going to be? Doing HRD [homologous recombination deficiency] testing in the NCCN is a level 2B. I get it, if you look at the historical past for GOG-0218, bevacizumab was a level 2B, and then it got FDA approval, and they said, “OK, fine, it’s a 2A.”

It’s a little awkward for me because I’d hope that the NCCN would look at the data and be independent. Maybe they’ve evolved and have changed their perspective. What do you think of the NCCN guidelines today for this upfront management of newly diagnosed advanced ovarian cancer?

In general, since you were hinting at the philosophy, the NCCN guidelines are important. They help us build consensus on how we should treat our patients. They’re most important when there’s an information void, when there’s a data-free zone. Payers probably use that more than clinicians to justify reimbursement for a treatment or access for patients to a treatment.

Now that we have this data from the FDA, the FDA labels will probably trump the NCCN guidelines, especially from a payer’s perspective. I see them going back to revise the guidelines, and I don’t think I’ve ever seen them revise the guidelines in anticipation of an approval based on data that are presented in a meeting and paper.

 The challenge with PAOLA, thank you for presenting that by the way, is that once you start on bevacizumab, you’re going to continue it. What the NCCN doesn’t do is they don’t add guidance or guidelines on when to use bevacizumab. How does that make you feel that they don’t give information about when to use and when not to add bevacizumab?

It makes me glad that I get to choose based on my patients. They’re flexible.

 They’re supposed to be guidelines, not options. It’s not the NCCN options. It’s the NCCN guidelines. Go ahead, Matt.

I would have to say that they are payment options, as Leslie’s point is. The NCCN is a broad group. It’s not like, this is the best practice at this time when this committee meets. In fact, when they get a 2B recommendation, it just means not everybody agreed. Less than 50% agree that this should be a standard at this point. If we look around at just the members of this call, we don’t all agree on the value of HRD and what the best test is. When should we do the test? It doesn’t surprise me that it got a 2B recommendation back in January. There will be a revote, and they’ll say, “Yes, the FDA says we need this test to combine these 2 drugs.” It’ll be a 2-way recommendation.

 They’re the most valuable in the rarer tumors, right? Like the sarcomas and those sorts of things. For these sorts of things that I see every day, it’s about reimbursement. What do you think, Michael, on the NCCN? Then I’m going to move on.

I agree with what was said. If you go back historically to when NCCN was created—when Bob Young, [MD,] led that whole movement—it was to get a group of experts outside of the NCI [National Cancer Institute] and outside of the FDA that would provide them with reimbursement. That’s got a lot to do with it. You’re right, the companion diagnostic will work its way in. There’s no issue about that.

NCCN Guidelines and FDA Approvals for Treating Ovarian Cancer

 Michael, one of the interesting things about PRIMA is that they changed the dose, they personalized the dose. Tell us about that.

The original observation extended from NOVA, where there was not an insignificant amount of thrombocytopenia. We were involved in that trial, and there were patients whose platelet counts went down below 10,000. A retrospective analysis of that study showed that a lot of those patients could be identified because of low body weight, thin habitus, and low platelets, called weights and plates.

When PRIMA was designed, as you know, this was then inserted into what we would call individualized dosing in that trial. Like a lot of things, there’s good news and bad news. The good news is that approach worked. There was much lower incidence of thrombocytopenia. Although I might add that when it was presented at ESMO [the European Society for Medical Oncology annual meeting], it was a little confusing because they had low platelets and thrombocytopenia, and I couldn’t quite figure out what the overlap was. That’s been clarified since then, and thrombocytopenia dropped from 38% to about 20%, which is pretty darn good. Most of us are using it; I use it if I see a patient like that. It starts at 200 mg rather than 300mg; you always have the option of going up if you want. There’s been some interesting mumbling in the background that maybe 200mg should have been the dose to begin with, but this is a perfectly reasonable compromise.

 One of the interesting things is that the label for prescribing information of PRIMA is if the baseline weight is less than 77 kg, or if the baseline platelet count is less than 150,000, then you should start at 200mg, but it was not applied to the NOVA or QUADRA data. Although, to what Michael said, that’s what we’re going to do.

In the label, they show that there was no difference because you got to the right dose a cycle earlier. Patients sort themselves out based on toxicity in the first cycle, but that is impactful. At ASCO [the American Society of Clinical Oncology annual meeting], we’re not going to get into the specifics, but they even showed that, in the curves, there was no difference in the efficacy with the personalized dose.

I’ll add, Brad, that it doesn’t mitigate the issue of getting weekly platelet counts when you put a patient on niraparib.

 Thank you. We do that in my practice for all PARP inhibitors just to keep it simple. If the platelet count is less than 100,000, which would mean 99,999 or less, you should stop and reassess.

Personalized Dosing of Niraparib in Ovarian Cancer

 In the HRP [homologous recombination proficient] subset, we have to decide if it is PARP inhibition alone, as you just said Leslie, or is it bevacizumab alone? You can’t give both. How do you look at that? I would even put it out there: let’s not do the biomarker. I know there’s all this controversy. The biomarker is important; it’s a CDx [companion diagnostic].

Matt, how do you look at this HRP patient? I get it it’s the highest unmet need. I get it that that’s where research really needs to be done. What’s the best treatment for the HRP patient, and how do you make that decision?

Wow, why do you keep asking me all these hard questions?

That’s far from true…. That’s the patient who has no real targetable options. Maybe they have bevacizumab as an advantage. We’re going to spend a lot more effort thinking about what to do with these proficient patients, or the patients who had an HRD [homologous recombination deficient] scar who is no longer deficient, who has reactivation of their homologous recombination pathway. Those patients are out there as well.

What we did learn is that HRD is not perfect. We learned from the VELIA trial that the assays, wherever we draw these lines, are really tricky. The nice part about it is that, in the recurrent setting, we have patients who responded to platinum twice, and I can see that patient’s probably a PARP inhibitor candidate. If you barely responded to platinum the first time, it’s hard to know what to do with that patient; we don’t have great options for those people.

 One of the best things about being in education is what I call reverse mentorship. I said Leslie Randall was my fellow, and now I’m her trainee. What she just said here is very impactful. What she said, if I’m interpreting you correctly, Leslie, is that this may be the only chance for that HRP patient to get a PARP Inhibitor.

There’s a chance, just as you said Matt, that she’s going to have a platinum-sensitive relapse. The medium time to progression with HRP is 5.4 months in PRIMA. Half of them for sure are going to be platinum-resistant. The other half may have a platinum-sensitive relapse, but they then have to respond; remember, they’re HRP, so the chance of responding to platinum again is low.

Does that resonate with you Shannon, that it may be that the HRP patient should get a PARP inhibitor? In this case, it’s niraparib. Biomarkers are not perfect, and there are some patients who even have HRP who can do well for a long time on a PARP inhibitor. What do you think about that?

I’ve had this conversation now several times with patients in the clinic where they want to do something. We’ve done the testing on the tumor; the tumor is apparently HR-proficient, but that’s exactly what I say, “This is not a perfect test. There was clearly a proportion of patients who were identified to be HR-proficient who had benefit from the PARP inhibitor. I don’t know who they are, and you might be one, so if you’re willing, and you want to do something, then this is an extremely reasonable thing to do.” The elephant in the room here is that we’re all eagerly anticipating what the results of all these new combination trials that are ongoing and accruing will be in that patient population. We had hoped that PAOLA would be positive in the HR-proficient, but here we are.

 I want our audience to notice, they’re like athletes; they have game faces. I love it when Shannon starts to talk to me like her patient. She gets to wear her game face.

Treating Homologous Recombination Proficient Ovarian Cancer

 I want to talk about the HRP [homologous recombination proficient] subset in a minute. I don’t give up that easily though, Leslie; you know me.

 is more common in ovarian cancer. The risk of ovarian cancer is greater if you have a 

 is more difficult to treat with a PARP inhibitor. If you look at the 

 efficacy data in SOLO-1, there’s slippage.

 does better, but if you look in PRIMA, the 

. That supports Dr Birrer’s hypothesis that niraparib is a more potent PARP inhibitor because it works in the subset of 

. I want your opinion and what you’re going to do in your practice.

Let’s say there’s a contraindication to bevacizumab, and let’s say your patient has a 

mutation, and you have the option, as per PRIMA, to use niraparib, or as per SOLO-1 to use olaparib. Both are NCCN [National Comprehensive Cancer Network] recommended, level 1. What are you going to do? 

 Which one? Olaparib or niraparib in the patient with a 

 mutation? They’re both NCCN recommended. We’re going to get to that too; I don’t agree with you.

I’ve been using olaparib for almost 2 years.

 On April 29th, the world pivoted. Some people say that you can now use the better PARP inhibitor in that patient with a 

Yes. We’ve been having this discussion ever since platinum-sensitive recurrence, right? Which PARP inhibitor is better? We’ve had 3 indicated in the space, but still, what am I going to do? Should I just say my preference? I would use niraparib in that setting.

 OK. I’m not going to hold the rest of everyone accountable, Leslie was my follow-up.

Do you want to know what I do in my practice? I choose for very practical reasons. I choose for a patient who can’t remember to take a drug twice a day. I choose for a patient who I need a lot of dose flexibility on. I don’t have that with niraparib because most of my patients are going to go down to 200 mg, or start at 200 mg, and then go to 100 mg, and then I’m done. If I need more dose flexibility, I’m going to be using olaparib.

 To support you, as I always try to do, I would make the same decision, but because I like bevacizumab, I would want that patient to get bevacizumab-olaparib. I backed you into a corner.

I’m not in a corner because I haven’t decided that yet. I feel like the PARP inhibitor is enough, but if you look at the analysis presented at SGO [the Society of Gynecologic Oncology annual meeting], trying to pull in that olaparib-only arm that was not included in PAOLA; is olaparib alone enough? In that analysis, it showed that even the germline 

 Give us the details of that paper, which did a weighted comparison. They could do it, right?

It has a very specific name, the analysis….

This was a population adjusted, indirect comparison combining SOLO-1 patients with PAOLA patients, so you get a 4-arm trial if you look at it in a simple manner.

You get your carboplatin-Taxol [paclitaxel] by itself; you get your carboplatin-Taxol [paclitaxel] with a PARP inhibitor; you get your carboplatin-Taxol [paclitaxel] with bevacizumab; and then you get the combination. If you look across there, you see that 24-month PFS [progression-free survival] for the best group, which is the combination group, was 82%. If you look at the bottom, which is the carboplatin-Taxol [paclitaxel] alone, it was 36%. You can walk through this population and see the benefit and the contributions of each intervention. You’ll see that it looks like it’s at least additive giving these 2 targeted therapies to the backbone of carboplatin-paclitaxel. It compels you to think about the combination that bevacizumab is helping. So 82% progression free at 2 years, it’s staggering.

 Leslie, if olaparib makes bevacizumab better in the non-

patient, or at least the HRD [homologous recombination deficient], why wouldn’t it work in the 

Olaparib alone may be enough. To add bevacizumab, you may not get much more benefit, and you definitely get more toxicity.

 What do you think Shannon? For the patients with 

 mutations: olaparib, niraparib, or the combination?

For me, I have to make the decision right from the beginning. The population that I feel is most at risk has stage IV and ascites. For those patients, I am going to give the bevacizumab and then continue it with the olaparib. To Leslie’s point, it’s hard to know. I wish we had a better way of selecting who needs that additional benefit. I thought the analysis presented was compelling, but I want to see the paper and all the details before I make some bold change. I do think there is a proportion of the patients with 

who are going to do just as well with the PARP inhibitor alone. If we can tease that out and understand it, it would be helpful.

patients do better, right? Michael, what do you think is the best treatment for 

2? What do you think about the bevacizumab combination?

As you know, I give a lot of bevacizumab.

A lot of my patients are going to get a PARP inhibitor, olaparib, with bevacizumab. The interesting issue now that we’ve had multiple indications, and insurance companies are trying to wade through all this; it’s even conceivable to me that people are going to extrapolate from AVANOVA that niraparib may be combined with bevacizumab in some of these patients. I don’t necessarily think that’s wrong.

AVANOVA was in the recurrent setting combining niraparib with bevacizumab, showing it was safe and showing an advantage. Now that’s been moved up front with the trial that you mentioned, and I don’t think any of us will doubt that it’s going to be safe. It’s a single-arm, phase 2 study, so there’s only so much information that you’re going to get out of about 105 patients, somewhere in that range. But could I envision it showing good numbers, PFS, and safety, so leading to an expanded indication? It’s conceivable.

 It was just presented at these meetings. Because it’s a single-arm trial, it probably won’t lead to FDA approval, but it might add to a compendium listing and create an opportunity.

Correct. Let me say one other thing on this, Brad, because I know you’re thinking it, which is, all of these trials, combinations versus control arms, the 800 pound gorilla in the room that nobody looks at is sequencing.

 That’s what I said, the NCI [National Cancer Institute] needs to do that.

PAOLA-1 shows the combination is good. Can you get the same bang for your buck if you did it single-agent sequentially?

We don’t know, and that’s the question that NCI should ask, right?

 That’s Dr Randall’s point. Leslie’s point is can we get them to do it in sequence?

Here’s the problem with sequence that I’ve thought about. I’ve had to iterate these trials over in my head many, many times to come up with some of these thoughts. One thought is that, right now, your PARP inhibitor in the recurrent setting has restricted benefits to the platinum-sensitive population. Especially if you have an HRP patient who is likely to be platinum-resistant, front line may be your only opportunity to give them the PARP inhibitor. You may not have the option of sequencing PARP inhibitors until we get better combinations, which we’ll talk about later. Right now, single-agent PARP inhibitors are best given in the front line.

Treating BRCA+ Ovarian Cancer

 Michael, the point is that, when we had recurrent maintenance, all 3 indications were the same. We didn't have to have this conversation. We could talk about nuances, and niraparib is once daily. For olaparib today, there are 7 indications and 4 tumor types for olaparib. That’s cool, right? Prostate, pancreatic, breast, and ovarian cancers. It’s not that way anymore. Now the indications are different. Are they different because of the way the clinical trials were designed, or are they different because the agents are different?

It’s a great question. Getting back to the PAOLA-1 comparison, the Europeans, particularly Eric Pujade-Lauraine [MD, PhD,] made a big deal of the bevacizumab comparative arm neutralizing that effect in HRP [homologous recombination proficient]. I don’t buy that. It should at least be additive. In addition, if you look at GOG-0218, if you remember what happens when you stop the bevacizumab, the curves come back together.

Yes, and that’s inherent. It’s in PAOLA-1. It raises the issue that there’s something’s fundamentally different between the PARP inhibitors. They are probably the best data we have to suggest there’s something there. I’m the eternal skeptic, so I don’t buy it yet, but it’s worth some thought. Maybe we should design a direct comparison trial.

 To your point, that’s what the NCI [National Cancer Institute] is supposed to do, right?

 You know the pharmaceutical industry isn’t going to do that. That’s ideally what the cooperative group should do, and I completely agree with that. And the NCI should look at sequencing. Giving PAOLA-1, it would just be chemotherapy, but if you give PRIMA, it should be bevacizumab as a second-line therapy.

Another way to address it is, after progression on olaparib, do patients respond to niraparib?

 Right. Leslie, I’m going to follow this and then I’m going to move on. I don’t mean to belabor the point….

 I’m going to give you a point right now. Why don’t you say something, then I’ll ask you a question? Go ahead.

I want to chime in on this discussion; it’s important. Shannon is spot on. This is a lot of trial design: it’s population, and it’s control arm. It’s PRIMA, no bevacizumab. It’s PAOLA, different PARP inhibitor. In the HRP group in PAOLA, it looks like bevacizumab alone may be enough for them, and they may just not benefit from PARP inhibition.

If you look at PRIMA, they did benefit, but we’re talking about all these subpopulations. These are underpowered, hypothesis-generating assessments. In the normal situation, we would use this to design the next trial, right? The fact that the PAOLA label is based on that subset is surprising to me for that reason, but I guess it’s because it was so negative.

Need for Frontline Comparison Trials in Ovarian Cancer

 Leslie, about the biomarkers, there are a lot of conversations about this HRD [homologous recombination deficiency] biomarker. We should try to keep it simple. I realize that the biomarker is not perfect. At least tell us what the biomarker was both in PRIMA and PAOLA.

The biomarker is called HRD. I call it a phenotypic assay of DNA repair deficiency, homologous recombination deficiency. It has a score that has 3 components: loss of heterozygosity [LOH], telomeric allelic imbalance, and a large-scale state transition. They then give you a score of which somatic 

 is a part. The scoring is complicated, and if you’re interested, I would have you look at the package insert for the Myriad Genetics, Inc myRisk test, but the cutoff is generally set at 42.

 MyRisk is germline, myChoice is somatic.

Right, myChoice. I misspoke, thank you for correcting me. The cutoff for PRIMA and PAOLA is set at 42. You don’t have to set the cutoff at 42. That’s not been stringently validated; it’s just a cut point that we’ve chosen to use. There are things that are not measured in the myChoice test that contribute to HRD, and we don’t know the exact cutoff yet, so it’s not yet a perfect test. But if you look at the totality of all this frontline data, it’s pretty discriminatory with regard to the outcome. Although we say it’s not a perfect biomarker, it’s not an imperfect biomarker by any means. It can be very helpful in an upfront setting.

 Because it’s a companion diagnostic for PAOLA-1, as Matt said, it is reimbursed by CMS [Centers for Medicare & Medicaid Services] and others. There is a second commercially available option. You said it nicely: it’s a functional test. Looking at the mutation is not acceptable. The other functional test is Foundation Medicine, Inc’s LOH test. Is that as good, Michael Birrer? I know you’ve spent a lot of your career on this biomarker. Should we stick with the companion diagnostic because we don’t want to interfere with the reimbursement? How do you integrate the Myriad myChoice with the Foundation LOH score?

You just hit on it. It’s more of a reimbursement insurance issue. If you look at fundamentally how the assays were developed, they’re both based on LOH. Myriad just added those 2 other elements that Leslie talked about. I’m not sure how much they add, but I tend to agree; the assay works. It’s not perfect. Hopefully with more research, we’ll get it better and more accurate, but I remind all of my colleagues, even estrogen receptor expression doesn’t predict a tamoxifen response in breast cancer all the time. No biomarker is perfect.

 Matt, we’re all friends, but I’m going to ask you one of the toughest questions of the day, and I’m sorry. When I look at PRIMA versus PAOLA, 2 very impactful frontline trials, I see, as Leslie very nicely said, that there was an impact in the HRP [homologous recombination proficient] group, in other words the non-HRD group. I didn’t see that in PAOLA.

I want to explore that. I know we’re not going to get to the answer, but these are the questions that everyone is asking. What is the reason that there was no benefit in PAOLA-1 in the HRP subset? I’m going to give you some choices, or maybe there’s another answer. Is it because niraparib works in HRP and olaparib doesn’t? Or is it because the bevacizumab negated the effect of olaparib? Maybe olaparib performed as well as niraparib in HRP and bevacizumab negatively affected that? Or is it something else?

Without a head-to-head comparison, we won’t know. That is the tricky part here. The dose might matter in the proficient population, and we are probably getting a little better dose or penetration to the tumor with niraparib. Perhaps it’s that slight difference; maybe they’re the same as far as PARP inhibition, but we can get a better dose in the patients who are wild type or are HR proficient.

Both things are probably in effect. The bevacizumab is helping the placebo arm to do better in negating the effects of what we see with olaparib in PAOLA. I don’t know the answer. It’s nice that we have an option for PARP inhibition in our proficient patients. I’ll leave it at that and say, “Thank you, FDA, for giving us some choices.”

 Shannon, I want everyone’s opinion on this because this is what our audience wants to know, and it’s what I want to know. This is not meant to be a commercial; I’m not trying to sell anybody anything. I’m trying to help patients. I get that I’m a coauthor in many of these papers, and I don’t want that to influence my opinion. That’s why I’m going to let you talk.

Niraparib is the only PARP inhibitor in treatment. We’re going to talk about it and proof beyond 

Shannon N. Westin, MD, MPH, FACOG

Articles