Guidelines for germline and somatic BRCA testing in ovarian cancer

DR LOVE: To begin our discussion, I asked Dr Monk for his thoughts on a survey question we posed: In general, when do you typically order mutation testing for your patients with newly diagnosed ovarian cancer? And the overwhelming consensus was, “At initial diagnosis.”

DR MONK: Family history and age at onset are not predictors of germline BRCA mutations, and so all 3 organizations, NCCN, ASCO and the Society of Gynecologic Oncology, recommend germline testing for BRCA1 and BRCA2 genes in every patient with epithelial ovarian cancer — early-stage, late stage, endometrioid, serous. So yes, at a minimum — and we’ll talk about somatic testing in a minute, but somatic testing now is equally important.

DR LOVE: That’s what I was going to ask you, because I think this is really interesting, which is, you say, okay, what kind of testing generally? And you kind of see a split between some of the docs saying just germline and others saying both, and yet maybe in sequence or whatever, but what I thought is interesting is, it looks like the gynecologic oncologists are a little more oriented towards both than the medical oncologists. And I kind of wonder if part of that is some of the stuff that came out at the SGO meeting in terms of that you miss patients either way, right?

DR MONK: So first of all, this is a rapidly evolving paradigm, and the world changed on December 19^th. December 19^th is PARP day, did you know?

DR LOVE: No, I didn’t.

DR MONK: So we have gotten 3 FDA approvals on December 19^th.

DR LOVE: Really. Wow. I did not —

DR MONK: Three of the 6 FDA approvals —

DR LOVE: Wow.

DR MONK: — of PARP inhibitors occurred on December 19^th. Maybe it’s just because the FDA wants to go on Christmas holidays, for the break.

But anyways, on December 19^th, 2019, we got olaparib in maintenance. FDA approved based on SOLO1, the New England Journal publication by Katie Moore, okay? A GOG study. But it was also for somatic. So somatic now is absolutely mandatory. Now, there’re only 2 somatic patients in that study, and it’s interesting that the NCCN divided the recommendation for germline as a Level 1 and somatic as a 2A. It doesn’t really matter, because there’re only 2 patients.

It will be interesting to see what Europe does, whether they get a germline and somatic BRCA mutation, olaparib maintenance. But we have both in the indication, and the key point is, we have no companion diagnostic. So they got this somatic indication without telling us what the companion diagnostic is. We think it will be the same companion diagnostic that’s somatic for rucaparib in treatment, but that is yet to be seen.

So you can test for somatic using cell-free DNA even. There’s no established test in the label yet.

Incidence and potential impact of other clinically relevant genetic abnormalities on clinical decision-making in the front-line setting

DR LOVE: So speaking of that, what else theoretically would you think about or maybe consider a PARP inhibitor for in the up-front setting? And you can see a lot of people bring up the issue of PALB2, RAD51, but also HRD, which the group seems to be pretty split, like, only 13 say yes to that. So it seems like there’s a lot of variability in going beyond BRCA.

DR MONK: Right. So why do doctors do what they do? They do based on a label, because a label informs reimbursement, and today we don’t have an HRD indication. But these genes that you see are part of the homologous recombination repair deficiency, the inability of the cancer cell to repair double-stranded DNA breaks.

And you probably saw that last week on the Myriad website, they announced that they had filed HRD as a companion diagnostic for treatment now for niraparib. So niraparib is the only one of the 3 ovarian cancer PARP inhibitors that does not have a treatment indication, and now they have filed that based on QUADRA, also a Katie Moore publication in Lancet Oncology, and that will be HRD treatment, which is basically all of these genes, and later in the year we’ll have an HRD front-line study report, again, niraparib, which is called PRIMA.

So we have 2 things happening here, and expansion of the molecular testing, which we’re talking about, and also moving earlier lines of therapy. And SOLO1 was the first first-line positive trial, but there are other front-line trials, and some of these are expanded all comer or even HRD populations.

DR LOVE: So another assay question we asked about up-front therapy is MSI testing. And we are saying it may be different if it’s in the recurrent setting, maybe. But it looks like more than half of the people in our survey are doing MSI testing right up front. And I’m just kind of curious what your thoughts are about it.

DR MONK: We like that, but that’s much easier, because that’s generally an immunohistochemistry test. That can be done at the point of contact. Just to be clear though, most of the mismatch repair deficiency or MSI-high tumors are endometrioid or clear cell. The typical 85% of the patients, high-grade serous, are almost uniformly not mismatch repair deficient. So MSI testing is easy, should be done, but most of those are endometrioid and clear cell cancers.

Patient selection for neoadjuvant systemic therapy in clinical practice

DR LOVE: So let’s go on to the next area of questions, which is neoadjuvant therapy. We asked these docs, in general, what fraction of patients are getting neoadjuvant? And I was interested by their estimate, close to 50%. Do you think that’s the same? Is that really happening, for example, in the community?

DR MONK: I do. And it’s a little bit of a difficult decision, but I would venture to say that most patients with ascites, carcinomatosis, get neoadjuvant. So 50% makes it sound like 50% of those patients with tense ascites and carcinomatosis get surgery.

Now that’s not true. Almost all of those get neoadjuvant. The patients that get primary surgery are the patients who have imaging, just an omental cake or a suspicious mass, and if there’s any question, just stick the laparoscope in and look and assess the patient for surgical opportunity. In the beginning, prior to making that decision.

DR LOVE: So we also asked about the specific type of neoadjuvant therapy, and I guess the main thing I wanted to mention that I noticed, and maybe it’s obvious, but I’m just going to put it out there. There’s a couple things. First of all, I don’t see, even in a patient with BRCA germline, I don’t see people talking about a PARP inhibitor in the neoadjuvant setting. Do you agree?

DR MONK: You can’t give full-dose PARP and chemotherapy together. As you know, veliparib is being studied with chemotherapy, but half the dose. And that study will also report later this year.

DR LOVE: Although you theoretically could do chemotherapy followed by PARP, or you could do PARP alone, theoretically. In any event, people aren’t doing it in the neoadjuvant setting.

DR MONK: Right. We’re going to study that, though. We’re going to study chemotherapy free with these combinations, but we’re not there yet, correct.

DR LOVE: And as you mention, there’s a big GOG study, the VELIA study, looking at veliparib.

DR MONK: That’s what I’m saying.

DR LOVE: And that’s a huge Phase III study. I hear that may be at ESMO this year. Is that what you’ve heard?

DR MONK: Right, so we’ll have VELIA, which is veliparib with chemotherapy in maintenance, and PRIMA, just maintenance, but all comers in HRD. Both of them. And then the third study, which is imminent, is adding olaparib and bevacizumab together, which is PAOLA-1.

Bevacizumab as a component of neoadjuvant systemic therapy for patients with ovarian cancer

DR LOVE: Interestingly, I’m going to point out, though, that in terms of neoadjuvant therapy, of these investigators, there’re 3, and that’s a distinct minority, who use bev in the neoadjuvant setting. Any thoughts about that?

DR MONK: Yes, so again, doctors do what they do based on a label, and that’s off label. So the bevacizumab label, as of June 13^th, 2018, was after surgery.

So when you give bevacizumab before surgery, although it’s a pretty good idea because that’s a sick patient, has ascites and needs help, technically it’s off label. So that’s why there’s some uncertainty here. I think doctors really kind of prefer it if it’s reimbursed, but because the label specifically mentions after surgery, technically it’s off label. I like it, but it’s a reimbursement risk.

DR LOVE: The other thing is, when we present a scenario where the patient has, in addition to extensive abdominal disease, bilateral pleural effusions, still you don’t see much bev — the same 3 or 4 docs, but kind of at that point you should be getting a better response if you add bev in.

DR MONK: So Tewari presented the final overall survival for GOG 218, and in those Stage IV patients, there was a 10-month improvement in overall survival in the Stage IV subgroup, which was consistent with the high-risk subgroup in ICON7. So I get it that those are post-hoc analyses, but many of us think that that’s the best patient, these patients who are sick, have widely metastatic disease, Stage IV if you will, who can benefit from bevacizumab. And I think we think that that survival advantage may actually be real.

Said another way, Stage IV patients behave like Stage III patients. Stage IV patients behave like Stage III patients if you add bevacizumab. You can make that negative Stage IV prognostic factor go away with bevacizumab.

DR LOVE: I think that when you start talking about, for example, liver mets, people maybe are more inclined — patient presents with metastatic disease, you see more people using bev. But theoretically, if they’ve got pleural effusions, I don’t know that it’s really that much different.

DR MONK: Right. And my experience is, and I think the market research would support, that the medical oncologists endorse more front-line bevacizumab than the gynecologic oncologists. And I’m not sure why. Maybe it’s their experience in colorectal cancer. Maybe it’s their familiarity with the toxicity. But If you look at the use of bevacizumab at the time of the label, it was about 30%, okay? It’s about double that now. So having that label was impactful in expanding the opportunity, particularly for reimbursement, and so I think that that was important.

Clinical approach to patient selection for neoadjuvant chemotherapy; perspective on optimal schedule and dose of neoadjuvant regimen

DR LOVE: So we also asked these questions about neoadjuvant therapy in older patients. So we said a 77-year-old patient as opposed to a 60-year-old patient, and otherwise very similar. And it kind of looks like the answers are pretty similar.

At what age — they say age is just a number, but, I mean, is there an age at which you start curtailing your therapy assuming the patient’s otherwise healthy? I mean, 90? I don’t know if you ever see 90-year-old patients.

DR MONK: So there’re 2 reasons to give neoadjuvant chemotherapy. One is, you don’t think that you can adequately resect the tumor, and second, you think the patient is too sick to take it. And too sick to take it includes these emaciated, malnourished and elderly patients.

So there’re 2 different reasons. And obviously there’s no number, but we all see patients who you think, oh, that patient is not going to tolerate a big operation. Let’s give her some neoadjuvant.

DR LOVE: So I want to talk now, in a minute I want to go on and talk about adjuvant treatment. But I thought we could just pause for a second and have you talk about a patient, just any patient that comes to mind, or if you prefer, you can just talk about a typical clinical scenario.

But I’d like to hear you walk through the way you approach actually delivering neoadjuvant therapy. The regimen you use. When you go to surgery. And you could either use a case or maybe a scenario to walk us through that.

DR MONK: Certainly. The first thing that we’ve learned is this idea of dose intensification of the paclitaxel through weekly administration probably is not necessary. Unless the patient is so infirm that you want to get a little dose of carboplatin at an AUC of 2 and a little bit of paclitaxel 60, which is not dose dense. So that AUC of 2, paclitaxel at 60 weekly for the really sick patient’s a good idea.

That aside, I think we would all agree that it’s an AUC of carboplatin of 6 generally, although some would use 5, and it’s paclitaxel 175 mg per square meter over 3 hours every 3 weeks. So most of us in the typical patient would not give dose dense weekly, would not give fractionated, as I call it, unless she’s infirm, and our preference would be every 3-week carboplatin and paclitaxel AUC of 6 and 175. And if you thought you could get it reimbursed, probably bevacizumab on cycles 1 and 2. And then no bevacizumab on 3, surgery between the third and the fourth doses, no bevacizumab on 4, and then the triplet therapy 5 and 6, and then bevacizumab maintenance.

So you would give 4 of the 6 doses with bevacizumab, 1, 2, 5 and 6. And then the other controversy is, should you do HIPEC with the interval debulking between the 3 and 4 cycles? Probably not, but there’s some fascination with that opportunity. Otherwise, intraperitoneal chemotherapy, like dose-dense weekly chemotherapy, has fallen out of favor.

Importance of the extent of response to neoadjuvant chemotherapy in the timing of debulking surgery

DR LOVE: One question I’ve heard from a number of gynecologic oncologists is when to take the patient to surgery. And do you factor in kind of how the patient is responding? If, I don’t know, if they’re responding quickly, do you take them quicker? Or maybe later? How do you factor in response?

DR MONK: I try not to overthink it, because you can talk yourself into anything. It’s really between the third and the fourth cycle, and unless she really progresses, do it. And if she responds really well, great! That means that the surgery is really important. Let’s clean her out, have a complete resection, get her in remission, maybe even cure her, although that’s less likely. And if she didn’t respond that well, let’s use surgery to catch up, because we’re behind.

So I don’t really overthink it, unless — if she’s progressing in the first 3 cycles, then I don’t. And those patients that progress during carboplatin and paclitaxel probably don’t have ovarian cancer. I don’t know what they have, but they have a biology that is so unique that it’s a very poor prognostic factor. The typical high-grade serous, high-grade endometrioid tumor, the response rate to neoadjuvant chemotherapy is 90% or more.

DR LOVE: So you know the use of surgery in ovarian cancer and oncology in general is a little bit unusual, if not unique. And I guess one of the questions I have is, so it seems like you and most of the other people in this survey take this same approach in a patient with pleural effusions?

DR MONK: Uh huh.

DR LOVE: Right? Even though there’s no way you’re curing that patient, right? So it’s palliation. And that makes sense. You’re debulking the tumor —

DR MONK: You’re correct.

DR LOVE: It makes sense.

DR MONK: That’s right.

DR LOVE: It makes sense.

DR MONK: Uh huh.

DR LOVE: I mean look, when you give chemo, you’re debulking. When you do surgery, you’re debulking. It makes sense.

DR MONK: Correct.

DR LOVE: It’s not that it doesn’t make sense to me, but again, same approach if the patient has a couple liver mets?

DR MONK: Correct. Three cycles and do it.

DR LOVE: And you still do the surgery as a way to globally debulk them?

DR MONK: And I hear your words. I understand your rationale, I just don’t have experience not doing surgery on that patient. I want every patient to have a chance.

Therapeutic approach for patients with newly diagnosed ovarian cancer, germline BRCA mutations, bilateral pleural effusions and liver metastases

DR LOVE: Actually, I don’t even have any questions. The only question I have — let me tell you the question I do have, because I totally understand what you do. I understand why you do debulking surgery. I mean, this came up. We did a symposium at SGO, and we were doing videos of people in practice. We had a medical oncologist saying, I don’t understand why my gynecologic oncologists, they debulk patients even if they have metastatic disease. I understand that. Makes total sense.

Here’s what I do question. It seems like people don’t incorporate a PARP inhibitor even if the patient has a BRCA germline mutation. That’s what I don’t understand.

DR MONK: They should in maintenance, and if they don’t —

DR LOVE: Would you do that? So, for example, you have a patient with a known BRCA germline. She presents with metastatic disease. She’s got pleural effusions. She’s got a liver met. So you’re going to give her chemo.

DR MONK: Yes.

DR LOVE: You’re going to take her to surgery.

DR MONK: Yes.

DR LOVE: And then what are you going to do postop?

DR MONK: I’m going to give her 3 more cycles of chemotherapy and olaparib in maintenance as per the approval. And those were exactly the sorts of patients that were included in SOLO1.

DR LOVE: So you would do that also if they had pleural effusions as well?

DR MONK: Absolutely, and those patients were also in SOLO1. What happens, though, is, it takes a little bit of time to get the BRCA. So in many of those patients, they will have started on bev, and then oh, I got the BRCA back, or I got the somatic BRCA back, which is even tougher to get. Then what do you do? I still think you stop the bev after the chemotherapy, and rather than do bev maintenance, you use olaparib maintenance, because you could always go back to bev again.

DR LOVE: Okay.

DR MONK: Because the difference in SOLO1 is so huge. It’s unprecedented. It’s 3 years of progression-free survival.

DR LOVE: Although, I mean, it’s also almost the exact same hazard rate that you see in metastatic or recurrent disease with BRCA germline.

DR MONK: It’s exactly the same, 0.30. But 0.30 of 5 is 19. Zero point three of 14 is 50.

DR LOVE: Yes. No. I got it.

DR MONK: So although the hazard ratios are the same, the medians in the clinical benefit is enormous.

DR LOVE: Absolutely. I’m totally on board with that. No, it’s a revolutionary change.

Lingering controversies regarding the use of intraperitoneal chemotherapy for ovarian cancer

DR LOVE: So let’s talk about adjuvant therapy. And we presented a situation of a patient who’s 60 years old, has optimal debulking surgery, and we said BRCA wild type. And incidentally, in these cases, when we said BRCA wild type we meant nothing, no HRD, no other mutations, just nothing.

DR MONK: Yes.

DR LOVE: Okay? We can debate about the other stuff.

But anyhow, in terms of what they would do postop, I think to me, of great interest is a couple of things. First of all, 6 people would use bev, although you wouldn’t. At least that’s what you answered. Five would use IV IP therapy, which I thought was gone but apparently is not gone, because 5 people, including 1 medical oncology investigator. Any thoughts about these responses?

DR MONK: So I don’t know when I answered that question, but the patient that is not a good candidate for bevacizumab is the patient that has a high cure rate, and that’s not very many patients. So complete resections can have a high cure rate, but it’s pretty unusual, and so I prefer bev in that setting. Now that it’s FDA approved — I don’t know why people latch onto intraperitoneal chemotherapy, though. I think bev is easy. It’s FDA approved.

DR LOVE: I was shocked. Totally shocked.

DR MONK: I was shocked. In the intraperitoneal chemotherapy trial that’s in the New England Journal, GOG 172, Deb Armstrong, it was intraperitoneal, increase the dose, and it changed the regimen. So they changed the schedule, the dose and the root, and intraperitoneal got all the credit.

So then we repeated the study, as you know, which is in the JCO, with bevacizumab, is completely negative. People say, it’s the bevacizumab. Listen, if it was the bevacizumab, then use it! Because bevacizumab is so much easier than intraperitoneal chemotherapy.

So either way, intraperitoneal chemotherapy doesn’t work, or if you blame it on the bev, that’s fine. Then use bev.

GOG-0252: Design and results of a Phase III trial of intravenous or intraperitoneal chemotherapy, each in combination with bevacizumab, for advanced ovarian cancer

DR LOVE: So this trial. This is the one that has bev in all the arms?

DR MONK: Yes.

DR LOVE: Which one? I forget the number. What is that?

DR MONK: 252. 252. It’s been presented at the SGO, and it’s in the Journal of Clinical Oncology.

DR LOVE: So what exactly do they look at? And what did they find?

DR MONK: Yes, so they asked the pure question, carboplatin in the vein? Carboplatin in the belly? I get it. It was carbo. There was no difference. And then there was the third arm, just like the prior study, with cisplatin and paclitaxel in the belly, and there was no difference.

It was 1,500 patients. And everybody agreed this was going to be the definitive trial. And then when it didn’t support your bias, you didn’t believe it. That is not the way to practice oncology. When we do these trials and they don’t support your bias, you need to change your mind.

DR LOVE: And again, I thought it was — I remember when I first heard about that trial, it was very — because there was still a lot of controversy about bev. And yet bev is in all 3 of these arms. Were you involved in the design of it?

DR MONK: Absolutely. So the reason that bev was in all 3 of those arms is because we thought that bev was a game changer. Remember, bevacizumab GOG 218 was published in the New England Journal in 2011. So we’re like, “Oh, we did it!” A targeted therapy. And then the FDA didn’t approve it. And so that’s when there became this concern, and was it cost or toxicity, what we don’t understand —

DR LOVE: Although it was approved in Europe, though, amazingly.

DR MONK: It was approved in virtually every other country, including Australia and Mexico and Canada.

DR LOVE: Right. And the UK.

DR MONK: And the UK, right. So in the UK, NICE didn’t approve it, but they said, ovarian cancer is so rare we’ll just pay for it, the cancer drug fund.

Perspective on the recent FDA approval of bevacizumab in combination with carboplatin and paclitaxel followed by single-agent bevacizumab for Stage III or IV epithelial ovarian cancer after initial surgical resection

DR MONK: So then in 2018 it got approved. So people ask, “Monk, what happened? What changed between 2011 and 2018? Why did bev not get approved in ’11, but it did in EU and other countries, and now it is in the US?” And there are 3 reasons. And I want to — just humor me.

DR LOVE: Go for it.

DR MONK: The first reason is that we helped message to the FDA that progression-free survival was a valuable endpoint. So the idea was, it had to be overall survival, but because postprogression therapy was so long, patients got lots of therapy, you could never reach an overall survival benefit. So that was a key process, and we had a consensus with the FDA, and the FDA is a coauthor on that paper. Herzog is the first author. They said, okay, fine. PFS is fine.

And then, as I said, in 2018, 30% of docs were using it anyways. Off label. And the FDA hates that. The FDA hates oncologists out there doing stuff that they can’t regulate.

So (1) PFS is valuable, (2) doctors were doing it anyways and the FDA couldn’t regulate, and the third reason was that there was some CA-125s as progression in there, which made the progression-free survival only 4 months. So when we took the CA-125 progressions out, censored them, if you will, the progression-free survival went from 4 to 6 months. And there was more traction about the magnitude of the clinical benefit based on the resist PFS endpoint rather than the 4-month CA-125 endpoint that was published in the New England Journal in 2011.

So PFS is valuable, docs were using it anyways and we looked at the real PFS difference based on CT scans, not serologic markers, and all 3 of those things led to the FDA approval on June 13^th, 2018.

DR LOVE: So, though, just to clarify, this trial that you’re talking about, that was presented at the SGO that just happened?

DR MONK: So 252, the IP trial, no. It was presented at SGO2 years ago. Maybe even 3.

DR LOVE: This is the one with all 3 arms getting bev?

DR MONK: Correct. Three years ago probably.

DR LOVE: Huh. Okay.

DR MONK: And it was a hot topic then. Like you say, we kind of forgot about it. It’s nobody’s fault. But when it comes out in the Journal of Clinical Oncology, it will recirculate.

DR LOVE: Oh, yes. That’s what I heard you say, that it was just getting published now.

DR MONK: Yes.

DR LOVE: Okay.

DR MONK: That’s right. Uh huh.

DR LOVE: Okay, so just to clarify, though, because although people were really focused on the higher-risk patients in 218, if I understand the approval, it didn’t specify that the patients had to be high risk, did it?

DR MONK: Brilliant. No. It didn’t. And I don’t know why.

DR LOVE: Let me ask you, could you describe a patient who had debulking surgery who you wouldn’t recommend bev for?

Importance of disease stage and amount of residual disease in the decision to use bevacizumab as a component of primary therapy after debulking surgery

DR MONK: Yes, so in order to go on GOG 218, you had to have residual disease. So the complete resections were not included. And so those were the situations where bevacizumab is most controversial, because the supposition is that those patients might be cured without it.

I think we’re increasingly becoming aware that even if you have an R0 or a complete resection that you may not be cured, and maybe bevacizumab is appropriate for that patient. But if you have residual disease, you’re not being cured. Or if you have neoadjuvant disease, you’re not being cured. Or in your point, Stage IV. So those are ideal patients for bevacizumab. The complete resection R0 primary debulking patient, where a cure rate might be a little higher, there’s still some controversy.

DR LOVE: And that is the case that we were talking about before, and —

DR MONK: That’s exactly the case.

DR LOVE: Right, and there —

DR MONK: That’s right.

DR LOVE: And you can see the controversy right in the way people voted.

DR MONK: Uh huh.

DR LOVE: But then when we flip the case and say, instead of having no residual disease they have gross residual disease, now you have most people saying bev.

DR MONK: Great point. That’s right. Uh huh. Excellent. And that’s a key point for our listeners. That front-line bevacizumab is approved, and basically in the gross residual or the Stage IV it’s a no-brainer. It’s a no-brainer.

Evaluating the optimal duration of maintenance bevacizumab

DR LOVE: What about duration of maintenance bev?

DR MONK: Right. So the duration was 15 months or 21 doses in GOG 218, because we thought that would be the median time to progression. And so I told you that there were 3 trials that were going to be presented this year, the VELIA, the veliparib, the PRIMA, the niraparib, and then PAOLA-1, PARP and bev. The fourth one is called the BOOST trial, which will randomize patients to 15 months of maintenance bev versus 30 months — 15 versus 30.

So the current indication is 15 months because that’s what was in the study, but we all want to do it longer. And that trial, 15 versus 30, will also be presented in the next year. And we wonder if prolonged maintenance bevacizumab in those front-line patients will have even greater benefit. And that’s called BOOST.

DR LOVE: Interesting. Any predictions on what that’s going to show?

DR MONK: I’m not good at that, no. I don’t know.

DR LOVE: Yes. I’m trying to remember if there’ve been any bev duration trials before.

DR MONK: So the study that John Chan did in the New England Journal called 262, which was the dose-dense weekly versus every 3-week, that did use bevacizumab until progression. And when you do it to progression, it’s about 15 months on average anyways. So I don’t know how many patients will remain progression free to get to 30 months. So that’s why there’s some hesitation about whether or not this will be a positive trial or not.

Effect of bevacizumab on progression-free survival for patients with high-risk disease

DR LOVE: When you have a patient, let’s say, who has gross residual disease, and you think about kind of what the impact of adding bev is, let’s say from the point of view of a hazard rate, what would you say in general it would be for progression-free survival? Not just from the trials, but from your experience, all the data you’ve sifted through globally. What do you think the impact of bev is both in progression-free and overall survival?

DR MONK: Yes, so the progression-free survival hazard ratio is 0.62. The final overall survival, we just got the review back from the JCO. It looks like it’s going to be published. We’ve got to make the manuscript better. And again, the overall survival really is in the Stage IV patients like we talked about, 10-month improvement in overall survival makes those Stage IV patients like Stage III. I mean, 0.62 is a pretty significant boost for a therapy PFS.

DR LOVE: Yes, but, I mean, 0.62 as a hazard reduction for a therapy that —

DR MONK: That’s pretty good.

DR LOVE: — not the most toxic therapy. I mean, it has to be infused, but that’s a fairly sub —

DR MONK: That’s right.

DR LOVE: — fairly substantial hit that you’re getting.

DR MONK: And we’ve been giving bevacizumab a long time in ovarian cancer. And everyone that has given it has seen it change women’s lives. And I’ve worked hard, and I’m sorry, but there’s no biomarker. But there are some patients with ovarian cancer where bevacizumab is transformational. And I’m sorry that I can’t figure it out other than lots of cancer. Lots of cancer needs lots of bev.

Risk of bevacizumab-associated gastrointestinal perforation; contraindications to bevacizumab

DR MONK: But that’s kind of why you get into this is, you’ll see — I saw a patient on Monday from Beverly Hills — I live in Phoenix — who came with a bowel obstruction a year ago, and she was bevacizumab naïve. I said, I’ve got nothing for you. Your risk of GI perforation is much higher, but if you want to go for it, get some chemotherapy and bevacizumab.

She came back on Monday, and she’s in remission, and she’s still on bevacizumab maintenance, and she’s seen her grandson graduate from high school. So again, she could have had a GI perforation and died. But she was going to die anyways, of a bowel obstruction from a high-grade serous cancer.

And so that’s an example. I’m sorry there’s no biomarker, but it changed her life. And her medical oncologist in California and her and her whole family are grateful for that opportunity. It changed her life.

DR LOVE: That brings up another issue that we tried to assess in this survey, which was how the presence of bowel obstruction or other intra-abdominal complications affect the use of bev. And it’s amazing you brought up this case, because I’m going to guess a lot of people would not have used bev in that situation.

DR MONK: That’s right. We were desperate.

DR LOVE: How do you evaluate the risk for intra-abdominal problems?

DR MONK: Right, so it’s GI perforation. GI perforations we mean diverticular abscess, ruptured appendix, duodenal ulcer. The biggest risk factor is inflammatory bowel disease. It should almost be a contraindication. The next is bowel obstruction. And the third is number of lines of therapy. The more number of lines of therapy, just — there’s too many things that can go wrong.

But if you use it early on, such as front line, in the absence of inflammatory bowel disease, where there’s no bowel obstruction obviously, the risk of GI perforation is less than 1%. So you can kind of look at a patient and say, in the Cannistra recurrent Phase II it was 11% GI perforation, 5 out of 44. Eleven percent GI perforation is a big deal. And again, these are very sick patients. They get intraperitoneal stool and peritonitis.

Perspective on combining a PARP inhibitor with bevacizumab as maintenance therapy for patients with advanced platinum-sensitive ovarian cancer

DR LOVE: So let me bring up another factor from this survey that’s, I think, interesting, which is — and I’ve seen this in other forums, when we’ve polled people, which is the high-risk patient who has a BRCA germline mutation, of course they’re going to get olaparib. But what we’ve seen in the community — and, actually, it’s interesting. I don’t see that much of it here in the investigators — are some people integrating bev.

So we see some people, and there were several people, investigators, who give bev during the chemo and then give the olaparib. And there are some people, and you talked about the PAOLA trial that’s going to look at this and report soon, but there are people out there right now who are giving bev during chemo and then bev and PARP together. Marked minority. Very few of these investigators do it. Any thoughts about that?

DR MONK: We kind of want to do it, don’t we? Because they’re not overlapping toxicities. And you’re right: We don’t have any data. But we kind of want to do it together. The good news is that the trial has been done, completed enrollment. The rumor on the street is that the events are being met. And so we’ll see if combination therapy versus sequence.

But listen, in ovarian cancer it’s all about combinations, right? We don’t give carboplatin first and then paclitaxel. We don’t give carboplatin, paclitaxel and then bev. We give them together, and then once you’ve had as much carboplatin and paclitaxel as you can take, we do maintenance bev. So I think there’s an opportunity here, and we’re very excited.

DR LOVE: So it’s interesting, we asked these investigators, have you or would you use a PARP inhibitor combined with bev as maintenance therapy in a patient with advanced ovarian? And, actually, three quarters of the people, including you and Dr Ledermann, said, “I have not and would not.” But about a quarter of people said, “I have not but would for the right patient.” So I’m not sure exactly what that — maybe a younger patient who’s kind of grabbing you, saying, “Can you do everything you’ve got?”

DR MONK: So when I answered that question 3 weeks ago, it’s been that long, I was counseling a patient, bev versus PARP in platinum-sensitive relapse. And she was a dermatologist, and she’s like, “Now, let me get this straight, Monk. Why don’t I just get both of them?” And she says, “Has that ever been done? Is that tolerable?” And I said, “Yes, it’s tolerable,” I said. And the labels don’t technically say, “Don’t give bev without PARP” and “Don’t give PARP without bev.”

DR LOVE: That’s right.

DR MONK: And I said, “Let’s start the bev with the chemotherapy, then we’ll do bev in maintenance, again, platinum-sensitive relapse. And then once you go into remission, we’ll submit funding for the PARP and we’ll see what happens.” So I don’t know what’s going to happen here in the funding situation, but I’m going to try. Because, like I said, you kind of want to do it.

DR LOVE: Yes, I’ve heard that story from other people. Even docs in practice say the same story. I’ve got this patient. She’s well read. She’s et cetera. It’s an obvious question. Hopefully it’s going to get easier once that trial reports.

DR MONK: Yes.

DR LOVE: It’s interesting. We asked these questions, would you, or if you haven’t, would you do it? And it’s interesting, because like you said, you said, “I have not and would not.” But then you get a situation that you don’t anticipate.

PAOLA-1: An ongoing Phase III trial of olaparib combined with bevacizumab as maintenance therapy for patients with advanced ovarian cancer after first-line platinum-based chemotherapy in combination with bevacizumab

DR MONK: So here’s the unique thing about that trial, though, that adds bev and PARP. The primary endpoint is all comers. So it’s the first randomized PARP study that doesn’t have a biomarker endpoint.

DR LOVE: Hmm.

DR MONK: So many people would have had HRD as the endpoint, or even BRCA, but it’s all comers. So it’s possible that the study will be positive, but it will be driven by the BRCA patients or the HRD patients. And then we’re going to have to dissect what the real benefit is of doublet therapy, that is, adding a PARP inhibitor to the bevacizumab in the completely molecular signature-negative patient.

So that will be a fun conversation to have. It’s true, it might be that bevacizumab and olaparib works well even if you don’t have a molecular mutation, but I don’t think so. I think that we know enough about PARP inhibitors that they work the best in patients who have BRCA, and then the next best in the patients that have HRD, and they work just a little bit in the non-HRD, non-BRCA subset. So it’ll be interesting, but in that PAOLA-1 combination, the primary endpoint is an intent-to-treat analysis.

DR LOVE: Hmm.

DR MONK: So that will be fun to talk about.

DR LOVE: Wow. That really is going to be interesting.

Practical approaches to mitigating carboplatin/paclitaxel-induced peripheral neuropathy

DR LOVE: I want to ask you also, we asked a couple questions about — it’s kind of a common situation, but just — we were curious what people do and whether or not they all do the same thing, relating to management of patients with peripheral neuropathy.

So obviously an old-time question. We’re just kind of curious what kind of homogeneity or heterogeneity we had. So we presented a situation of a younger patient, 56 years old, who is on every 3-week carbo/paclitaxel after debulking surgery and after 4 cycles develops Grade 1 peripheral sensory neuropathy.

So, actually, there was a split in that actually most people would continue treatment and not make any modifications with just Grade 1. But about 40% would reduce the dose of paclitaxel for Grade 1. Can you talk about what you do and what you think about this approach?

DR MONK: So it’s a tough situation. Probably for Grade 1 I wouldn’t do much either. For Grade 2 probably dose reduce. Certainly there’s an opportunity for docetaxel. And I don’t know if you saw on the NCCN, but the NCCN now moved up carbo/PLD as a 2A opportunity for those particular patients who have neuropathy.

So going from 175 to 135 in a dose reduction, particularly when you only have a couple of cycles left, probably doesn’t do much. But some patients get neuropathy early on, and you’re going to have to make a sustainable change, particularly in Grade 2. And that’s where docetaxel or PLD is an option.

In my practice I’ve been doing docetaxel for years, but I’m starting to think more and more about pegylated liposomal doxorubicin as an opportunity for replacement. Yup.

DR LOVE: When you start out a patient on carbo/paclitaxel who’s a diabetic but does not have peripheral neuropathy, do you approach that patient any differently? Are you quicker to dose reduce?

DR MONK: So the other problem with the diabetic patient is, the dexamethasone would cause a lot of hyperglycemia.

DR LOVE: Right.

DR MONK: I just have a higher sense of awareness. I do not approach them differently other than the sense of awareness.

DR LOVE: And what do you do with a diabetic or a patient who has peripheral neuropathy from some other cause to start with?

DR MONK: You just have to assess the severity, and if the severity is worth modifying the treatment, then you can either start docetaxel or, as I said, you could use pegylated liposomal doxorubicin.

Comparison of treatment-associated neurotoxicity with nab paclitaxel and solvent-based paclitaxel

DR MONK: We, within our specialty, have not used a lot of front-line nab paclitaxel. I’m not exactly sure why, but in many settings that — breast cancer, that’s what they do. You saw the atezolizumab/nab paclitaxel approval in breast cancer, and that was related to the steroids with the paclitaxel, because they didn’t want to give steroids with the checkpoint inhibitor. So I don’t know. But we generally don’t approach them differently, but we have a solid sense of level of awareness and suspicion.

DR LOVE: Nab paclitaxel’s been kind of interesting for me to observe for a long time out there. It’s been out there for breast cancer, lung cancer, and I still —

DR MONK: Pancreatic. Yup.

DR LOVE: — yes, pancreatic for sure. There are people who believe that it is less neurotoxic, but a lot of people don’t. And when I ask people for data, it’s kind of hard to say, so I’m not really clear whether it’s less neurotoxic or not.

DR MONK: Let me just tell you. So weekly paclitaxel/bevacizumab is a very special regimen in platinum-resistant recurrent ovarian cancer. And in that setting, many of these patients are heavily pretreated and have some existing neuropathy. So our go-to regimen here in Phoenix in that platinum-resistant recurrent ovarian cancer patient is nab paclitaxel/bevacizumab.

Nab paclitaxel/bevacizumab is a very special regimen. Eighty mg of nab paclitaxel weekly without a break, bevacizumab every 3 weeks, and away you go.

Clinical experience with bevacizumab for ovarian cancer; management of hypertension and proteinuria

DR LOVE: So one other thing I was going to ask you about is, we’ve been talking a lot about bev, and you were commenting on the issue of intra-abdominal issues. In terms of other issues that are more common, particularly hypertension and proteinuria, any experiences with patients where that was an issue or a problem?

DR MONK: This bevacizumab idea was kind of — emanated from Bob Burger and me in 2000. So we’ve been working on this for 19 years. And having grown up in OB/GYN, the first is, we have every patient on an anti-VEGF therapy, whether it’s a TKI or bevacizumab or even aflibercept, to check their blood pressures at home, so that’s very important. And we also try to have the oncologist manage the blood pressure, because if you don’t jump on it immediately, it can lead into headaches and all these other problems. So that’s the first thing.

The second thing is, with proteinuria we used to do these urine ratios and all that. Now we just dip the urine for proteinuria, and in 1+ we don’t worry about, but if it gets to be 2+ we treat the patient on that day, and then we make sure she doesn’t have a UTI or it’s not a stent or something. And then we do a 24-hour urine, and if the proteinuria is not greater than 2 grams in a 24-hour period, we continue. And frequently my experience is, that’s been reversible. And so those are notable side effects or adverse events attributable to bevacizumab that can be dealt with, with careful monitoring.

Activity and ongoing investigation of anti-angiogenic agents alone or in combination with PARP inhibitors for advanced ovarian cancer

DR LOVE: You were talking about other anti-angiogenics and your longstanding interest in this whole area, and every now and then I start to see papers on cediranib. What about TKIs? It seems like they’re effective, but somehow they don’t kind of get there.

DR MONK: So for pazopanib, for example, we did a maintenance trial with pazopanib just like SOLO1, but it was all comers with pazopanib —

DR LOVE: This was for ovarian?

DR MONK: Uh huh.

DR LOVE: Ovarian.

DR MONK: Published in the JCO. Uh huh. And there was also a 6-month improvement in progression-free survival.

DR LOVE: Wow.

DR MONK: And we submitted it to the EMA, and the EMA had all of these concerns about hypertension, and it spun out of control and it never got labeled.

DR LOVE: Wow.

DR MONK: It’s still listed in the NCCN guidelines.

DR LOVE: Huh.

DR MONK: So there’s this anti-VEGF oral cediranib versus olaparib combination that’s being studied in the NCI and other places. So we talked about bevacizumab/olaparib, but there’s an opportunity for cediranib/olaparib. It’s the same sort of idea, anti-VEGF/PARP. The beauty of cediranib and olaparib would be that it would be oral. The disadvantage, as you know, is that olaparib has some other adverse events, particularly diarrhea. So we’ll have to see how these things play out, both from an efficacy standpoint but also a toxicity standpoint.

DR LOVE: You also mentioned aflibercept. Has that been looked at in ovarian?

DR MONK: It has. Michael Birrer had a study, and it looked at 2 different doses and it worked, and it was just never pursued.

FDA breakthrough therapy designation for lenvatinib in combination with pembrolizumab for previously treated advanced microsatellite instability (MSI)-nonhigh/mismatch repair proficient endometrial cancer

DR MONK: Let me tell you, our TKI experience today in gyn oncology is with lenvatinib, because we —

DR LOVE: Wow.

DR MONK: — have breakthrough designation. They’re studying lenvatinib with pembrolizumab in endometrial cancer. So we have pembrolizumab alone in MSI high, but when you add lenvatinib to pembrolizumab, it might actually work in endometrial cancer that’s not MSI high. And so that has breakthrough designation, and that will be filed to the FDA and hopefully will be an expansion in second-line endometrial cancer, pembrolizumab/lenvatinib. And as you know, that combination is being studied in multiple other tumor types. But that’s the TKI that we have in the clinic today in the endometrial cancer space.

DR LOVE: That’s interesting you mention lenvatinib. We’ve been talking about that a lot with HCC — hepatocellular cancer —

DR MONK: I know you have.

DR LOVE: — also, but what about lenvatinib alone in ovarian?

DR MONK: It’s been studied. It has some activity. But as you know, there’re a number of oral anti-VEGF TKIs. None of them really have traction, but this VEGF/PD-1 combination is special, whether it be axitinib and avelumab in renal or even bevacizumab/atezolizumab in lung. So we’re excited about these combinations, and that’s what we’re working on daily, in ovarian cancer as well as endometrial and cervix.

DR LOVE: What about ramucirumab? Has that been looked at in ovarian?

DR MONK: Not in a systematic way.

DR LOVE: Interesting.

Rationale for the design of ongoing studies combining immune checkpoint inhibitors with anti-angiogenic agents and/or PARP inhibitors

DR LOVE: You were talking about the idea of combining anti-angiogenics, for example bev, and TKIs, with checkpoint inhibitors. Can you talk a little bit about where that is right now? And you mentioned the accelerator or the FDA thing there with lenvatinib/pembrolizumab. But generally where that strategy is in gynecologic oncology.

DR MONK: Yes, so we have 3 active categories. You’re talking about them. Anti-angiogenics, PARP and IO. And we’ve had an epiphany in the IO space that agents such as KEYNOTE-100, which is pembrolizumab, wasn’t very active. And you could get a little bit of enhancement through a PD-L1-high expression, but it really wasn’t very good.

And then we said, that’s okay, we’ll add chemotherapy to it, and so avelumab added to pegylated liposomal doxorubicin just at the SGO we just went to, negative. And then we added it to front-line carboplatin/paclitaxel, like you do in lung, stopped by the data safety monitoring committee.

So the microenvironment in ovarian cancer is very immunosuppressive. So now we think, okay, IO/chemo isn’t that great, let’s do IO/bev, so atezolizumab/bevacizumab, that study is enrolled front line. It’s done. We’ll have to wait. Or IO/PARP, for example, with rucaparib and nivolumab. And certainly pembro and olaparib, durvalumab/olaparib.

So now there’s this great hope that IO alone isn’t enough, and maybe we need to do IO/VEGF or IO/PARP or even all three, because some of the studies just say, if you can get bev paid for, we’ll layer bev over these IO/PARP studies, which is really a triplet approach, and that would, for example, be the study called FIRST. So it’s very complicated right now.

I can tell you, the thing I know for sure is that the microenvironment in ovarian cancer is very immunosuppressive, and it is not breast cancer. And certainly it’s not lung cancer and all these other IO-approved tumor types. There’s something about that microenvironment that is very different, and PD-L1 expression is a little bit of a signal but probably is not enough to be a biomarker in developing IO alone.

DR LOVE: So if I hear you, it sounds like what you’re saying is, you believe that even though it hasn’t been that helpful yet that theoretically immunotherapy could be very effective in ovarian cancer.

DR MONK: With PARP or angiogenic, yes.

DR LOVE: That the biology, you think, is — we just haven’t found the right intervention yet.

DR MONK: Right. We do these large, expensive trials at a Phase II risk. So what happens, we get some safety data, and it’s such a competitive landscape that we launch into a very expensive thousand-patient trial. The idea of doing a randomized Phase II for biomarker discovery and signal finding, we don’t do that anymore, because there are 6 FDA-approved IO agents, 6 FDA approved PD-1 or PD-L1, and they’re all trying to compete.

Modulation of gut microbiome to enhance response to anti-PD-1 immunotherapy; impact of antibiotics on response to treatment with immune checkpoint inhibitors

DR LOVE: So I’m trying to remember if over the years I’ve ever asked you this question, because you’re really a Talmudic scholar of oncology research, but I’m curious what your thoughts are about the issue of the microbiome and the immune system and in general whether you see that as a potential factor that may be useful in gynecologic cancers.

DR MONK: So we have a grant here at the University of Arizona to study that. It’s so, so interesting, whether it be the vaginal microbiome and how the HPV virus causes cervical cancer or the gut microbiome in the GI toxicity that you have with your chemotherapy or the gut microbiome and how that sensitizes or desensitizes you to IO. So I’m really intrigued with this. I think it’s in the infancy stages, but I love that sort of thing.

And personally, I believe in probiotics. I took them just this morning. What do I know? But I’m a fan of probiotics.

DR LOVE: Everybody’s fascinated by it, but I have to say that I was shocked. We did a symposium at AACR. I don’t know if you know Naiyer Rizvi from Columbia, lung cancer, he’s a checkpoint —

DR MONK: Very famous.

DR LOVE: — checkpoint guy, and we were showing this study where it looked like people on IO did better if they didn’t have an antibiotic.

DR MONK: That’s exactly right.

DR LOVE: And I said, does that translate to your practice? He goes, yes, I kind of think twice before I use an antibiotic. I never heard anybody say that before. But I guess once —

DR MONK: Yes, that got a lot of press, particularly for the upper respiratory tract infections that are viral, where you don’t need an antibiotic anyways.

DR LOVE: Right.

DR MONK: Exactly.

DR LOVE: Right.

DR MONK: Very intriguing.

DR LOVE: Anything else you want to add to what we’ve talked about in terms of this issue of up-front management of ovarian cancer? Anything that we haven’t talked about you want to add in?

DR MONK: You’re so impactful, and you have such a large footprint, I just want to remind our medical oncologists that before chemotherapy is started, the patient should be evaluated by a surgeon to participate in the decision to do neoadjuvant chemotherapy or not. All too common, the medical oncologist will have started the chemotherapy before the gynecologic oncology consultation. ASCO has actually chimed in on that and has an opinion that every patient, before the chemotherapy is started, should be seen by a surgeon.

Biology and histopathologic subtypes of ovarian cancer

PROF LEDERMANN: I think we’ve always struggled with the concept of primary peritoneal cancer looking exactly like ovarian cancer, and we’ve made up some sort of hypothesis that it starts in the peritoneum, and the tissues are very similar to the ovary — never quite hung together. And so we see the same histology in both primary peritoneal, ovarian and fallopian tube cancer.

So the hypothesis that was then developed of this Type II ovarian cancer to represent the most common type, the high-grade serous ovarian cancer, was that it actually originated from the fimbrial end of the fallopian tube. And then the initial STIC lesions that one can see, these altered p53 mutations in the fimbrial end of the fallopian tube then actually either fall onto the ovary and then advance to a cancer or remain in situ and become fallopian tube cancer or go to the peritoneum where they become primary peritoneal cancer.

So this is a unifying hypothesis for the high-grade tumors. And there’s good evidence now building up from the pathologist’s reviews to support that. And that’s very different from the ovarian cancer that we see, the less common Type I, which originates in the surface epithelial of the ovary, characterized by certainly the low-grade serous cancers but also the mucinous and some of the clear cell cancers too.

So I think conceptually, it helps us to understand the development of the 2 tumors, the different development, and actually in terms of high-grade serous it gives us clues as to how we should be looking for the lesions earlier on, and we shouldn’t be looking in the ovary, because that’s too late, in a sense.

So I think that’s helpful. I think at the moment, of course, we’re tending to treat all ovarian cancers, irrespective of the histology, in the same way. And that’s a problem. And we need to find specific treatments for specific disease entities.

Structure of gynecologic oncology clinical practice in the United Kingdom

DR LOVE: Again, I want to kind of shift into the discussion of evaluation of the newly diagnosed patient. Can you talk a little bit about your own clinical setup there in terms of, do you see a lot of patients who kind of come in de novo, or most of them have been seen by some other physician? And how is their visit with you set up in terms of who they see and how decisions are made when a patient presents?

PROF LEDERMANN: Okay, so about 15 years ago now in the UK — because 90% of all the practice is through the National Health Service system — about 15 years ago the treatment of gynecological cancers became centralized. And designated surgical centers were set up. So it’s no longer possible for gynecologists to operate on ovarian cancer in the community or in the smaller hospitals. So if they’re suspicious of a patient having ovarian cancer, they refer her into the center, and we drain a population of around about 2.2 million. So every single case of ovarian cancer will come in to us.

And they will be reviewed, together with surgeons and the oncologists and the pathologists and radiologists in our multidisciplinary meeting, which you would call a tumor board, and a decision would be made at that point whether the patient goes straight for surgery or may have a biopsy and then have neoadjuvant chemotherapy.

And we function very much as a group, with 6 medical oncologists, 2 clinical oncologists who do the radiation and some chemotherapy and 5 or 6 cancer-dedicated surgeons. So that is our medical group within a hospital environment.

DR LOVE: And how many newly diagnosed patients do you see a month or a year?

PROF LEDERMANN: So within the group we will see around about 150 new ovarian cancers a year.

DR LOVE: So it’s interesting that you approach it that way. I imagine obviously you think through a lot of decisions that are being made on an individual basis by docs in individual practices, for example, in the United States.

Approach to mutation testing for patients with newly diagnosed ovarian cancer

DR LOVE: How do you as a group decide how you’re going to approach mutation testing in general in a newly diagnosed patient, and how does it vary based on the patient’s family history or ethnicity?

PROF LEDERMANN: So again, at the moment we’ve got funding through the National Health Service system for testing or offering to test every single newly diagnosed patient with ovarian cancer for a BRCA mutation. So the funding is there. It’s just a question of setting up the mechanisms of doing that.

Now until recently, when we could only make use of this in the recurrent disease setting by using PARP inhibitors for recurrent disease, we had some time to get patients tested for BRCA mutation. The situation now is different.

With the data emerging from SOLO1, we need to make decisions much earlier, particularly whether or not we’re going to use bevacizumab, so that we are ready by the time the patient completes her chemotherapy to offer her maintenance olaparib if she has a BRCA mutation.

So that does put an increased challenge on the system, and I have to say that at the moment it’s still in evolution, because we’re only just getting hold of olaparib in the first-line setting. So we haven’t yet fully set up the robust mechanisms of testing, which will really need to be done at the time of surgery, the time of initial diagnosis, even before the patient comes to the medical oncologist for their chemotherapy. Because we need about 6 weeks’ turnaround, 4 to 6 weeks’ turnaround, for these tests to come through.

DR LOVE: So putting aside the issues in terms of what you yourself are able to access, maybe just kind of looking at it from a theoretical point of view, optimally, what kind of mutation information would you like to have in a newly diagnosed patient? And specifically, is it just germline BRCA1 and 2? Is it a germline panel? And what about somatic testing like NGS? Ideally would you like to have all of that?

PROF LEDERMANN: Right. So in an ideal situation, I would like to have both germline testing and somatic testing, because we would otherwise miss around about 5% to 7% of patients who have a somatic mutation in the BRCA genes. And currently, unfortunately, we’re not funded for that.

So in an ideal world, I would do both somatic and germline testing. Why both? Because there is a false-negative rate of germline testing in the tumor. That is to say, you can miss a large germline deletion if you just look at the tumor. So I think you have to look at both the tumor and the germline.

Primary debulking surgery versus neoadjuvant systemic therapy in the up-front management of advanced ovarian cancer

DR LOVE: So let’s talk a little bit about clinical evaluation of the newly diagnosed patient. And as you know, we did a survey of I think about 25 clinical investigators, both gynecologic oncology investigators as well as medical oncology investigators, to try to assess how they were approaching it. We asked you and Dr Brad Monk, the other co-chair of this project, also to take this survey, so I kind of want to allude to that as we go through this.

But one of the first questions that we tried to tease out, and we looked at this based on whether the patient has BRCA germline or not just to see how that would affect it, but one of the first questions we looked at is the question about whether therapy would be given neoadjuvantly or not. Can you talk a little bit about how you approach that decision and what the key clinical research databases are that have led you to your approach?

PROF LEDERMANN: Yes, yes. So this is a very complex question and a very thorny question. There’s still a lot of discrepancy in terms of people’s views as to whether or not it’s completely equivalent to offer a patient neoadjuvant chemotherapy or primary surgery on the one hand or whether one should strive to operate in all those patients where it’s possible. And of course the confounding factor there is the quality of the surgery and the quality of the support structures that exist for that surgeon.

So you see up and down the country differences in the primary surgery rate. So we would offer probably around 30%, 35% of patients neoadjuvant chemotherapy, because we have very good surgeons, very good support structures. Whereas you could go to another part of the country, even though it’s a designated center, and about 80% to 90% of the patients would be offered neoadjuvant chemotherapy and surgery halfway through.

And that’s not to say that necessarily it’s because one surgeon is better than another, but that may be a factor. But it’s also to do with the timing, the space that surgeons have on their operating lists, the support structures in terms of GI surgeons on hand, ITU beds, et cetera, et cetera.

So there is this debate that continues as to whether or not they are equivalent or not, and we’ve recently completed within Europe, led by the German group, the TRUST trial, which has been a trial of neoadjuvant chemotherapy versus primary surgery, and you might say, “What? Not another one.” But yes, this is different, because this is only being done in centers that are highly specialized in the field of surgery. So you’re controlling, if you like, for top quality surgery, and this trial has completed recruitment, but we don’t yet have the outcome.

But I think many surgeons still feel that primary surgery is the way forward, if you can do it and get the patient to no residual disease. Certainly in Europe we feel that no residual disease is really the divider between good and less-good prognosis. We no longer would accept that less than 1 cm residual disease is a reasonably good prognosis. It’s either zero residuum or some residuum in terms of the prognostic division.

Value of neoadjuvant chemotherapy versus up-front debulking surgery followed by adjuvant chemotherapy

DR LOVE: I see these practices in the light the way, say, a general medical oncologist sees them, because we’re doing CME on all parts of oncology. And what you call neoadjuvant is kind of a lot different than, say, how breast cancer people approach neoadjuvant. And weirdly enough, it ties into the concept of is there a role of debulking surgery in a patient you can’t cure, and particularly in a patient who has disease outside the abdomen? When a patient presents with a pleural effusion, I mean, what would you call neoadjuvant therapy in a patient like that?

PROF LEDERMANN: Yes, first of all, I agree with you that the term neoadjuvant is not a good term. It’s primary surgery or delayed surgery, I think, is preferable. Many of us do use that term and try and avoid this concept of neoadjuvant.

DR LOVE: That’s interesting.

PROF LEDERMANN: In terms of the value of undertaking debulking surgery, whether primary debulking surgery or an interval debulking surgery, in patients with disease out of the abdomen, there are data to show the patients’ progression-free survival and quality of life are improved by the surgical component. But it tends to be in those patients who are responding. So we wouldn’t undertake interval debulking surgery in a patient with Stage IV disease if there hadn’t been a good response to the chemotherapy.

But certainly we do see improvements in progression-free survival. Although that surgery itself in that group of patients is not curative, it adds to the extension in control of disease.

DR LOVE: Yes, I mean, I don’t know, maybe everybody got this a lot earlier than I did, but it took me a long time to realize that, I guess, really ultimately, a lot of times when you do debulking surgery it’s kind of similar to giving systemic chemotherapy, or, I could say, giving chemotherapy in a noncurative situation is like debulking. And it kind of makes sense that it might be easier to just scoop it out than to give somebody 10 cycles of chemotherapy. But it’s a different concept, I think, than most surgery for cancers.

PROF LEDERMANN: I think that’s right, and I think that’s one of the arguments that the surgeons would use to say, “Go for surgery first.” Because this scooping out by surgery reduces the tumor burden, therefore potentially reduces the population of cells that are resistant to chemotherapy. So it gives chemotherapy, if you will, an easier time, because you’ve got less disease to deal with.

The counterargument is that you make the patients very sick with the surgery. They have long hospital stays. Shrink down the tumor with a bit of chemotherapy, then you get the tumor out more quickly and then you complete with chemotherapy. So that’s the counterargument. And of course we don’t know who’s right.

Higher incidence of visceral metastases in patients with ovarian cancer and BRCA1/2 mutations

DR LOVE: You were nice enough to participate in a CME meeting we did in Honolulu, where SGO decided to have their 50^th anniversary, so we got to fly for, whatever, a long, long time to get down there. But in any event —

PROF LEDERMANN: It was a long way.

DR LOVE: — as you know, in the meeting we were showing videos I had recorded of docs around the country who had cases, and there was one we didn’t actually show because we ran out of time, but it was an oncologist in Maine, medical oncologist who had a patient who had liver mets. And the surgeons wanted to do debulking, and they did. And he was asking — I didn’t get a chance to ask you all the question — his question was, like, from a medical oncology point of view, it seemed a little counterintuitive, but within this model that you just described, it kind of makes sense.

PROF LEDERMANN: Yes. I mean, I think the parenchymal liver metastases at presentation, I think that is a slightly different situation in terms of this decision-making process, and we would certainly go with primary chemotherapy in a patient like that. Because that patient, firstly, may not respond very well to chemotherapy, in which case we certainly wouldn’t subject her to surgery.

If she had, however, a very good response to chemotherapy and after 3 cycles we saw, let’s say, just 2 or 3 very small liver metastases but still some bulky disease in the pelvis, then you could argue that actually you might give her better symptom control, pelvic control, by doing the surgery. She’s a lot fitter. The tumor volume is smaller. It won’t be so morbid. Do the surgery, then carry on.

But we do know, of course, that unfortunately patients with parenchymal liver metastases do worse. They are Stage IV, IVB disease.

DR LOVE: Did not know that. How much difference is there? I never heard that before.

PROF LEDERMANN: So there’s a paper that was published by Charlie Gourley and colleagues in the JCO some years back now, where he described a series of patients with parenchymal liver/splenic metastases, and a higher proportion of those patients had BRCA mutations.

DR LOVE: Wow. That’s interesting.

PROF LEDERMANN: It’s a case series description.

Optimal timing of debulking surgery with (neo)adjuvant chemotherapy for newly diagnosed ovarian cancer

DR LOVE: So another issue about neoadjuvant therapy that I’ve heard a lot of people bring up is the question of at what point do you send the patient to surgery? Is it maximal response? A lot of docs in practice have questions about — they’re concerned. Also the issue of bev in the neoadjuvant setting. Bevacizumab. Is that ever a consideration?

PROF LEDERMANN: Okay, so 2 things there. First, the timing of chemotherapy in the primary chemotherapy or neoadjuvant setting. I mean, the data from the trial suggests that it should be done halfway through, which most commonly is after cycle 3. But there are occasionally diary issues that make it delayed until cycle 4.

So I think 3 or 4, but 3 preferably, which also gives time for more chemotherapy post surgery. Because we know that in the vast majority of patients, although the surgery can be successfully completed and the patient, let’s say, has no macroscopic residual disease after the interval surgery, there is still live tumor under the microscope. And they need to continue with chemotherapy. And so they need to have another 3 cycles, I would say, of chemotherapy. So that is how we’ve commonly divided it, right in the middle, so that patients get chemotherapy after surgery.

Efficacy and safety of bevacizumab-containing neoadjuvant therapy followed by interval debulking surgery for advanced ovarian cancer; duration of maintenance therapy with bevacizumab

PROF LEDERMANN: In terms of the bevacizumab, there were several questions that were raised. Firstly, is it safe to give bevacizumab before you operate on patients in that neoadjuvant space? And secondly, is it of any value to give patients bevacizumab in that time?

In terms of the safety, there are now several trials. There was a trial conducted by the French group, a trial conducted by the Spanish group, that specifically looked at neoadjuvant chemotherapy with bevacizumab, and we’re doing that as part of the big ICON8B trial at the moment, which is a prospective study looking at bevacizumab and weekly paclitaxel and 3-weekly paclitaxel. So that’s story’s not completely gone away yet.

And the bottom line is that it does seem to be safe to do that, so long as you miss the bevacizumab out on the cycle before surgery, so the third cycle, and that you don’t restart it at cycle 4, but you start it at cycle 5. Which then raises the question, if you’re giving bevacizumab at cycle 1 and cycle 2, is it of any value to the patient? Does it improve the response rate? We know that bevacizumab given over a period of time with chemotherapy will improve the response rate, and of course that can be important in the neoadjuvant setting, where you want maximal cytoreduction by chemotherapy before surgery. But I think it’s questionable whether 2 cycles of bevacizumab are going to improve the response rate to make the neoadjuvant chemotherapy — to make the surgery easier.

So many people now in practice might defer the decision about adding in the bevacizumab to cycle 5. So you put it in at cycle 5 and 6, after the surgery, and then, of course, the greatest value in the bevacizumab is during the maintenance phase. And that will continue beyond chemotherapy. But I think it is safe, providing you manage it correctly, but it’s questionable whether it will add value in that presurgical phase of treatment.

DR LOVE: So in terms of bevacizumab maintenance, how long do you continue it?

PROF LEDERMANN: Again, in the UK we use the lower dose of bevacizumab, which was from the ICON7 trial, 7.5 mg/kg, and we use it for 12 months not 15 months, because again, that was part of the ICON7 trial.

And we’re not alone in doing that. Some of the Nordic countries do that. So there’s a bit of a mixed picture across Europe, with some European countries using it as in GOG 218 but others using it as in ICON7.

ICON8: Results of a Phase III trial evaluating response to neoadjuvant weekly dose-dense chemotherapy as first-line therapy for epithelial ovarian cancer

DR LOVE: You mentioned the issue of including the bevacizumab in the neoadjuvant therapy. I wonder if, particularly in patients with ascites, whether it makes a difference.

PROF LEDERMANN: It is supposed to be the case that if you use bevacizumab that it will help to control ascites, and we’ve seen that certainly working well in patients with recurrent disease. But actually when you’re dealing with a chemo-naïve patient, the chemotherapy, if it’s really going to work to the point where the patient gets surgery at cycle 3, I think you’ll see the ascites dry up very quickly. And I think if it doesn’t, with or without bevacizumab, the patient is going to have a much harder time with surgery and is likely to do less well. In fact, it’s one of the indicators that we would use after cycle 3 when we reevaluate the patient for surgery after neoadjuvant chemotherapy, is whether or not they have persistent ascites.

And until recently, certainly if patients had significant ascites at the time of reevaluation, we would probably not go ahead with surgery. We are moving a little bit away from that point, because the data from the ICON8 surgical component of the trial, this is without bevacizumab, were presented at ESMO last year and showed that those patients who did not have a particularly good response but went ahead with surgery after cycle 3 actually had a pretty reasonable progression-free survival.

So I think the message coming from the trial, at least, was that we shouldn’t be too restrictive in deciding who should and shouldn’t have surgery purely on the basis of whether there’s ascites. It’s whether you can operate to zero residuum. That’s the critical thing.

Influence of performance status and age on the choice of neoadjuvant versus adjuvant chemotherapy

DR LOVE: Another factor that I’ve heard people talk about in deciding about neoadjuvant versus adjuvant that we tried to tease out in the survey a little bit was the issue of age and performance status. How does that factor into your decision?

PROF LEDERMANN: The surgeons in reviewing the case and reviewing the radiology will, of course, bring that into account, and comorbidities in particular. And they may feel that although technically it is a resectable tumor the comorbidities are such that they would rather the patient have 3 cycles of chemotherapy and then do the operation. Because certainly the trials of neoadjuvant chemotherapy have shown that the period of hospitalization is certainly shorter if you operate later. So I think that comorbidities, and there are various scoring systems that are used in terms of assessing patient’s fitness for surgery, would be more of an indicator about whether to go down the primary chemotherapy or primary surgery route rather than age, per se. So I think age, we try not to use age per se as a determinant.

DR LOVE: So I want to talk a little bit about the patients who start out with surgery and how you determine treatment postop, but before we do, anything else you want to say about neoadjuvant therapy? Maybe any myths, misperceptions that you think may be out there about that? Or any important points we haven’t talked about?

PROF LEDERMANN: I think it’s still debatable as to whether the two are equivalent. And I’m a little anxious about saying at the moment that they definitely are. Although that slightly flies in the face of the clinical trials that we’ve seen, which show the two are equivalent.

I really think the TRUST trial will be very important in helping us decide this, where it’s only being done in highly specialized surgical centers. Because I think there’s a good rationale for surgery, if you can do it, and you can do it well and safely for doing it first. But we need to see.

Activity and ongoing investigation of PARP inhibitors in combination with chemotherapy and/or bevacizumab and as maintenance therapy for patients with newly diagnosed ovarian cancer with and without germline BRCA mutations

DR LOVE: So 1 other question about neoadjuvant therapy, which is, if you know if the patient, let’s say, has a BRCA germline mutation from the beginning, how does that affect the way you might approach neoadjuvant therapy, and would you ever include a PARP inhibitor?

PROF LEDERMANN: So I think I would, at the moment, really only include a PARP inhibitor at the end of chemotherapy. Certainly from what we’ve seen of combining PARP inhibitors, particularly olaparib and chemotherapy, is that it increases the toxicity of the chemotherapy.

So the trials that combine carboplatin/paclitaxel and olaparib, and this was the study in platinum-sensitive recurrent disease, showed that you had to compromise not only the dose of olaparib but you had to compromise the dose of carboplatin. And that’s something you don’t really want to do in patients receiving primary chemotherapy. So I think the mixture of getting the response with chemotherapy, getting the tumor reduced either at the beginning or halfway through, is the best way of getting the patient into at least a partial response so that she can have olaparib maintenance.

Bringing the olaparib in further forward, at the moment, I think does not make sense. I mean, there are discussions going on about whether one can use olaparib and sacrifice drugs like paclitaxel, which I think is an intriguing question and may be something to look at, perhaps not in the patients with the most bulky disease, because we know that patients with a BRCA mutation will often respond better to their initial platinum than a patient without a BRCA mutation.

So those ideas are being floated of combining carboplatin and a PARP inhibitor without paclitaxel. But there are other ways that we could do this to avoid the toxicity of paclitaxel, namely, bring in pegylated liposomal doxorubicin and carboplatin at the beginning. And there’ve been studies already that have shown that this is a reasonable approach.

So that, I think, is the way we may well move forward as PARP inhibitors become more firmly embedded in our management of first-line treatment. And I think they will, because my prediction is that the results of the PAOLA-1 and PRIMA study, and perhaps too the VELIA study, I don’t know, will show a benefit of PARP inhibitors not only in the patients with BRCA mutation but also in those with BRCA wild type.

DR LOVE: Also, I think you mentioned the PAOLA study that combines bev and PARP inhibitor too, right?

PROF LEDERMANN: That’s right. Yes. Yes, so all those patients will have bevacizumab, and half the patients get olaparib in addition.

DR LOVE: Right. Yes, and I was going to kind of get to that. So I was actually thinking about the fact that if the patient gets neoadjuvant therapy but has a BRCA germline mutation, then they go to surgery and then they get their postop therapy, I don’t know whether that fits the indication, at least in the US, or even the clinical trial. I don’t think patients like that could go on SOLO1, could they?

PROF LEDERMANN: Certainly patients with neoadjuvant chemotherapy could go on SOLO1, yes.

DR LOVE: Oh, is that right?

PROF LEDERMANN: Yes. I think 25% of the population of SOLO1, 25% to 30% had neoadjuvant chemotherapy.

DR LOVE: So is it generally accepted then that — maybe the indication is this way too — that a patient like that, again, if they have a germline mutation, would be eligible for olaparib maintenance after they finished their chemo. Right?

PROF LEDERMANN: Correct.

DR LOVE: Okay.

PROF LEDERMANN: It’s what the situation is at the end of your primary mix of surgery and chemotherapy.

DR LOVE: Right.

Consideration of bevacizumab as a component of (neo)adjuvant systemic therapy for patients with ovarian cancer

DR LOVE: And then also, we’re going to get to the issue of the patient who goes to surgery first, but it also does bring in the issue of bevacizumab. In these patients who get neoadjuvant chemotherapy, let’s say they’re BRCA wild type, just to make it simple, in general, are all of them getting bevacizumab? And bevacizumab maintenance postop?

PROF LEDERMANN: So at the moment, I mean, it’s slightly complicated in the way in which it’s available within the United Kingdom. That is, with patients —

DR LOVE: Yes, maybe just is that — rather than whether that happens in the United Kingdom, more is that considered kind of the standard, from your point of view, ideally? Theoretically?

PROF LEDERMANN: Okay, let us work to the European label of bevacizumab, which is for any patient who has Stage IIIA or greater disease. So in those patients, therefore, they would technically be able to have bevacizumab. I think the big question is, is it really of value? And how much value? And is that value better than adding a PARP inhibitor? So at the moment, we only have that information for a patient with a BRCA mutation. And what we see when we look at the SOLO1 data in terms of progression-free survival is just way different from what we see with bevacizumab.

So, I think, with few exceptions, but I think generally what we would do with a patient who has a BRCA mutation, and that’s why it’s important to know this early on, is that we would not give the patient bevacizumab. That we would give them olaparib maintenance.

When we see the results of PAOLA-1, of course we may change our minds. Because we may see a benefit in both BRCA wild type and patients with a BRCA mutation. And that’s on the back of bevacizumab. Now the difficulty, of course, in PAOLA-1 is that no one has done a comparison of PARP inhibitor versus bevacizumab. Because both arms of PAOLA-1 get bevacizumab.

But we’ll have an idea of what happens in PRIMA, although the populations are different. The PRIMA populations are selected to be patients with a poor prognosis, because they’re patients who’ve got bulky residual disease or Stage IV disease. And so their expected rate of recurrence is going to be higher than it is in PAOLA-1. But no patients in PRIMA are getting bevacizumab.

So there’ll be some inferences that one could draw across the 2 trials, notwithstanding the fact that cross-trial comparison is quite a dangerous thing to do. But people will start, I think, then to make decisions about whether or not they should be or shouldn’t be using bevacizumab so much in the first-line setting.

Choosing between bevacizumab and olaparib after primary debulking surgery; importance of timely BRCA mutation analysis to aid in this decision

DR LOVE: So that really kind of does lead into the whole story of the patients who go to surgery first, and you alluded, which I’m really curious to hear you talk a little bit more about, about at least how in Europe the regulatory groups kind of respond to this. But again, can you talk about the way you approach the patient who’s going for primary debulking surgery and how you decide what you’re going to be doing postop?

PROF LEDERMANN: So the patient who goes for primary debulking surgery, we will only use bevacizumab if they’ve either got Stage IV disease, let’s say pleural effusion, or that they have had unsuccessful surgery. And by that I mean they’ve got more than 2 centimeters of residual disease. So those patients we would use, and we are funded to use, bevacizumab. Now, at that point, of course, this is the difficulty that we’re going to run into: If that patient turns out to have a BRCA mutation, we would rather not give the patient bevacizumab but wait until she has had the 6 cycles of chemotherapy and then put in olaparib.

So to do that, we would have to get the results of that BRCA mutation analysis very quickly, before we make that decision — latest really cycle 2 — cycle 2, cycle 3 of the postoperative chemotherapy so that we could actually maximize the effectiveness of bevacizumab or not give it so that we give olaparib.

It’s slightly easier in the neoadjuvant population, because if we work to the principle that maybe we don’t need to give bevacizumab for the first 2 cycles because it’s not going to do very much, we’ve got time to wait for the BRCA mutation analysis, cycle 3, to know after surgery whether or not we now introduce the bevacizumab or we wait until cycle 6 and introduce the olaparib.

DR LOVE: As you know, we really love asking people clinical scenario questions and kind of see what they do. We ask investigators. We ask docs in practice. And one of the things I’ve noticed, more in the recurrent setting, but it’s just kind of an interesting idea, is, I’m starting to see a bunch of people, not the majority, maybe 20% of docs talk about, again, this is more in a recurrent setting, but I kind of wonder about it even in the up-front setting, using carbo/paclitaxel/bev for the 6 cycles, then stopping the bev and adding PARP inhibitor maintenance. Any thoughts about that strategy?

PROF LEDERMANN: I mean, if you look at the results of GOG 218, there was an arm that had bevacizumab with chemotherapy and then placebo maintenance. And that arm did no better than the placebo arm, really, in terms of progression-free survival. So I don’t think that really makes sense from the point of view of improving progression-free survival.

Where you could argue there may be benefit is if you’ve got a patient with bulky disease at the end of her surgery. You will get an 18% to 20% higher response rate, for whatever that’s worth, because it doesn’t translate into progression-free survival benefit if you give them bevacizumab with chemotherapy and then stop.

So I don’t think that really, as progression-free survival is a stronger, although an endpoint with problems of itself, is a stronger endpoint that response rate, I don’t think there’s much value in giving bevacizumab with chemotherapy and then stopping. And really if you’re doing that just because you haven’t had a chance to test her for the BRCA mutation, I don’t think that’s probably the best way to practice. What you need to do is to develop the logistics so that you can get the BRCA mutation analysis quickly and make that decision early.

DR LOVE: So as you said, these things are all bubbling around. A bunch of trials are going to be reported. Incidentally, the PAOLA study, is that going to be presented this year at ESMO? Or we don’t know?

PROF LEDERMANN: I think it’s likely that it’ll come to ESMO. I mean, what I’ve heard is that it’s likely to come to ESMO, yes.

Progression-free survival benefit with bevacizumab in combination with and after chemotherapy for patients with surgically debulked disease

DR LOVE: So in any event, you alluded to that and a number of these other — so we’ll see what happens, but perhaps there’s going to be more discussion of bevacizumab. But I would like you to kind of go through the data that’s out there.

You mentioned the GOG 218 study, also the ICON7 study, about the issue of bevacizumab in first-line therapy. What was found in terms of clinical research? And then maybe if you could talk a little bit about how the regulatory agencies responded to it and how it really affected clinical practice.

PROF LEDERMANN: Okay. So we had 2 trials, 218 and ICON7, that ran more or less at the same time and were, in fact, published contemporaneously. But they were different in the sense that ICON7 used half the dose of bevacizumab, 7.5 mg/kg as opposed to 15 mg/kg, and gave the maintenance of the drug for 12 months as opposed to 15 months.

But both trials showed overall there was improvement in progression-free survival of a matter of a few months if you gave bevacizumab with chemotherapy and then as maintenance afterwards. And really the maximum difference in progression-free survival was just at the point where you’re stopping the bevacizumab, which in itself has raised some questions as to whether the drug should have been given for longer.

But notwithstanding that, neither of the trials showed an improvement in overall survival. And I think it was this difference that led to a difference in the regulatory review, if you like, across the Atlantic. And, of course, it was only 218 that was submitted as the definitive trial to the regulators in Europe, the EMA.

ICON7 was not submitted for regulatory approval, although the data, of course, were made available. And the EMA decided at the time to grant a license for bevacizumab on the basis of this progression-free survival difference, even though there was no difference in overall survival.

But my understanding is that in the US it was not even submitted to the FDA, because at the time, the FDA would not entertain a situation where you saw an improvement in progression-free survival of a matter of a few months without any improvement in overall survival. So we had this difference across the Atlantic that lasted for several years.

DR LOVE: When you move away from the regulatory issues and just take a look at your vision of the risk/benefit, how do you see it from a patient point of view?

PROF LEDERMANN: My experience of using bevacizumab has generally been good. I mean, there are patients in whom I’ve had to stop the drug early because of side effects, hypertension, proteinuria probably being the most common. But the majority of patients will tolerate it well. It’s inconvenient to come in every 3 weeks to have an infusion, and that has an effect on the quality of life that they have to come up to a hospital or a facility to have an infusion. But generally speaking, it’s well tolerated.

But the dropout rate of both the trials, there were patients dropping out through toxicity, through the fact they were fed up with the treatments before they finished their 15 months or 12 months. But it’s a manageable drug.

Incidence of gastrointestinal toxicities with up-front bevacizumab for ovarian cancer

DR LOVE: When bevacizumab first came out in ovarian cancer, it wasn’t used in far-advanced disease. There were concerns about things inside the abdomen, fistulas, et cetera. How has that shaken out over the years? And how much of an issue is it in the earlier-stage disease?

PROF LEDERMANN: That’s a really important point, because yes, we definitely saw in patients with recurrent disease a higher incidence of perforation and fistula formation. And there was concern that that might translate into the first-line setting.

But that wasn’t actually borne out by the data. We did not see significant increases in fistula rates. There were some patients who had fistulas or perforations, but the numbers were small. Interestingly, and whether it’s relevant, they were lower in the patients in ICON7 than in 218. Maybe there’s a dose effect there? It’s not because the patients were of necessarily different types of tumor in terms of the amount of residual disease. But again, it’s quite difficult to make those comparisons.

But I think we’re still nervous — at least in our unit — we’re still nervous about using bevacizumab in patients with miliary disease on the serosa of the intestine. So if the intestine is coated with disease, I think we are a little bit nervous about using it. But certainly in those patients who’ve had surgery, where we know what the state of the intestine is at the end of surgery, we haven’t been caught out by an increase in perforations or fistulas.

Approach to discontinuation of maintenance bevacizumab before surgery

DR LOVE: You mentioned the fact when we were talking about neoadjuvant therapy that if you do use bev, of course you have to stop it before surgery. I’m just kind of curious, as these patients are receiving maintenance bevacizumab, how do you approach the issue of people who need minor surgical interventions, major surgical interventions, dental extractions, et cetera?

PROF LEDERMANN: So for minor procedures, we just try and time it so if it is a dental extraction, we’d say, try and do it 3 weeks after your last dose, and then we’ll delay you a week or 2 weeks or something like that. If it’s a surgical emergency, you have to deal with the emergency and face the consequences. Actually, they’ve not been too bad. I don’t recall any patients that have gotten into really serious problems.

I think our concern about stopping before patients have definitive debulking surgery and not starting too early is, particularly in patients who have bowel resections and primary anastomosis, you want to minimize the chance of an anastomotic leak. So that’s why we say, don’t start for another 6 weeks postoperatively, particularly if you had a bowel surgery.

MSI/DNA mismatch repair testing in ovarian cancer

DR LOVE: So I want to finish out talking about your cases, but I had jotted down a couple of notes of things I wanted to ask you. I just want to go back before we get to your cases. We were talking about this issue of looking for BRCA mutations, HRDstatus, et cetera. What about MSI testing in ovarian cancer? How do you approach that?

PROF LEDERMANN: We don’t, really. I mean, we haven’t been doing it. We, of course, do it in endometrial cancer, but we haven’t been doing it as a routine in ovarian cancer. And I think we need to accumulate more data on MSI and checkpoint inhibitors in ovarian cancer. It’s an interesting question, but it’s not one that I have any information at the moment about.

DR LOVE: Actually, our survey showed a lot of people are doing MSI testing, particularly in advanced disease. We’re seeing people test almost — it seems like any metastatic adenocarcinoma of any type nowadays, even if it’s looking for less than 1% with the idea, as you say, to try to look for a checkpoint inhibitor.

PARP inhibitor-associated insomnia

DR LOVE: The other question I wanted to ask you about, we were talking about PARP inhibitors earlier. I heard something, actually, I think it came out at the meeting you were at, and I can’t remember what you said or if I even asked you. Remember when we talked about this issue of insomnia with PARP inhibitors?

PROF LEDERMANN: Yes.

DR LOVE: I forgot, did you say you’ve seen it? Or you think it exists?

PROF LEDERMANN: Oh, yes.

DR LOVE: Really.

PROF LEDERMANN: Oh, very much so. And I think I didn’t notice it so much when we were going through the trials, and it may be because we weren’t specifically asking the patients those questions. But now that we’re not focusing on the standard adverse event-type questions, when we’re outside of the trial, patients begin to volunteer these symptoms, and it just becomes clearer and clearer.

I think it’s difficult. We’re using a lot more niraparib at the moment because of the way that it’s funded. So we’re certainly seeing a lot of it with niraparib. But I think you do see it with other PARP inhibitors, too. Not in every patient, of course, but it is certainly something that is, I think, an under-reported adverse event.

DR LOVE: Is there anything that’s particularly characteristic about the insomnia? And have you been able to figure out what it might respond to in terms of management?

PROF LEDERMANN: No. So we’ve tried. For instance, with the niraparib patients who take the drug once a day, I’ve said to them, “Maybe you should take the drug in the morning.” And that doesn’t seem to make any difference. So no. I’ve not found any solution that I can recommend.

DR LOVE: Interesting.

Case: A woman in her late 50s is diagnosed with Stage IV high-grade serous ovarian cancer with a germline BRCA1 mutation

DR LOVE: So let’s talk about your cases, and we can begin with your first case, the 59-year-old woman. Can you talk a little bit about her?

PROF LEDERMANN: So this was a lady who presented actually very ill to another hospital, with a pulmonary embolus, pleural effusions, ascites, and on CT scan she had an omental mass. And so she was actually transferred into our center from that referring hospital, having had a biopsy of the omental mass that showed a high-grade serous carcinoma. We carried on her treatment for a pulmonary embolus, drained the fluid off her chest —

DR LOVE: Can I just ask, in retrospect, had she had a progressive history of being ill? Or was this a very rapidly developing —

PROF LEDERMANN: It was a relatively short history. As I recall, it was a matter of 4 to 6 weeks or something like that. She also had a past history of breast cancer.

DR LOVE: Hmm.

PROF LEDERMANN: And this was, of course, in the time before we had the SOLO1 data. But we nevertheless immediately set about testing her, and she does have a BRCA mutation.

But by that time, we were already into the chemotherapy, and we then operated on her halfway through her treatment and then gave her bevacizumab.

DR LOVE: So you started therapy with carbo/paclitaxel?

PROF LEDERMANN: Yes.

DR LOVE: And what happened once you started the therapy?

PROF LEDERMANN: So for the first 3 weeks, not a lot. By the time the 3 weeks came up and she had cycle 2, we were beginning to see a little bit of symptomatic improvement, and she was hospitalized, which tells you how sick she was. She was hospitalized for the first 2 cycles of treatment.

But then at around about 5 weeks from the start of treatment, she was well enough to go home. She had lost a lot of weight. She had been in bed a lot. But she was clearly responding, such that she was then able to go for surgery after 3 cycles. We then added in bevacizumab. And at the end of her treatment, she basically had a remission on CT and normalization of CA-125, which was very high at the beginning, and then she has continued on bevacizumab. And she’s just about to finish her bevacizumab now. She’s just almost a year from starting the bevacizumab and remains in remission.

DR LOVE: Did she have an R0 resection?

PROF LEDERMANN: Yes, she did.

DR LOVE: I was looking at the write-up here — she had pleural effusions.

PROF LEDERMANN: Yes.

DR LOVE: Because I was thinking maybe it was related to the pulmonary embolus, but were these malignant pleural effusions?

PROF LEDERMANN: They were malignant. So she had a Stage IVA ovarian cancer.

DR LOVE: And what happened with them with treatment?

PROF LEDERMANN: They went away very quickly.

DR LOVE: Did they go away with the neoadjuvant therapy or the postop therapy?

PROF LEDERMANN: No, with the neoadjuvant. My recollection was, at the time of surgery, after 3 cycles, there were no pleural effusion on the CT scan.

DR LOVE: Just out of curiosity, if they were still there, would you still send her to surgery?

PROF LEDERMANN: I think that’s an important question, and it’s the same question that relates to if there was still ascites there, we would have gone to surgery. We would have been a little bit more circumspect about doing that. And I think we would have taken into consideration other aspects like the bulk response, like the CA-125 response. Her CA-125 levels were very high at the beginning and really came down very precipitously. So we felt this patient was really on the trajectory of a very good response to platinum-based chemotherapy.

And therefore I think even if she had small effusions, we would have probably gone ahead with the operation. But, in fact, she did not have effusions.

Counseling patients with ovarian cancer with germline BRCA mutations who did not receive PARP inhibition as a component of initial therapy about potential use of these agents in future treatment

DR LOVE: Can you talk a little bit about what you see for her long-term outlook and how you see managing her as she moves forward in the future? Also, she has a BRCA mutation. Would you consider adding olaparib? Even delayed?

PROF LEDERMANN: So that is the very question that she asked me last month. So she was 9 months into her maintenance of bevacizumab, and she said, but what about the SOLO1 data? Should I now not switch to olaparib? To which I answered, first of all, we wouldn’t be able to get it for her at this point in time. Secondly, we just don’t know about adding in olaparib late on in terms of maintenance, and would we go on for another 2 years?

It’s all uncharted waters, so I said to her we would, of course, use the PARP inhibitor if she had a recurrence. I also said to her that a small percentage of patients with a BRCA mutation will go through chemotherapy with or without bevacizumab and not have a recurrence.

And we’ve all got patients on our books that are 10, 14 years out with a BRCA mutation that had 1 line of chemotherapy and have not had a recurrence. It tends to be patients with BRCA, but not entirely. There are some patients without a BRCA mutation who are long-term survivors. That tail of about 20%, 22%, existed before we ever had PARP inhibitors.

DR LOVE: This thing when patients find out about new studies, and this woman finding out about SOLO1 and asking about adding in olaparib, I was just flashing — if the PAOLA study’s positive for combination, then you’re going to have patients, BRCA germline, who are on olaparib are going to say, can we add in bev? Right?

PROF LEDERMANN: I think that’s going to be harder, because there is no comparison against bev in that PAOLA-1 study. It’s a shame. We really need a study. The only study that would have done that was the JAVELIN 100 PARP study that’s now stopped. And that had a bev arm versus a talazoparib arm, and that would have been nice. We would have seen the data comparing PARP versus bevacizumab. But we’re not going to see that now.

DR LOVE: Interesting.

Impact of BRCA mutation status on patient prognosis and potential responsiveness to platinum-based chemotherapy

DR LOVE: You referred to the fact or — and I’ve been hearing this for a long time, about ultimately that the prognosis of ovarian cancer is more favorable in patients with BRCA germline. I’ve always been curious, is that an innate feature of the disease? Or it is because they respond so well to treatment — platinum?

PROF LEDERMANN: I really don’t know that answer to that question. And people would also dispute that. When they look at epidemiologic data, they say, yes, the prognosis is better for the first 10 years, but then the curves come together.

I think there’s some doubt. I mean, I think what is clear is that the vast majority of patients with a BRCA mutation have a better response to platinum, certainly first line, but even second line, than patients without a BRCA mutation.

There are, of course, exceptions. There are patients who have profound platinum resistance even though they have a BRCA mutation. But in the main, these patients respond well to platinum. And that’s because the mechanism of repairing platinum damage is not that dissimilar from repairing the double-strand breaks that occur in the presence of PARP inhibitors and with HRD being the common deficiency in these patients with a BRCA mutation. You can imagine that you get good responses to platinum-based treatment as you would with PARP inhibitors.

Case: A woman in her early 70s with Stage IV high-grade serous ovarian cancer with a germline BRCA1 mutation receives olaparib on the Phase III SOLO-1 trial

DR LOVE: So I’d like to hear about your second case, the 70-year-old lady, because in your write-up it kind of hit a nerve in the deepest part of my brain of memory with a Sister Mary Joseph nodule. I vaguely remember that from med school, and I don’t even remember why they call it Sister Mary Joseph, but can you talk about what happened with this patient?

PROF LEDERMANN: She was a nurse that described it, I think.

DR LOVE: Oh, that’s what it was?

PROF LEDERMANN: Yes. She described this occurring in patients with advanced gynecological, particularly ovarian cancers.

DR LOVE: Wow. So what happened with this lady?

PROF LEDERMANN: So this is a very interesting lady, because she presented with this nodule around her umbilicus and bulky disease —

DR LOVE: That’s how she presented? She came in asking about that?

PROF LEDERMANN: She actually presented with it, yes, she did.

DR LOVE: Wow.

PROF LEDERMANN: And that was biopsied and shown to be high-grade serous cancer. The problem was that she was a Jehovah’s Witness.

DR LOVE: Wow.

PROF LEDERMANN: And we discussed about whether or not we should operate on her at all. In fact, she started off with neoadjuvant chemotherapy. Had a moderately good response. But the surgeons felt after 3 cycles that actually it was still too risky for her to have surgery without a transfusion.

And so she went through 6 cycles of chemotherapy and had a very good partial response. Now, this was just at the time of SOLO1. So she was able to enroll on SOLO1 because she also was found to have a BRCA mutation.

DR LOVE: But she got on SOLO1 without surgery?

PROF LEDERMANN: Without surgery. That was allowed.

DR LOVE: I didn’t know that.

PROF LEDERMANN: So long as patients had a partial or complete response to first-line treatment, they were eligible.

DR LOVE: Wow.

PROF LEDERMANN: So she was enrolled in SOLO1, and because she did not have a normal CT scan at the end of chemotherapy, and indeed she still had an abnormal CT scan, although it was barely abnormal at the end of 24 months, she has remained —

DR LOVE: What happened to the nodule with the chemo?

PROF LEDERMANN: That completely disappeared.

DR LOVE: Completely went away, okay.

PROF LEDERMANN: So she went onto the trial drug, which we know is olaparib, because after about 2 or 3 months she developed raised transaminases to the point where we had to reduce the dose of the trial drug, and therefore we know it was olaparib. And she had 1 or 2 other side effects that made us suspicious that this was olaparib rather than placebo.

But at the 24-month point, she still had some abnormalities on the CT scan. And so at that point she was allowed to continue, as part of the protocol. She represents 1 of the 10% of patients who have gone beyond 24 months. And she continues on olaparib to this day.

DR LOVE: Wow.

PROF LEDERMANN: And the question is, when should we stop? She is not very keen to stop this, because she’s tolerating it very well, and the disease has remained in remission. So she has done extremely well. But she’s never had surgery.

DR LOVE: Hmm. Interesting. Wow. Remind me of the pathophysiology of these Sister Mary Joseph nodules. It’s like direct extension from the tumor directly? Or is it through lymphatics? Or what?

PROF LEDERMANN: This is spread through the peritoneal cavity into the urachal remnant.

DR LOVE: Oh right. It’s the urachus.

PROF LEDERMANN: It’s the urachus. Yes.

DR LOVE: Wow. Amazing. Hmm. And that never came back? That’s just gone away?

PROF LEDERMANN: No. No. It’s just it, that’s it. Gone. Yes. She’s got a little bit of pelvic abnormality on the CT scan, which is why we’re continuing with the trial drug.

DR LOVE: Huh! Wow. Interesting.

Case: A woman in her early 60s with suboptimally debulked Stage IV high-grade serous ovarian cancer receives adjuvant chemotherapy/bevacizumab followed by maintenance bevacizumab

DR LOVE: How about your next patient, the 62-year-old lady.

PROF LEDERMANN: This patient has FIGO Stage IV disease, and so she would certainly be a patient that we would treat with chemotherapy and bevacizumab. Although having said that —

DR LOVE: Could you backtrack a little bit and talk about how she presented?

PROF LEDERMANN: She actually presented with abdominal swelling and discomfort. She didn’t volunteer that she had axillary lymph nodes or inguinal lymph nodes. But they were found on presentation. And that clearly made her FIGO Stage IV. And on the basis of that, we felt that she should be given primary chemotherapy, because she has widespread disease beyond the abdominal cavity, and only consider surgery if she has a good response.

And she did have a good response to 3 cycles of treatment. And it’s interesting and actually not all that well recorded, but certainly my experience over a number of years is that patients with inguinal lymphadenopathy often do very well. And it’s a recent thing that they’ve been designated Stage IV. They used to be Stage III patients. Now they call them Stage IV patients. But they actually do quite well.

And she certainly did do very well, and she went for surgery after 3 cycles and had resection of all of the disease except a little remnant behind the liver, which the surgeon was just not able to take off. So on the basis of that remnant of disease, she then went on to have bevacizumab in the postoperative setting. And she continued with chemotherapy 3 cycles, bevacizumab, and then bevacizumab maintenance. And she is now just, I think, just finished her bevacizumab and is now in follow-up.

DR LOVE: But now how long has it been, 4 years since she presented?

PROF LEDERMANN: Yes. So she finished her bevacizumab in August ’16, so yes, it’s 2 and a half years, which again I think just reinforces this observation that we’ve had over the years that patients with inguinal lymph nodes, notwithstanding the axillary lymph node, but the inguinal lymph nodes tend to do very well. And I think actually that that is true for patients that have predominantly lymphatic disease rather than serosal disease and ascites.

They’re a different biology. I mean, we spoke about unifying Type I, Type II hypothesis at the beginning of the discussion, but actually, there’s more. Within that Type II, there are clearly different biologies.

And there’s a biology that seems to be mainly located in the lymph nodes that behaves in a different way to the biology that is widely disseminated within the peritoneal cavity with ascites and serosal disease. And those lymph node-only patients, in my experience, often do very well with chemotherapy and surgery.

Biologic rationale for and ongoing evaluation of anti-PD-1/PD-L1 antibodies in combination with PARP inhibitors for newly diagnosed ovarian cancer

DR LOVE: So I want to finish out our discussion here, and you’ve referred to a number of ongoing trials that some of them maybe are going to be presented soon. But can you kind of just take a minute and just take a breath and talk about some of the trials that are out there that you’re most excited about? Clinical trial strategies in the up-front setting?

PROF LEDERMANN: Okay. So, I mean, I really am excited about the PARP inhibitor trials. So with the three we’re expecting results. The VELIA study, the PAOLA-1 study and the PRIMA study.

Because I think at the end of that we’re going to be bringing PARP inhibitors into the first-line setting, for not only patients with a BRCA mutation, but BRCA wild type. That’s my guess. Which will, of course, have knock-on effects in terms of how we manage recurrent disease. But that’s another issue. So I think that’s going to be the most exciting thing in the next 18 months, when that all comes through and we digest it all.

Of course the next thing behind that is the question of whether or not the checkpoint inhibitors are going to have an impact in first-line treatment. Now we’ve seen what I think are really rather disappointing results so far in patients with recurrent disease, and we can conjure up all sorts of hypotheses as to why those trials may not have been as successful as we would have liked them to be.

I think one of the issues, of course, is that we’re not biologically selecting patients and tumors, and I think that’s a big mistake. We made that mistake with bevacizumab when we didn’t have a marker. We didn’t make that mistake with PARP inhibitors where we had a more focused approach with BRCA and HRD, but at the moment we’re just testing all patients with checkpoint inhibitors. But I think the area where we may see a benefit of checkpoint inhibitors in the first-line setting is in combination with PARP inhibitors. So there are some good scientific arguments and rationale for combining a PARP inhibitor with an immune checkpoint inhibitor that may enhance the immunogenicity of the tumor and the immune response.

So my hope is that the trials, the ongoing trials with checkpoint inhibitors in combination with PARP inhibitors, will show an improvement in outcome. So I think that next charge of trials, which are recruiting, and there are now 4 open trials. One trial opened and then closed. The company pulled out, so we’re left with 4 trials all looking at the same concept of combining the PARP inhibitor and the checkpoint inhibitor. So it will take 2 or 3 years to recruit these patients, another 2 years or so to get the results, but I think that will be a very interesting result.

DR LOVE: Couple final questions. When do you think the VELIA trial might be presented?

PROF LEDERMANN: I’m not sure. I suspect for ESMO. And if they take it to ESMO, I think it will be ready by then. But I’m not involved in recruiting that trial, so I’m not sure when the data are going to be sufficiently mature for analysis.

Benefits and limitations of commercially available HRD (homologous recombination deficiency) testing platforms

DR LOVE: And another question, just for my own background, maybe naïve but I’m still trying to figure out. When you talk about HRD measurement, are you measuring mutations or something else?

PROF LEDERMANN: So the commercial HRD tests that are run at the moment don’t look at mutations in the homologous recombination repair genes. So I think there are 2 separate things here. We know that mutations in the BRCA gene almost always lead to this phenotype of homologous recombination deficiency.

We suspect that some of the other mutations in the HRR genes might also lead to homologous recombination deficiency. Let’s say the RAD51B or C mutation or BRIP1, probably, right?

We know there’re also other homologous recombination genes that become mutated and may not lead to the phenotype of HRD. And that phenotype is measured not by the mutation analysis but by looking at the amount of genomic scarring that occurs in the DNA, which is a function of the homologous recombination deficiency.

So it measures the degree of loss of heterozygosity within the DNA. And the Myriad test has some other things added in too to produce a score, an HRD score, but it’s not based on mutations of HRR genes. Which is different from the BROCA test, which looks at mutations in HRR genes.

DR LOVE: But when you say it measures scarring, like, what is it measuring?

PROF LEDERMANN: Yes.

DR LOVE: Is it, like, proteins or DNA? What is it?

PROF LEDERMANN: It measures loss of heterozygosity in the genome.

DR LOVE: In the genes.

PROF LEDERMANN: What happens is, because of the repair deficiencies, you get this loss of heterozygosity.

DR LOVE: Hmm.

PROF LEDERMANN: And that correlates most closely, we are led to believe, and actually the data from the trials does suggest that from both the NOVA trial and from the ARIEL3 trial that the HRD status correlates with the benefit of the PARP inhibitors.

DR LOVE: And how would you define heterozygous recombination?

PROF LEDERMANN: I mean, if you take the BRCA gene, so there’s 1 allele that is mutated. The other is not initially, but at the time the patient develops the tumor she will lose the other allele, the nonmutated allele, and therefore she becomes homozygously deficient, if you like. And that’s associated with the cancer and the response to PARP inhibitors. Whilst the argument would go that in, let’s say, the white blood cells they’re still heterozygous and therefore you don’t get devastating effects of PARP inhibitors on the normal tissues, because they’re still heterozygous. So it’s the tumor that becomes homozygous for the BRCA deficiency or the BRCA mutation.

DR LOVE: And again, are they looking at the BRCA gene specifically when they measure that, or not necessarily?

PROF LEDERMANN: So the commercial assays for the HRD analysis don’t look at the HR gene mutations. They separately will, if you ask them to do a mutation analysis, they do mutation analysis.

DR LOVE: Right.

PROF LEDERMANN: But that’s a separate question, separate investigation.

DR LOVE: Hmm. Wow, that’s interesting.

PROF LEDERMANN: But I think that both the tests, the HRD tests and indeed the HRR test, they’re still not able to exclude patients who’ll benefit. So you may say the patient is HRD-positive. That’s great — she’s got a better chance of responding.

DR LOVE: Right.

PROF LEDERMANN: But if she’s HRD-negative, you cannot say I’m not going to give the patient a PARP inhibitor, because she could still have a very good response, even though they’re HRD-negative.

DR LOVE: Right. Right.

PROF LEDERMANN: So the tests are not good enough at the moment to be able to distinguish patients who will or won’t respond.