Ovarian Cancer — Kathleen N Moore, MD, MS

DR MOORE: Good day, everyone. It’s my great pleasure to have been invited to give this Year in Review for gynecologic cancers, specifically for ovarian cancer, with Research To Practice. I’m Dr Kathleen Moore. I’m the Associate Director of Clinical Research at the Stephenson Cancer Center in Oklahoma City, and I direct Phase I clinical trials here, and I’m a GYN oncologist. And there was a lot to talk about this year, so we’re going to get started.

The first thing I’m going to talk about is PARP inhibitors. PARP inhibitors aren’t new, so I’m not going to belabor a lot of the old data, but we did have some new data presented this year, some good and some thought provoking, that I’ll take you through.

So the first thing I want to show you is just a summary slide, and this is — these are the 5 kind of clinical trials that have been presented in front-line ovarian cancer with the use of PARP inhibitors as maintenance therapy amongst patients who have completed front-line paclitaxel/carboplatin and have either a complete or partial response.

And so this is the subsets who are BRCA. Now I say that with the one exception, which is SOLO-1, which you see in the middle of your screen, which was only done in BRCA. The rest were all comers, and these are the BRCA subsets.

So looking at the right — left-hand side of your screen, in the kind of dark blue, this is the niraparib series of studies, PRIMA done in the US and Europe, and PRIME done in China. I’ll come back to that. In the middle is olaparib, SOLO-1, as I mentioned only in BRCA, PAOLA-1, done entirely ex-US, this was bevacizumab and olaparib versus bevacizumab. And then on the right-hand side, presented in 2022, was ATHENA-MONO. There’s another ATHENA coming, ATHENA-COMBO. But this is rucaparib, so in green, versus placebo.

So a couple points to make here. One is that even though with SOLO-1 I’m talking about a study that was entirely BRCA and the rest are subsets, we pulled out the BRCA, the hazard ratios demonstrating an unprecedented benefit with the use of PARP inhibitor over placebo is incredibly consistent. So there’s no way you could say, and I don’t think anybody would at this point, that SOLO-1 was an outlier.

Irrespective of the clinical risk, SOLO-1 versus PRIMA, for example, kind of high, high, high clinical risk with PRIMA, a lot of Stage IV and neoadjuvant versus SOLO-1, and then ATHENA is kind of in the middle. The hazard ratios are very consistent, 0.3 to 0.4. So that translates to a 60-70% reduction in the hazard of progression or death with the use of a PARP inhibitor. And this really is the standard of care for women with BRCA-associated ovarian cancer, either germline or somatic, which is why testing for germline, and if germline is negative, testing for somatic, really is the standard of care for high-grade serous and high-grade endometrioid tumors.

Now things to note that are specific in 2022, I mentioned ATHENA-MONO. This is the first time we’ve seen data in the front line with rucaparib. ATHENA is a much bigger study. It’s kind of 2 studies in 1. One was showing rucaparib versus placebo was effective, and we’ve seen that here. The second, which has not yet reported out, is rucaparib versus rucaparib/nivolumab, and that is pending.

The other new data we saw earlier in 2022, on the lower right — left-hand side of your screen, is the PRIME study. Now I’ve called it niraparib, and it is niraparib, but it is manufactured in China, so it may be just a little bit different. And the homologous recombination deficiency test, which I’ll show you in another slide, was also mirrored on the Myriad test but is not exactly the same. So it’s a little bit different, but I think you can still consider it a study of niraparib. But here I’m just showing you the BRCA data, which again, very consistent. And that’s important to point out because everything else I’m showing you is predominantly US and Europe, and this is a study in China, really just showing, again, the consistency of benefit here.

Now even more important, this is what we saw in late 2022 at ESMO, is that we’re starting to see long-term follow up in the front line. This is really what we’re all waiting for, is we are impacting PFS, it is unprecedented, but are we changing overall survival? And we think the answer is yes, although what was presented is not quite definitive, but it’s a pretty good signal.

This is SOLO-1 overall survival at 7 years. We have not reached the prespecified number of events to do the overall survival analysis, the final analysis. This is not the final. This is looking at it at 7 years, which was prespecified. And what you can see here is clinically meaningful benefit in overall survival, again this is only in BRCA-associated cancers, with a hazard ratio of 0.55. Wow. And that even accounts for the appropriate 44% of patients who crossed over and got PARP later in their lines of therapy, appropriately. But the message here is number 1, we are seeing a signal of OS, and number 2, is you can’t catch people up by waiting to use PARP inhibitor in a BRCA setting later. The best place to use a PARP is in the front line, you don’t catch them up later, which was kind of a question when SOLO-1 first came out, but I hope this puts that to rest. Front line, front line, front line.

Now the other question is, is it just BRCA or are we seeing a benefit outside of BRCA. And we did see some signal of this with PAOLA. This was also presented at ESMO again, not an event-driven analysis, but it was a prespecified landmark analysis at 5 years. Three years after the PFS endpoint was the statistical analysis plan, that was at 5 years.

You can see on the left in the intention-to-treat arm there was maybe a little bit of a difference, but really overall not statistically different for OS. But on the right-hand side you see the homologous recombination deficient subset, inclusive of BRCA, so this includes BRCA but also includes BRCA wild type HRD. And you can see the difference in the number of patients who were alive at 5 years, 65.5%, versus 48.4%. So that’s a striking difference. I mean we can say we definitely are seeing a signal here that we are improving overall survival with the use of PARP really in biomarker-selected populations, and I do think that’s true.

But OS is one thing. And I love OS, don’t get me wrong, but now I’m greedy. So OS is one thing, but am I curing more patients? That’s — we’ve never talked about that. Well let’s start talking about it. So here is the surrogate for cure, which is time to first subsequent therapy from SOLO-1, meaning if you never got a subsequent therapy that meant you didn’t recur, and therefore you can kind of say you’re maybe cured.

So that’s what I’m showing you here. This was updated at ESMO. This is the placebo group at 7 years. So at 7 years we have 20% of patients who were randomized to placebo who’d never recurred at 7 years. And so there is a group of patients who don’t need PARP, and I wish I could tell you who they were, because I would love to just leave people alone, less can be more. And it’s 20% in this population. They didn’t need anything.

But look what happens when we give PARP inhibitor to this population. At 7 years we have 45% of women who have not recurred. And that slope is very horizontal, so the events are really coming slowly. So as we start to approach 9 years, 10 years, which is when we might be able to really call cure, it’s probably not going to be that much lower than 45%. So we definitely — we really feel that we’ve cured women with this regimen. We can’t call it yet, but we really do feel like this is cured for some women, and so it’s very exciting.

The flipside to that is 55% of women did recur, and so we’re not done. We’re not done. We still have to do better. But for 45% they have been off therapy for 5 years, not recurred, living their life, and that’s really what we want for all of our patients, to not just be in this hamster wheel of forever treatments. So this is pretty exciting.

We do have an update on this same idea from PAOLA-1. This is an update, now it’s at 5 years. Now I’m going to go back 1. So look at 5 years here, where 51% versus 22% olaparib versus placebo, all BRCA.

So let’s look at PAOLA, HRD, inclusive of BRCA, not entirely, 46% versus 20%. Very similar. And that slope’s starting to get horizontal. We’re going to follow this for longer and see, but this is 3 years off assigned maintenance without recurrence in almost half of the patients with HRD-positive tumors on PAOLA, and that’s very exciting for them to be off therapy and hopefully having great quality of life. So this will get updated, but I think this just reinforces use of PARP or PARP plus bevacizumab in front line, especially in BRCA, and in my opinion HRD as well.

Now I mentioned that we had some new data at — in the last few years, or just some other things to mention. PRIMA did also report out prolonged progression-free survival, and so this was — this is just 1 slideshow from Dr Gonzales-Martin’s presentation at ESMO. They have not shown OS data yet. We are anxiously awaiting that.

And I’ll remind you that PRIMA’s an incredibly high-risk population, so it’s really hard to compare to PAOLA and SOLO, but you can see in the intention-to-treat population the rates of 1-, 2-, 3-, and 4-year progression-free survival for ITT. This is a very different population. But we do hold with benefit even in the entire population with use of niraparib, which makes it a very reasonable option irrespective of biomarker.

And similarly here’s ATHENA-MONO. In the intention-to-treat population, again I will point out, this is a clinically different risk group than PRIMA, so you can’t compare them. It’s kind of between SOLO and PRIMA. They had more neoadjuvants, but not as many as PRIMA, and that’s just reflected in what you see here. This is a much less mature study. It’s mature, but it’s less follow up than the others, so really we’re just at this 24-month benchmark, 45% disease-free versus 25.4%, which is certainly different but will require additional follow up. But I think the take home is these are all good medicines, is what I would say to you.

So just to look at — we talked about BRCA. I’m just showing you this same kind of figure for BRCA wild type, but homologous recombination deficient. So of course SOLO-1 drops off because they didn’t have these patients, and you see PRIMA, PRIME on the left for niraparib, PAOLA and ATHENA, with hazard ratios that range from 0.43 to 0.66, so still very solid improvements in reducing the hazard of progression or death with the use of a PARP inhibitor. And I showed you the long-term data for PAOLA.

So I do really feel like for HRD-positive, BRCA wild type, PARP really is — it’s a standard of care, but in my practice it’s the standard of care unless otherwise contraindicated.

But this group is really our challenge. This is the group of tumors that are homologous recombination deficiency test negative or unknown. And it’s plus/minus as to the benefit of PARP for PRIMA. We do see about a 30% reduction in the hazard of progression or death, which is reasonable. PAOLA we saw no benefit of PARP/bevacizumab over bevacizumab active control. And ATHENA-MONO looked very similar to PRIMA.

So there is some benefit there for patients who are in response, their tumor are in response to their front-line chemo, we do get some benefit. It’s nowhere near the magnitude of biomarker positive, but it is clinically and statistically significant and is an option.

But the bottom line is we need to figure out why these tumors aren’t homologous recombination deficient and fix that or come at it with other medicines so that really we can change the appearance of these progression-free survival curves. These are much more vertical than what I showed you here. Look at the difference. Here’s homologous recombination test positive, here’s negative. These tumors recur very quickly.

Now I will point out, again, this was presented at SGO in March, PRIME, in your lower left. This was done in China with a different assay. But here the homologous recombination deficiency test negative tumors did really well with niraparib. So we don’t understand this really. The manuscript will be coming out.

I’m not saying it’s not real because there certainly are pharmacogenomic differences between China, Japan, Korea, where it’s a much more pharmacogenomically homogeneous population than a European clinical trial or a US clinical trial. And the metabolism of drug may be different. The biomarker may be subtly different. We don’t really understand why this is, but were I practicing in China I would not need the biomarker test to tell me to use a PARP inhibitor because clearly the benefit is there. And this was a well-done study.

So that one is a little bit of an outlier that we’re interested in thinking about more. For US populations I think there’s still clinical equipoise as to how best to treat maintenance in the homologous recombination deficiency test negative tumor, and this is really a place where clinical trials are key.

So I’ve talked about this, and I talked about PAOLA, which is the bevacizumab/PARP versus bevacizumab, which I get a lot of questions about this, especially for BRCA. Do I need to use bevacizumab, or can I just use a PARP? Do I need to use everything at once for BRCA wild type, HRD? Does that make PARP work better versus PARP alone? What do we do?

Well, the truthful answer is none of the trials answer that question. We — as you can see on the left-hand side of this figure, in PAOLA-1, because of when it was designed, it’s a great study, Isabelle Ray-Coquard I’m a big fan, I think she would say she would design it differently today with a monotherapy PARP arm. We all would, but it doesn’t have one. And in SOLO, PRIMA, and ATHENA there’s no bevacizumab combination arm, and so we can’t answer this question with clinical trials.

Fortunately there are 3 ongoing trials, and you see them listed here, NIRVANA, AGO-OVAR 28, and MITO25, are testing this question, so we will have prospective data at some point, a long time from now. But patients are in front of you today. You’re going to go see them right now. So do we have any evidence of to guide practice today? We do have a little bit from PAOLA, but also from OVARIO.

OVARIO is a single-arm study, we do have to acknowledge that, but it’s maintenance niraparib/bevacizumab in a population that is more like PRIMA than not, so there’s a lot of patients with Stage IV neoadjuvant tumors who received chemo plus bevacizumab and then went on to receive bevacizumab and niraparib for maintenance. And they did use the individualized starting dose for niraparib in this particular protocol, and you can see that in the schema here outlined on the left-hand side of this slide. Bevacizumab was for 15 months, which is the label for bevacizumab, and then niraparib was really until progression or 3 years.

And the endpoint was progression-free survival at 8 months — 18 months. This is a high-risk population, so they said if we get to 18 months that’s kind of already 6 months beyond what you see the median progression-free survival for very clinically high-risk population, that will give us a signal. And they certainly hit that.

So this is Melissa Hardesty’s — Dr Hardesty’s work, and she’s out of Alaska. So you can see the summary of the results here on the right-hand side. And this paper actually has been published. You can pull it. I think the key point is the PFS rate, which you can see — I like landmarks for single-arm studies better than medians, so you can see at 2 years in a very high-risk population you have about 60% of patients still progression free. That’s a nice signal. It’s hard to make too much of it because it doesn’t have a control arm, but that’s a pretty nice signal. You really don’t expect that otherwise.

You see the median PFS in the upper right-hand side, and then on the bottom you see by HRD status it’s a little hard to read it for you, but if you pull the paper you can see it. But looking out to 24 months on the X axis, the square is HRD test positive, and the triangle is HRD test negative. So HRD test positive, you’re approaching 75% progression-free at 2 years with combination PARP/bev. So I actually do like this combination, especially for BRCA wild type, HRD. This gives us nice safety data for niraparib/bev, we have safety data for olaparib/bev, and so if you like to use bevacizumab in the front line this gives you some data to counsel your patient if you want to layer on a PARP.

DR LOVE: How do these data compare to the olaparib/bev arm of PAOLA?

DR MOORE: Very similar. So if you compare — I wish — I should have made a slide about that. If you compare landmarks at 2 years, and we can go back to that here, for HRD test positive you’re maybe 70-75% progression free. And if we go back to the PAOLA data here in the HRD test positive but BRCA wild type, and you look at that 24-month mark you’re about 60%, but there are BRCA patients on OVARIO, so it’s going to bump it up a little bit.

So I think it’s — I think it’s probably ballpark the same for both of them, honestly. I don’t think there’s vast — I don’t look at those curves and see vast differences. I would right now consider them pretty much equivalent to one another. It just comes down to which PARP inhibitor a particular investigator is more comfortable using.

DR LOVE: What about your question of do you gain additional benefit from bev over PARP by adding bev in BRCA patients?

DR MOORE: Clinically trial wise we just have to say we don’t know the answer. We have not done the trial. We did do an exploratory analysis that’s been published. Ignace Vergote is the first author on that, and I’m on it, so you can PubMed me. I’m probably easier to spell. But we took the BRCA patients from PAOLA out, and then we matched them with patients from SOLO who got olaparib who clinically looked like the PAOLA patients. So we did this propensity weighted matching to try and recreate the missing arm of PAOLA from the patients enrolled to SOLO-1, like as if they might have been enrolled, and then we redid the progression-free survival analysis for that. And the hazard ratio did favor the combination, with a hazard ratio of 0.71, did cross 1, which you would expect for an exploratory analysis, but it looked to be 29% better with bevacizumab. And if you recall, we didn’t put this in the slideshows because it’s decade-old data, but that’s what you get with bevacizumab. You get about a 30% reduction in the hazard of progression or death.

So our take-home hypothesis for a BRCA population is that it’s additive and not synergistic. So it’s okay to do. If you’re using bevacizumab by all means add on olaparib, but if you don’t use bevacizumab we don’t feel like there’s evidence that it’s synergistic or curing more patients that you have to use it at this point.

Now in a homologous recombination deficiency test negative, BRCA wild type tumor I am very — and I really want to see the results of those 3 studies that I pointed out. It’s going to be a bit. But I’m very curious about that because we have seen data in the recurrent setting that anti-angiogenesis — targeting angiogenesis with antiangiogenic agents plus PARP in a BRCA wild-type tumor makes them look like a BRCA tumor. We saw that with Joyce Liu’s work with olaparib/cediranib and Mansoor Mirza’s work with niraparib/bev versus niraparib.

So if there is some synergy magic in that particular population, which is higher risk, maybe in front line you would see that. But we don’t have any way of assessing that right now, and these studies that are ongoing will be quite important. Because if there’s synergy that would change a lot of people’s behavior. If you knew you were synergistic you probably would start using more bevacizumab with PARP with an intention to add PARP on in that group than just leaving it up to your normal practice pattern. But we don’t have that data to answer it just yet.

DR LOVE: All right. Please continue.

DR MOORE: Very well.

So other things that came up in 2022 were frustrating and provocative, and we’re still chewing on this. But there was the — the audience should know there was an updated ASCO Guideline that came out, very late 2022, that was in follow up to one that was done in 2020 when all the PARPs came out. It was updated initially in 2022, really just to include the — the intent was to include the ATHENA data. That’s why I’m on the review. That’s why we called it. But while we were revising it all of these Dear Doctor letters, HCP letters (health care provider letters), and retractions started to come out, and so we had to respond to that as well.

And so what was happening and what continues to happen? Well we saw overall survival data come out from these 2 studies you see listed here, ARIEL4 first, and then — or SOLO-3 first and then ARIEL4. These 2 studies were, for those of you that don’t remember, the planned confirmatory trials for the original accelerated approvals for olaparib and rucaparib. So they were approved in 2014 and 2016, I think, based on single-agent data in BRCA-associated tumors with 3 or more lines of therapy.

And then these were the confirmatory trials that were designed in the recurrent setting, BRCA-associated cancers only, PARP naïve, PARP versus chemo. So SOLO-3 was platinum sensitive, olaparib versus chemo. You had to have at least 2 lines of platinum before you could come on, so this was third line and beyond. It’s a very heavily pretreated population. And interestingly they didn’t have platinum as a selection. ARIEL4 allowed platinum sensitive and platinum resistant, and they had appropriate chemo to use for that arm, versus rucaparib; again all BRCA, all PARP naïve. And the progression-free survival for both of these studies was positive, as you would expect.

Neither of these studies was powered for overall survival, and I’ll talk about some of the questions we have about this, but they were made to present them, and there appeared to be a detriment. And you can see that here, this is the ITT population for ARIEL4. I’m going to show you another slide on this in a moment, but the blue lines’ the rucaparib, and the gray line is chemotherapy. So that confidence interval includes 1, but the hazard ratio is on the wrong side of 1, so 30% worse overall survival than — with PARP than with chemo. And then in SOLO-3 we saw the same thing, very consistent, 1.3 both of them.

And so we had to update the guidelines, and you can see I edited it just for space, but you should pull the guidelines, or wait, because we’re going to have to update them again, so maybe wait. But really we had to say that we don’t recommend using PARP as treatment instead of chemotherapy in the recurrent setting for BRCA-mutated tumors or even HRD per the QUADRA, that was retracted, as well, that we can no longer support that approval, and in fact that approval has been withdrawn.

So this is really no longer something you should be considering except in very rare circumstances, perhaps in a patient that’s still BRCA — PARP wild — PARP naïve, BRCA mutated, and can’t get a platinum, then of course I would use a PARP, but it would be a lot of conversation. But it’s not a planned therapy anymore based on this data, and this was presented in 2022, so very recent data, and the papers are coming out.

Why are we questioning this? Because you might say well that’s just what it is, why are you made about it, Dr Moore. Well, these were never designed to look at OS. So let me show you just with 1 example why we think there’s a problem. This is ARIEL4. I told you it’s platinum sensitive, platinum resistant, partial platinum sensitive, which isn’t really a thing, but it was in the study. Rucaparib versus either weekly paclitaxel if you’re platinum resistant or platinum-based chemo if you’re platinum sensitive, until progression. There was optional crossover for the patients who received chemo to get rucaparib, one-sided crossover, not the other way around, which is important, and I’ll come back to that, and then follow up. So the optional crossover is key here.

So here’s the breakdown for OS across the 3 categories, platinum resistant, platinum sensitive, fully platinum sensitive. Partially platinum sensitive is 6 to 12 months, and then fully platinum sensitive greater than 12 months.

Okay. Here’s the problem. For the intention-to-treat group for ARIEL4 19% of the patients randomized to chemo initially received no further therapy, so once they progressed they got nothing else. They went on hospice or died or something happened; only 19%. But for the patients randomized to rucaparib 42% got not further therapy, nothing, after that rucaparib. That’s very different. Remember, platinum resistant or platinum sensitive, we treat a lot of — we keep going, treat things. So 19 versus 42, that’s — there’s something going on there.

So what happened? Let’s look at it by arm. For platinum resistant, rucaparib, 42% got no further therapy, for the chemo arm only 23% no further therapy. The chemo arm who did get chemo, over 90% of them crossed over to rucaparib because it was supplied. In the partially platinum sensitive 38% versus 16% no further therapy. Of those in the chemotherapy arm who did get something, 96% got rucaparib because it was supplied. And the same thing here in fully platinum sensitive, fully platinum sensitive. Rucaparib 52% — or 48% got no further therapy, platinum sensitive. Why didn’t they get another platinum? What happened here? Versus 15% for chemotherapy, and then of the chemotherapy arm about 63% got rucaparib, and the others got a platinum.

So what happened here is that the study was done ex-US because rucaparib was approved, and you can never randomize it in a place where it’s approved, and it was done in countries that don’t have the same access to therapies as we do here, or maybe they do in France or Spain. And so for the patients that got randomized to rucaparib, which was supplied, they didn’t have access to any other therapy, as opposed to the patients that got randomized to chemo, because in many of these countries the companies are required to supply that chemo, and then in the one-sided crossover to rucaparib they got another line of therapy there. So you have completely unequal treatment post progression that is going to impact your overall survival.

So this isn’t interpretable data, in my opinion, and makes it hard for me to like look at that ITT, and the same things true for SOLO-3 by the way, and be upset about it or not upset about it because I don’t feel like it’s interpretable because these studies weren’t designed to look at OS. So that’s just one reason statistically we wonder what happened here.

DR LOVE: As a result of these and other data rucaparib did not get approved in front-line therapy, so my question is why didn’t olaparib and niraparib get pulled off the market if that’s the case?

DR MOORE: Well, now you may know something I don’t. I don’t think that the FDA has opined yet on the ATHENA approval. I don’t think it was denied, I just don’t think we know yet.

DR LOVE: I thought they said you have to wait a couple — no. I thought — no. Public knowledge. You have to wait. That’s what I’m saying.

DR MOORE: They have to wait.

DR LOVE: Yeah. I’m not saying they turned the approval down.

DR MOORE: Right.

DR LOVE: But they want to see more data, and they don’t want to see more data with olaparib and niraparib.

DR MOORE: They want to see more data.

DR LOVE: So just — I don’t know.

DR MOORE: Well, I would say that you’re right. So they want — and this is going to be an ongoing requirement by the FDA. I think we’ll see that with other studies. There’s 3 other front-line studies, DUO-O, FIRST, and KEYLYNK, who’ve all finished accrual and are maturing, and we expected readouts this year at some point and say they’re positive PFS. We’re all expecting that none of that will unfortunately translate into an approval because the FDA is going to require some percentage of events for OS before they are comfortable with an approval.

A lot of times we’ll come in with a PFS, and maybe there’s 20% maturity for OS, and you kind of just wait for it to read out. That’s been my experience as an investigator for a long time. I do not think that is going to hold any longer, and so unfortunately, I mean this is just my opinion, this is all my opinion, I think that rucaparib just came in late and are suffering because of a change in assessment.

DR LOVE: Right. Right.

DR MOORE: So right now we’re going to have to wait for more OS to come out. Now I don’t think, and again, I don’t know a lot of insider things, I don’t — I would be surprised if they are not looking at and waiting for the OS from PRIMA. They have the OS from olaparib, and that looks okay, and they have the OS now from PAOLA, and we’ll see what they think of that. I think it looks pretty convincing, and it is analyzed consistent with the statistical analysis plan that was put forth at the beginning of the study.

So I think, in my opinion, olaparib’s probably safe in the front line, but we haven’t seen the PRIMA OS data yet. In that, I bet they’re going to look at, and the challenge there that makes me a little worried is that that population was very clinically high risk, Stage IV and neoadjuvant, partial responses. They were very — they were women with very advanced disease, very different than the other studies, so the OS may look different. Hopefully it looks different in the same way in both arms, but we’re all a little bit on pins and needles to see that result. I don’t think they’re off the hook, unfortunately.

So this is one of the several statistical reasons that we question the analysis of OS for some of these studies that were designed really long ago. But the flipside to that is could it be true. That’s the other thing we want to know. If the statistics are true, and I do have a problem with the crossover, but could it be true? Because I don’t want to harm patients any more than any of you do. And there is some biologic plausibility here. I think we just have to acknowledge that.

And this was also presented last year as one of these things that evolves and we’re thinking about, but this is one hypothesis for why this could be true, especially amongst BRCA-associated cancers. This is data from SOLO-3, which they wisely looked at reversion mutations.

Again, these were patients who’d received at least 2 lines of platinum when they came on SOLO-3, but many of them had — like almost 40% had 4 or more lines of therapy. So you would consider them heavily pretreated, often with platinums, and still platinum sensitive. So they said well how many of them come on with a reversion mutation and then how many develop it during exposure to either chemotherapy or PARP.

And so this is interesting and a little bit scary for us in that both arms had about 4% — 3-4% reversion mutations at the time they started SOLO-3, so both on the olaparib arm and the chemo arm, very evenly distributed kind of by accident, but fortunate, that they saw fewer reversion mutations. So these aren’t super common, but on olaparib you saw develop 18% reversion mutations, and no additional reversion mutations were developed on the chemotherapy arm.

I will remind you they couldn’t use platinum though, so it was weekly paclitaxel, PLD, or topotecan I believe, which I guess could cause a reversion but probably not as much as platinum. So there was some difference there, but 18%. And if you develop a reversion mutation you’re not going to respond to PARP, and there is some overlap with response to platinum. So that could impact long term how you respond to things, potentially, and we’re following this and learning this as we go. But that’s plausible that that could impact overall survival.

So I think right now the message would be no longer recommended to use PARP for treatment, which honestly we shouldn’t be. It’s all front line in biomarker-positive patients. And reversion mutations happen either on or off PARP. PARPs aren’t entirely responsible, but they do appear to play a role. And how we track for these reversions and what we do about them is ongoing intensive research by fortunately brilliant scientists, so we will have more data on that coming forward.

The second line of follow up of data that came out in 2022, I talked to you about the monotherapy. All of those were monotherapy use of PARP versus chemotherapy, but what about combinations? There’s so much work being done with combinations, either to make PARP inhibitors work better in tumors that are inherently homologous recombination proficient or use them again; in a tumor that’s become resistant to PARP can you overcome that.

And so there was some data presented, and this is MEDIOLA. This is probably the one we’re most excited about, and I’ll tell you why in a moment. But this is Dr Susana Banerjee’s work. MEDIOLA is platinum sensitive, BRCA wild type, PARP naïve, and they took — it’s randomized, and they looked at a triplet of PARP inhibitor, immune checkpoint inhibitor, bevacizumab, so olaparib/durvalumab/bevacizumab, versus olaparib/durvalumab, so basically PARP/IO without bev. And it’s platinum sensitive. You can see the demographics here on the right-hand side. It’s a very kind of standard population for a platinum-sensitive study. It’s small, 30 patients per arm, so keep that in mind.

But these are the results. And this is not new, but I’m just reminding you. This was presented a bit ago. These are the response rates, so the waterfall plots. You can see the triplet on the left and the doublet on the right, and I would want my patients on the left, on that triplet arm, where only 2 patients didn’t benefit. Everybody else had some tumor reduction if not a response. And even more impressively, if you look at the chart on the bottom, LOH positive is the homologous recombination deficiency, 100% response there, which you kind of would expect.

But in the group that’s homologous recombination test negative 75% response. That’s compared to 17% with PARP/immune checkpoint inhibitor, which is what we’ve seen. I’ll remind you TOPACIO, OPAL, every other study we’ve done is 17%. In this platinum sensitive it’s still 17%.

So is there some magic with the triplet? That remains to be seen. But Dr Banerjee updated the data at ESMO just a few months ago, and this is the median overall survival. Again, we’re really starting to focus more on overall survival, at 56 weeks. Triplet on the left, doublet on the right, and you can see the 2 — and these are recurrent, so platinum sensitive, sometimes several lines of platinum. At 24 months you had 65% of patients alive post study versus 50% with the doublet.

This is a little bit hard to interpret because it’s small numbers. You expect — that’s about what you expect in a platinum-sensitive setting, that’s on average about half of patients will be alive at 2 years, although we hope it’s longer. So I think the thing here is that it doesn’t look like we harmed patients with this. It’s not incredibly short, but without a control arm it’s hard to know how to interpret this. But this is some updated data.

But just the reason we’re excited about this signal, especially here, as I mentioned earlier in our discussion, Dr Love, that we have 3 — 4 studies that are maturing in the front line, one that’s I mentioned, rucaparib versus rucaparib/nivolumab, that’s ATHENA combo. The other 3 are triplet. So first DUO-O and KEYLYNK. They’re all a little bit different. Some required — one requires bevacizumab, the other 2 allowed it, so it’s about half the patients had bevacizumab/immune checkpoint inhibitor/PARP.

So we’re going to have 3 huge studies reading out with this triplet in the front line, which immune checkpoint inhibitors haven’t worked yet for us, so this is like the final push. So if any one of these reads out positive, which I hope they do because in full disclosure I’m the US PI for FIRST, it will bring immune checkpoint inhibitors into the foray for ovarian cancer. And if they work in homologous recombination test negative, per our conversation about potentially are we going to be reassessing current approvals outside of biomarker positive, this could really throw a huge wrench into what we’re using in biomarker-selected populations.

So more to come on that. I think we’ll see things read out in 2023 so when we do our year in review in 2024 we’ll be able to have this conversation again.

So I mentioned prior PARP inhibitors, and we did have an update on the OReO study in 2023. Just to remind you, OReO was presented in 2021. This is Dr Pujade-Lauraine’s work, which asked the simple question can you use a PARP for maintenance a second time, and will it work, which is a very important question. This study was very well done. It unfortunately does not answer the question we need it to now, because I’m showing you this.

If you look at the patients who enrolled. I mean look at this, like 50% or more had 4 or more lines of chemo. So they clearly got PARP inhibitor in the platinum-sensitive recurrent setting, those were the first approvals, that makes sense, and progressed on a PARP, as opposed to starting it front line and getting it for 2 years and then maybe 2 years later they recurred and got it again. There’s different ways we look at PARP duration now. So this is really a group of tumors that had been exposed to PARP until they grew, heavily pretreated.

And I like to say it’s the most negative positive study ever because these are positive. The hazard ratios, these are nonproportional, so even though the medians are terrible, and they don’t differ. The area under the curve is significantly different. The hazard ratios are quite positive because it’s not proportional. There is a subset who benefit, and you see them in the blue. But the majority of patients, in either arm, they just progressed the moment you stopped their platinum. And so this is, again, another call to action that we just need better, newer agents rather than reusing things.

But the table at the bottom is the updated data that was presented at ASCO 2022, where they just looked at OReO and said well, can we tell who would benefit from a second PARP? Is there anything we can glean from this to guide use? And really they just identified things that we all kind of all know are prognostic factors. And I’m not belittling this, it just didn’t show us anything landmark, that if your CA-125’s markedly elevated, or you have visceral disease, like liver metastases. And then patients who had received prior bevacizumab did not do as well, and I think we all know that. So it unfortunately doesn’t tell us the group of patients for whom repeat PARP should be used.

So this is a can do. You can do this. I’ve done this. I don’t think it’s wrong, but really I want new things to use here so I’m not reusing old drugs. And so to that point we’ll talk about antibody-drug conjugate next.

DR LOVE: Just a real quick question, a real basic question, when I see a hazard rate that’s below 1, like 0.43 like you just said, okay, I understand how to apply it. It’s a 57% reduction. How do you use like a positive, like 1.82? What does that mean? Like how — how do you convert that the same way you convert the other way?

DR MOORE: 80%. So if it’s 1.82, which I hope we never see, that’s an 80% increase in the hazard of whatever your endpoint is. So if you’re looking at PFS that means you had an 80% worsening of progression-free survival with whatever you did as compared to your control arm. So it tells you if you’re above 1 and statistically significant, sometimes you’re above 1, but that confidence interval’s huge, like if it’s 1.13, but it goes from 0.7 to 1.2, I don’t know what to make of that. It just depends on where the point estimate falls. But if you’re really over 1, and that confidence interval’s all over 1, that’s a sign of harm.

DR LOVE: Well how to you get 2.85 though?

DR MOORE: Where do you see that?

DR LOVE: Well, it’s part of the standard deviation of visceral disease, 2.85. So like what does that mean?

DR MOORE: Oh. So basically what that means — so this is saying if you have visceral disease this is telling you how much worse you did based on progression-free survival.

DR LOVE: I’m just saying — but what I mean is how do you convert that to a relative risk reduction the way — when you said 80%, like what percent is it with 2.85?

DR MOORE: Oh. I mean visceral disease yes versus no. If you have visceral disease it’s 2.04 as your hazard ratio, so you have 2 times the likelihood, or 200% likelihood of having progression as compared to not having visceral disease.

DR LOVE: Okay. Please continue.

DR MOORE: So take homes from PARP, and then we’re going to move to antibody-drug conjugates, is (1) the overall survival signals, we can’t say final data for SOLO yet, but signals from SOLO-1 and PAOLA really do indicate that first line, here olaparib maintenance, because that’s the data we have thus far, so I’m not trying to be a commercial, that’s just the data we have thus far, maintenance does, I think, enhance the potential for cure and certainly enhances length of progression-free survival.

And HRD testing, I really do feel like it’s going to become more essential because these patients, I think, we really do want to get to PARP.

Now for some patients they missed the opportunity to get PARP in the front line, and they’ve recurred, and they’re very platinum sensitive, and they recurred as second-line platinums, there is some question about use of PARP inhibitors in that setting, as well, and in fact I didn’t have time to go into this, but the approval for niraparib in the platinum-sensitive recurrent maintenance setting was retracted for BRCA wild type because of overall survival data that’s even more confounded than the ATHENA 4 data.

So that’s a challenge. It’s still approved in olaparib and right now still approved for rucaparib, and in my opinion in a patient who’s biomarker positive this is what I’m using until — and I don’t want to harm people. But in that case I really do think the overall survival data is flawed. So in a PARP naïve, platinum sensitive, biomarker-positive patient I’m using a PARP for now. For BRCA wild type I think we really need to develop some additional agents. I’m going to talk about those next.

PARP for late-line treatment, not maintenance, treatment instead of chemo, we are no longer recommending, and I think I’ve covered that pretty well before.

So antibody-drug conjugates are really what’s coming. I keep saying well we need new things. Let’s not reuse everything. Let’s use new things, develop new drugs. Antibody-drug conjugates are answering the call. We have 2, one that has FDA approval. So this is new.

This is mirvetuximab soravtansine. This is an antibody-drug conjugate targeting the transmembrane protein folate receptor alpha, and it is conjugated with a cleavable linker, DM4, which is a maytansinoid, so a microtubule toxin. Lots of data, lots of work done with mirvetuximab over the years culminating in 2 studies, one of which read out in 2022.

This is SORAYA, led by my very good friend Dr Ursula Matulonis, and Rob Coleman. This is a single-arm Phase III, I bet you hadn’t heard about that, of mirvetuximab in platinum-resistant ovarian cancer, 1 to 3 priors, prior bevacizumab was mandatory because this was an accelerated approval intent study.

And you can see the results below. The overall response rate in this population was 32%, expected benchmarks around 15%. And then we broke it down by subgroups into number of lines and prior exposure to PARP, and really everything is very consistent, and this medicine works irrespective of prior lines of therapy.

And then this is the duration of response. You can see the progression-free survival curve, but really duration of response is what I like to look at. It’s almost 7 months, which in a platinum-resistant setting is quite nice, and what we were looking for. This was done only in folate receptor alpha high, I should emphasize. It did lead to an accelerated approval in November of 2022.

These are the treatment-related adverse events. This isn’t new. We’ve published extensively on mirvetuximab, so I think the audience has probably heard about this ad nauseum, but just to reiterate, it’s incredibly well tolerated, no alopecia, far less neuropathy than weekly paclitaxel, almost negligible to no hematologic toxicity. But we do see some low-grade diarrhea, so you have to be prepared for that.

And then about 50% of patients have Grade 1 to 2 ocular toxicity, which is keratopathies predominantly. And the mitigation plan is well spelled out in the package insert, every-other-cycle ophthalmologic or optometrist exams are required for the first 8 cycles just to insure safety, and that is also in the package insert. But these ocular toxicities are very well mitigated by the strategies, lubricating eyedrops and steroid eyedrops, as has been outlined.

And the confirmatory trial is MIRASOL, which we expect to read out in 2023, so another thing to talk about at our 2024 year in review.

Dr David O’Malley presented at ESMO an update on FORWARD II, that’s mirvetuximab and bevacizumab. This is quite exciting data. I want this paper to be published so that we can hopefully get it on NCCN. You can see the response rate in the overall population is 44%, quite high. Folate receptor alpha, they allowed medium and high, but it works in both. But more interestingly is that this program allowed platinum sensitive or platinum resistant. Sometimes patients with platinum-sensitive disease can’t get platinum again. They have allergies or other things, or hematologic challenge.

And you can use this medicine safely, and you can see this here, platinum-free interval, platinum resistant on the left, platinum resistant on the right — I’m sorry, I had them flipped. But the response rates are quite high. So this is a very active combination option to use maybe instead of weekly paclitaxel/bevacizumab, and you kind of lose the hair loss and some of the neuropathy. So I’m very excited about this data that Dr O’Malley presented and how we move it forward.

DR LOVE: What’s a Phase III single-arm study?

DR MOORE: So this is interesting. It’s a Phase III study because it was done with registration intent, and it’s how the statistics were designed. So this wasn’t a signal-finding study like most accelerated approvals are kind of signal-finding and it looks good so they give you an approval and do the confirmatory trial. Like this was designed, power analysis based on all of the prior data to really be a definitive study to look at ORR and duration of response. And so it got a Phase III label. It doesn’t change anything, but just the power and the statistics around it are much more precise than a Phase II.

DR LOVE: That’s interesting. I’ve never heard of that.

DR MOORE: For the nuance that people didn’t pick up. I know, none of us had. We thought it was a typo, but it’s actually real.

DR LOVE: So does that mean that it’s not going to be an accelerated — it is an accelerated approval, right?

DR MOORE: It is an accelerated approval still, yeah. Yeah. It just was a much more definitive power analysis to come in with. We’ve been working on mirvetuximab for a while and FORWARD I was a disappointment, so it was designed to really be a definitive answer for accelerated approval.

MIRASOL will be the ultimate confirmatory trial, and we’re waiting on that.

DR LOVE: Please continue.

DR MOORE: The other rising antibody-drug conjugate is upifitamab rilsodotin, which we call UpRi because it’s hard to say. And this is targeting the sodium-gated phosphate transporter, which actually we’ve tried to target this before with lifastuzumab for those of you that remember that. That looked quite good but didn’t quite kind of make it out of the gates. This is a different molecule. This is not lifastuzumab. Same target but a very different structure, very different scaffold you can see on this antibody-drug conjugate.

So the drug/antibody ratio is high, much higher than other antibody-drug conjugates, and this scaffold allows the protein to fold normally and not get stuck in the kidney and lead to the toxicities that we see with other high DAR ADCs. So that’s novel about it, and it’s also novel in that it has this payload that is — it’s a microtubule toxin but it’s this DolaLock mechanism that controls the bystander effect. So you see some exit of the toxin into the surrounding tumor cells to kill them, which we want for bystander, but with a DAR of 10 or higher you don’t want all of that getting out. And so it locks some of it into the cell, which prevents some bystander but also locks it into the cell to hopefully kill the cell a little bit better.

So this is an interesting molecule. It’s a very different structure than a lot of the other ADCs that we have seen, and it’s cruising through development. So this was data that was presented at SGO by Dr Debra Richardson, who happens to be my partner. This is the waterfall plot. And this is what you like to see as a clinician, the majority of patients are getting some tumor shrinkage.

So you can see the partial response bar there, and it’s a pretty nice percentage of patients who are responding, but 67% of some shrinkage, which is what you want to see clinically when you’re using something in the recurrent setting. This is clinical benefit. So I really like this waterfall plot.

And that has led to — so that has led to a study called UPLIFT, which is the accelerated approval Phase II of upifitamab in platinum-resistant ovarian cancer. It’s very similar to SORAYA, with an overall response rate endpoint. That study has completed accrual and is maturing, and I hope that we will see it this year. So that is exciting.

The confirmatory trial for that is this one. This is UP-NEXT, GOG-3049. This is actually led by Dr Richardson, platinum sensitive, where we need things. Platinum sensitive; you respond to platinum. Here they are testing for sodium gated transporter expression. In UPLIFT they allowed all comers, here they’re testing for high, which is about two thirds of high-grade serous, it’s not rare, but you have to be high, and then you respond to your platinum and randomized 2:1 to upifitamab or placebo.

And this study is open and accruing right now, and so patients can just be started on whatever chemo that they want to be on — that the provider wants them to be on, and they can send tissue testing to see if they’re positive. And if they’re not you can kind of do what you want, and if they are, then you can talk to them about this exciting clinical trial. So this is open and accruing, and we’re excited about this.

DR LOVE: Could you comment a little bit on tolerability issues with UpRi?

DR MOORE: Early on we did — in the Phase. And this, in all fairness, compared to mirvetuximab, where we spent a lot of time kind of really working out the mitigation strategies. Upifitamab developed pretty quickly, and we went up pretty high on the dose, I think in retrospect a little too quickly. I was involved in the Phase I. And so we did run into some toxicity issues with tolerability, a little bit of pneumonitis at high doses, which we’re not seeing at our current recommended Phase II dosing, so I think we have the dose right now, and some nausea/vomiting and fatigue at high dose levels. So we were at 42, which was just too high.

So the maintenance study is at 30, because you responded. You don’t need to like crush somebody with high doses. We were trying to keep — get tumor under — to respond. You just need to maintain it, right? So we’re using a dose that’s very well tolerated but very active too. This has a pretty wide therapeutic window. So in the maintenance study we’re using a dose that’s effective but not the same dose you would necessarily use in a platinum-resistant setting where you’re trying to get something to shrink to help someone’s symptoms improve. Here we’re just using it truly as a maintenance. I want to maintain the benefit of the platinum.

So I really feel like with the amount of work that has been done by the study team, with mitigation strategies, and also dose optimization, which they took very seriously, we have a tolerable dose for both settings. But it’s actually going to probably be an individualized dose selection based on what clinically you need to achieve. If you want to maintain something start low. If I’ve got to shrink someone’s liver metastases or their ascites because they’re really struggling I’m going to start higher, and I’m going to have the data to justify that.                  

DR LOVE: In the webinar I’m going to ask you to elaborate a little bit more on what you were saying about the mechanism of this ADC because I’ve never heard that. That’s really interesting.

But broadly, I mean we talk about ADCs all the time in oncology, like many different cancers, and some of them have specific issues like ophthalmic issues, not just mirv. Belantamab in myeloma, and for that matter tisotumab, right, has eye issues.

DR MOORE: Yup.

DR LOVE: But there’s another thing, too, and the most I’ve heard this has been with T-DXd anti-HER, which has a big bystander effect also, where they talk about chemo side effects like nausea/vomiting, and even alopecia. Is that kind of what you’re seeing here with UpRi or what you saw when you had the higher dose?

DR MOORE: A little bit. We’re not seeing, fortunately, the alopecia, but we were seeing the nausea/vomiting and fatigue. And we saw that, quite frankly, with mirvetuximab as well. When we ran FORWARD I — in the Phase I setting, for whatever reason, we just didn’t pick up on the kind of infusion-related nausea, and so we came into FORWARD I without specified premedications. And we actually saw pretty decent levels of nausea associated with the infusion, so that in SORAYA and MIRASOL we’re premedicating now, just like for chemotherapy, and it mitigates it. It’s done.

So I think it’s that same thing that we’re seeing with upifitamab, as well, is that you just need appropriate premedication. We try to use — you don’t want to throw a million things at patients, make them sleepy, with the diphenhydramine and the steroids. Patients hate the steroids, and we don’t need it here, fortunately, but we try to use less. But for this drug we just may need to premedicate, like we do most things, with antiemetics so that patients can stay comfortable and have good quality of life and gain the benefit of the medication. So I think it is very similar, and the more we use them the smarter we’re getting about mitigation and quality of life.

DR LOVE: Okay, please continue.

DR MOORE: So really I think the conclusion there is antibody-drug conjugates are — there’s a lot of things that are in development right now that we’re excited about that are coming in Phase I. But things that are in Phase II and Phase III, other than PARP inhibitors, are really these antibody-drug conjugates. And for ovarian cancer these are very promising. We have a lot of tumor-associated antigens on ovarian cancer, folate receptor alpha, sodium-gated phosphate transporter.

These are our 2 kind of frontrunners but coming up close behind we have cadherin 6. There’s Erika Hamilton. Dr Erika Hamilton presented data on a novel agent, DS-6000 at ASCO that looks really quite intriguing. HER2 targeting. Who would’ve thought about that in ovarian cancer. It’s not widespread, but again we’re looking at — you’re looking for the needle in the haystack, so if you find HER2 low, just like breast cancer, we’re going to see some data coming out on trastuzumab deruxtecan soon from the DESTINY-PanTumor that has an ovarian cohort. That may move forward, and so that’s a totally different payload and a totally different target.

TROP2, ALPP, ALPP2, there’s a number of these antibody-drug conjugates that are in development, different targets, different payloads, and patients may end up eligible for several of them with very differentiated safety profiles from traditional chemotherapy that gives us more power to individualize the right therapy for the right patient at the right time.

So ADCs, I think, are really on the brink of transforming things for us in ovarian cancer and probably in endometrial, as well, as Dr Tewari may mention, and probably will be a big focus of our next year in review. And I think then people should just be excited about the quality of science that’s coming behind that. We have engineered T cells, cellular therapies entering ovarian cancer treatment. Again, talk about a niche population and matching the right patient to the right treatment, but this would be really transformative for our patients too. Early on, but we’re really pushing the envelope to try and get away from empiric treatment and really more towards individualizing what each tumor needs, and we’re finally making some headway there for ovarian cancer. So it’s pretty exciting.

Endometrial and Cervical Cancer — Krishnansu S Tewari, MD

DR TEWARI: I’m really excited to present this Year in Review, the GYN Oncology Edition, and I understand Katie Moore will be also part of this. I’m going to focus on endometrial cancer and cervical cancer today.

I’m a Professor at UC, Irvine. I direct the Division of Gynecologic Oncology there, and I hold the Philip J DiSaia Endowed Chair in Gynecologic Oncology. I’ve been at UCI now, on faculty, for 16 years.

So we’re going to start off talking about endometrial cancer. We have a whole set of very interesting abstracts and papers that have been presented and/or published in the past year. And I think you’ll find many of these very insightful.

The first one we want to talk about is the KEYNOTE-158 study. Now this is an update that was presented at ESMO last year. The abstract itself was published in The Annals of Oncology. And if you remember, KEYNOTE-158 is the study that looked at colorectal cancer and non-colorectal cancers that had microsatellite instability high or defective mismatch repair genes and the use of pembrolizumab in those patients.

And in this update, you can see amongst the 68 women that had endometrial cancer in this trial, the objective response rate was nearly 49%. You can see stable disease, another 19%. And the duration of response, that has not yet been reached. Median PFS was 13 months. And over a third of the patients — or exactly a third of the patients had not progressed in 3 years or more of follow-up. Overall survival at 3 years and greater, was 62%. Really exciting update.

And the authors concluded that in this Phase II KEYNOTE-158, the update as stated, we have a median duration of response with the update of 47.5 months. And toxicology was manageable amongst these patients that had non-colorectal cancer MSI-high or defective mismatch repair.

Jumping into our next study, this is the GARNET study, and this is the study — it’s a Phase II — looking at another checkpoint inhibitor, this time dostarlimab, not pembrolizumab. And, again, this is in the mismatch repair deficient population of women with endometrial cancer, advanced or recurrent disease, or microsatellite instability high.

In this GARNET study, you can see that if you looked at the mismatch repair deficient MSI-high, 45.5% objective response. And then if you look at the MSI-stable or mismatch repair proficient, there still was activity, 15% activity amongst these patients. And just extrapolating to cervix, that activity of a checkpoint inhibitor in a Phase II study was enough to get pembrolizumab approved for second-line in recurrent cervix.

So, it was 15%, although lower in the mismatch repair proficient patients, is not altogether unexpected, but may be clinically meaningful.

You can see there were some complete responses, particularly in the MSI-high patients, in that middle column.

In summary, dostarlimab in this Phase II, the GARNET study, demonstrated durable anti-tumor activity in both subgroups of patients, the mismatch repair deficient and the proficient patients. You can see that the median follow-up time was 27 to 33 months, depending on which of the 2 cohorts we’re looking at. And the probability of remaining in response in the MSI-high patients was 83% at 2 years of follow-up.

The authors for this update were also interested in pointing out that the q6wk dosing in endometrial cancer was a first to be studied with a checkpoint inhibitor. And of course, that inhibitor was dostarlimab.

Once again, safety was very manageable.

Now let’s talk about a Phase III randomized trial, this time using a non-chemotherapy double of pembrolizumab to lenvatinib.

There was a study, prior to the KEYNOTE-775 study, that led to accelerated approval of this combination for the mismatch repair proficient patients. So, as you know, pembrolizumab, as a single agent, had been approved by the FDA under the accelerated approval program as an agnostic indication for MSI-high, but the combination of pembrolizumab plus the oral tyrosine kinase inhibitor lenvatinib, was approved under the accelerated program from Study 100 for the MSI-stable. And this paper, or this study, KEYNOTE-775, was the confirmatory, randomized Phase III trial.

And you can see that both the PFS curves and the overall survival curves, on the left and right, for the mismatch repair proficient patients, seen on the top. And then for the intention-to-treat population, seen at the bottom, was statistically significant and we believe clinically meaningful. This converted the accelerated approval of the combination to full approval for patients that have MSI-stable disease.

The authors concluded that one other attractiveness of this combination was, it was a non-chemotherapy doublet, and it led to significantly longer PFS and OS than chemotherapy, the physician’s choice chemotherapy, that was provided among patients with advanced disease.

So this was a serious game-changer in endometrial cancer. Non-chemotherapy doublet, an option for patients with MSI-stable disease.

DR LOVE: I was just kind of curious, has lenvatinib/pembro been looked at in MSI-high disease, for example, compared to pembrolizumab alone or IO alone?

DR TEWARI: Yeah. In about 9 patients in the original Phase II study, they were MSI-high, and there was activity there. But whether it’s been studied — the studies have not been powered enough to determine whether the combo in MSI-high is superior to just pembro alone in the MSI-high patients.

DR LOVE: I mean the response rate is great and duration of response is great with MSI-high, but not everybody responds. What about anti-CTLA4 plus PD-1? Has that been looked at in MSI-high endometrial?

DR TEWARI: Not yet, but that’s a — we’re very interested in that in cervical cancer. We’ve studied nivolumab and ipilimumab. And, as you know, in other fields like melanoma and in lung cancer, that combination has been supported. We haven’t studied that yet in endometrial cancer, but that would be interesting to pursue.

DR LOVE: Just out of curiosity, if somehow you could do a randomized study in MSI-hi — I know that’d be hard to do — looking at IO or pembro alone versus pembrolizumab/lenvatinib, what do you think you would see?

DR TEWARI: So, the real-world experience with lenvatinib, it’s very difficult to tolerate. So I think we’re going to see a lot of dose reductions, maybe people dropping the lenvatinib, especially if they’re MSI-high. So I’m not sure that study would get done.

If we had a different thing to add to it, and CTLA-4 inhibitors are also — they do have their associated toxicity, but if there’s another checkpoint, maybe a LAG3 or a TIM3 or maybe even a TIGIT, maybe that may be the way moving forward. Because as you correctly point out, pembro alone isn’t enough for many patients that are MSI-high.

DR LOVE: Okay, please continue.

DR TEWARI: Thank you for that question.

So, this was a mini oral session. This is an updated efficacy and safety of the non-chemotherapy option of lenvatinib plus pembrolizumab versus treatment — physician’s choice chemotherapy. And this is an update of the KEYNOTE-775, and this was a mini oral that was presented last year at the ESMO Congress.

And you can see here median PFS, 6.7 months for the MSI-proficient patients — I mean the MSI-stable/MMR proficient patients. The hazard ratio of .6. The median overall survival of 18 months. So with extended follow-up, this combination had durable clinical endpoints that were meaningful. And even though, as I did mention in the Q&A just now, that combination is associated with quite a few dose reductions. With lenvatinib, the starting dose being 20 mg daily, and oftentimes patients go down to 14 or 10. Still, even with these dose reductions, with the continued follow-up, the data from KEYNOTE-775, the updated data presented at ESMO is consistent with the primary analysis.

RUBY. This is a clinical trial. This is a Phase III, randomized trial for patients with measurable disease with newly diagnosed endometrial cancer that have measurable disease after surgery. This was studying whether adding a checkpoint inhibitor to primary systemic chemotherapy versus placebo — or chemotherapy alone plus placebo would be efficacious.

I don’t have all the data for this yet because all we have at this point is a press release that was dated December 2^nd, that the clinical trial met its primary endpoint at the planned data interim analysis. And so, this is pretty exciting because once this is presented and ultimately I would assume, reviewed by the US FDA, this may lead to a new indication for newly diagnosed advanced or recurrent endometrial cancer.

All we have right now is that it met its primary endpoint. And we’ll have to see whether it gets presented in the coming months, and maybe perhaps at the Society of Gynecologic Oncology in March or at ASCO in May.

DR LOVE: Okay, so this is going to be really practice-changing if it’s impressive. First-line therapy in endometrial. I guess, sort of similar to lung cancer — chemo plus IO.

DR TEWARI: Yep, exactly. And what’s also interesting, the NRG has their own doppelganger version of this study with pembrolizumab, chemotherapy plus pembrolizumab versus chemotherapy plus placebo. And that study is maturing right now. So I’m not sure when that will be ready.

DR LOVE: Thinking about lung cancer and other cancers, too, another thing I guess will be interesting, not only to see the MS status, but also PD-1 levels. At this point, I haven’t really heard very much about PD-1 levels in endometrial cancer. Is there a correlation of benefit?

DR TEWARI: The PD-1 levels is really, in our world, in the GYN world, is really in cervix cancer where they look at the CPS score less than 1 versus greater than 1 or greater than 10. And we haven’t looked at the levels in endometrial cancer because MSI, microsatellite instability, is such an important driver or predictor of response in this field. But I’m sure, especially if we get a first-line indication for checkpoint with chemotherapy, it’s going to be important to dive in and try to look at PD-1 levels.

DR LOVE: That’ll be really interesting. Okay, please continue.

DR TEWARI: So here’s another important paper. Now we’re just switching gears and we’re looking at a very novel oral agent, selinexor. And what it does is, it prevents extravasation or nuclear export of wild-type p53. So p53, as we know, is a tumor suppressor gene protein and it has many, many functions. And this is a study, it’s a prospective, double-blind, randomized Phase III study, the SIENDO study, of oral selinexor versus placebo, as maintenance therapy after first-line chemo for advanced and recurrent endometrial cancer. So a very unique medicine.

And the results are really interesting. If you look at the ITT population, you can see there is some separation in the curves, the PFS curves. But the real interest is in the right side of this graph, when we look at those patients that had a wild-type p53, you see a striking separation of the median PFS curves. I think that’s 13.7 months versus 3.7 months. Remarkable. And it follows along and harmonizes with the mechanism of action of this oral agent, in that it prevents the nuclear export of wild-type p53, so p53 can do what it needs to do.

So you can see here, there was a 30% reduction in the risk of progression and/or death in this p53-wild-type subgroup; the reduction in the risk of progression or death was 62%.

It appears that endometroid histology benefited the most. And the adverse events were all manageable with either supportive care, with or without dose reductions.

So pretty exciting study that was presented.

The subgroups were looked at in this same study, the SIENDO study. And by subgroups, we’re looking — we mean to talk about the Cancer Genome Atlas for endometrial cancer. Because from the Cancer Genome Atlas for endometrial cancer, we learned that there are 4 very important molecular subgroups, the POLE hypermutated patients, the MSI-high, the copy number-low and the copy number-high. And they each are associated with different prognostic subsets.

And here you can see the different subgroups in this data that was — when the SIENDO study was re-analyzed. And if I just switch over to the summary, the benefit of selinexor was, of course in the wild-type p53, as we saw in the original presentation, but also in the copy number-low endometrial cancers. Didn’t seem to be any benefit when you looked at the aberrant p53 proteins or in patients that had copy number-high or POLE hypermutated patients. But in the MSI-high, MSI-stable, wild-type p53 and copy number-low endometrial cancer, there appears to be a signal there.

So, this is a new medicine. A novel mechanism of action. We’ve heard about p53 for the past 40, 50 years, and how important it is in oncology. Now we’ve finally got a medicine that is associated with the prevention of nuclear export of the drug — of the protein.

DR LOVE: What fraction of patients are p53 wild-type? And would you, based on what you’ve seen here, would you like to be able to use this drug? And when? Third line?

DR TEWARI: Yeah. I think third-line would be reasonable, maybe even second-line. Because it has a more favorable toxicology profile than lenvatinib. But there’s no head-to-head comparisons.

As far as what percentage of patients have mutated versus wild-type p53? It’s difficult to determine with certainty because if you’re able to measure p53, that often means it’s aberrantly expressed because the wild-type molecule is unstable. And so it’s really a function of very sophisticated IHC, and even genetic sequencing, because oftentimes when you measure p53, you’re measuring the mutant. But we know, like, for example, in ovarian cancer, serous ovarian cancer, upwards of 70% are p53-mutants. In endometrial cancer, it’s much less, especially in the endometroid histology.

DR LOVE: In terms of tolerability, of course we’ve heard about selinexor, particularly in multiple myeloma. And there, we heard about a lot of toxicity. And it seemed like when they decreased the frequency and the dose, they saw a lot less. But they have struggled with it. I’m curious whether you’ve used it? And if you have any sort of qualitative impression about how it’s going to be using it in these patients?

DR TEWARI: We did not participate on SIENDO. I’ve not used it yet. But I’m looking forward to gaining experience with it. Because I will definitely use it if it’s made available for our patients, especially with the toxicity being very manageable with supportive care.

DR LOVE: That’s what I want to find out from you, if that is the reality. Because, like I said, it seems like it has been a challenge in myeloma. Although, again, the myeloma investigators, I think they kind of have a way to do it and we’re trying to get that out to the community. Premedications.

DR TEWARI: We can definitely learn from them.

DR LOVE: For sure.

DR TEWARI: We can definitely learn from them. I mean there’s many examples in clinical trials, like with niraparib in ovary, the dose was 300 mg and saw terrible platelet problems. And then the FDA changed the dose, and all of a sudden those platelet problems essentially minimized.

So we should learn more from the myeloma folks about dosing.

DR LOVE: Yeah, I don’t know if you know Joseph Mikhael at the City of Hope, but he was on one of our webinars recently. And he was actually speaking very highly about selinexor, but again, they’re very aggressive in his clinic with the nurses and all about premedication and following them. And, again, they use it every week. So it’ll be interesting to see how it plays out in gynecologic cancer.

DR TEWARI: Yeah, I absolutely agree.

DR LOVE: Please continue.

DR TEWARI: Thank you.

So here’s a paper that was presented. This is looking at a very novel combination of medicines for patients with advanced metastatic endometrial cancer. This is the ENDOLA study. It’s combining a PARP inhibitor, olaparib, with metronomic cyclophosphamide and metformin. So, it’s like they threw these 3 things together but there seems to be some scientific rationale for why this combination would be active.

So let’s talk about endometrial cancer in general. Many endometrial cancers have aberrations of the PI3K, AKT, mTOR pathway and also have PTEN mutations. These types of mutations, particularly with the PI3K, lead to problems with homologous recombination. And so that’s a nice entry point for a PARP inhibitor like olaparib, which is active in patients with HRD or homologous recombination deficiency.

There is some invitro data that metronomic cyclophosphamide would be synergistic with olaparib in this type of environment. And then metformin also is active in patients with these PI3K aberrations. And so, the 3 of these medicines together, may be an opportunity for an oral or synergistic triplet regimen for patients with advanced disease.

So, you can see, in the Phase I study, which was dose-seeking, the median age was 69 years. More than half of the patients had had 4 lines of prior therapy, so a very heavily pretreated group. And you can see 10% of the treatment-related adverse events were predominantly hematologic, although there was Grade 3 asthenia and fatigue in 13%. And this was what the recommended Phase II dose was. It turned out to be 300 mg twice daily of olaparib, 50 mg of metronomic cyclophosphamide daily, and then metformin 500 mg 3x a day.

If we jump to the Phase II, using the Phase I established dose, you see the median PFS, 5.1 months. The median PFS among endometroid cancers, 7.5. Among the serous cancers, 4.3. And if you compare this with the patients from KEYNOTE-775, the lenvatinib/pembrolizumab population, it compared favorably. But it should be noted that the KEYNOTE-775 patient was a healthier subgroup — they were younger, and they didn’t have — they weren’t as heavily pretreated.

So it’s not a Phase III randomized study, but it’s a promising triplet regimen, all oral. And we’ll have to see where this goes.

Question.

DR LOVE: I was trying to pull up the paper, but do you know if they saw objective responses?

DR TEWARI: They didn’t report that. They did see objective response — I mean they saw partial responses; they saw stable disease. I didn’t see any CRs.

DR LOVE: Just the fact that they saw an objective response is kind of interesting.

DR TEWARI: Yep. I absolutely agree. The importance of this study for me is that science works. Throwing these 3 things together, I didn’t understand what was the – I mean I know metformin is potentially active in endometrial cancer, but olaparib and metronomic cyclophosphamide?! But there is a scientific rationale to put these together. And it’s a very high-risk population, which is why they were part of the Phase I in the beginning. And so, it’s another promising regimen.

And this is one presentation that I’m going over today that was actually presented at the AACR. So it was nice to have a paper from that.

DR LOVE: Please continue.

DR TEWARI: Now we have an antibody drug conjugate. This is HER2-tareting antibody drug conjugate. I know you’ve heard of this and you’ve probably used it in your practice. The payload is etopoisomerase-1 inhibitor, deruxtecan. And this is being studied in this study, in those HER2-positive uterine and ovarian carcinosarcoma. So rare subgroup of patients that represents a clinically unmet need.

This is the cartoon of the drug. You can see that the actual linker is connected to the payload on the cysteine residues of the HER2-targeting protein.

And you can see in some of these invitro cell line data, when the HER2 — so the top line here is the inner control and you can see there’s no activity in the HER2-positive and HER3-positive ovarian/uterine carcinosarcoma cell lines when there is no drug being used. But when the drug is given, you can see that the cells die.

And then this is a double-control on the right side where the HER2 expression was 0. And it shows the drug doesn’t work as a negative control in the HER2-0 expressed uterine carcinosarcomas. But nice cell survival plots when the HER2 expression is present and verified with the control.

Let’s switch gears and talk about cervical cancer now. Most of these papers that we’re going to present with cervical cancer, I believe they’re all immunotherapy with the exception of an antibody drug conjugate that we will end the presentation on.

So let’s see what we’re going to talk about first. This is the KEYNOTE-826 Phase III randomized trial, front-line randomized trial of chemotherapy with or without bevacizumab, plus pembrolizumab, versus chemotherapy with or without bevacizumab, plus placebo for first-line treatment of recurrent or metastatic cervical cancer.

This was presented 2 years ago. The primary analysis led to US FDA approval of this, and converted the accelerated approval of single-agent pembrolizumab in second-line to full approval.

And this was a subgroup analysis that we presented at ASCO last year. Subgroup analysis of this group. And you can see for the PFS forest plots in the all-comers or intention-to-treat on the left, and the CPS scores, combined positive score, greater than or equal to 1, with different prognostic factors: tumor histology, the choice of platinum drug used by the investigator, the use or non-use of bevacizumab and prior chemoradiation only.

The forest plots, the error bars are all to the left of the ordinate, suggesting that there is activity in these subgroups that were analyzed.

If we jump over to the overall survival forest plots, we see the same thing. For the non-squamous tumors, it’s not as present in the intention-to-treat and the CPS greater than 1 as with cisplatin. Also, the error bars cross the ordinate. But overall, these seem to favor the combination with pembrolizumab. And again, forest plots are hypothesis-generating. They’re not powered for each of these subgroups. But I think the value of this subgroup analysis, it does provide a little bit a real-world experience because not all patients are getting carbo as part of their chemotherapy; some are getting cis; 60% of the patients were getting bevacizumab in this study. Sometimes bevacizumab is not safe to use, which is why when the FDA did approve the regimen, they specified in combination chemotherapy plus pembrolizumab with or without bevacizumab.

So this is a real-world experience embedded in the KEYNOTE-7 — excuse me — KEYNOTE-826 study and presented at ASCO last year.

If we look at the quality-of-life study, this was presented at the SGO last year from the KEYNOTE-826 study, essentially, if you look here, and you can see this is one of the instruments that were used, the EORTC QLQ-C30 global health scale, quality of life measure. And there was a couple of other instruments used. But basically, there was no significant deterioration in quality of life with the addition of pembrolizumab. You can see here the hazard — or the confidence intervals, and right here they even come together. And actually what’s interesting is that in some subgroups, not statistically significant, patients receiving pembrolizumab did better. And that makes sense — your tumor starts shrinking, you’re going to start feeling better.

And here you can see the different physical functioning, symptom experience. Lots of different tools were used to measure quality of life. And in some areas, quality of life was not just only stable with pembrolizumab, but actually it improved somewhat. And every improvement seen by these green bars was essentially greater than the improvement seen with the placebo in many of the cases.

Improved OS and PFS were not accompanied by any substantial deterioration in health-related quality of life. And that’s the bottom line from the study.

We’ve not found the best way, in my opinion, to measure quality of life in patients, in oncology, not necessarily GYN oncology. But I think the more instruments that we have available to measure it, the better. And I’m very happy with this trial because so many different instruments were used.

This is another checkpoint inhibitor, and you’re looking at a Phase III randomized trial of the anti-PD-1 inhibitor, cemiplimab, versus physician’s choice chemotherapy. And this is in the second-line setting. So the KEYNOTE-826 of chemotherapy plus pembrolizumab with or without bevacizumab was in the first-line setting. This is in the second-line setting.

And you can see here that in the intention-to-treat population, there was a significant separation in the overall survival, 12 versus 8 months, with reduction in the hazard of death by 31%.

And then when we look at the squamous cell population alone, again we see a significant separation. And the statistical hierarchy, the squamous cell population was studied first and analyzed first and if it was significant, only then did we go to the intention-to-treat.

Now what was interesting, the adenocarcinoma subgroup was not part of the statistical hierarchy. And so, the study was not powered to study adenocarcinoma, but I show these survival curves just because the most striking benefit with cemiplimab was in this group of patients. But again, it was not part of the statistical hierarchy. And so this is really hypothesis-generating. But 13.3 months OS versus 7 months in the second-line setting. So, pretty remarkable.

Now I have to say that with the FDA approval of chemotherapy plus pembrolizumab in the first-line, that’s made the clinical value of this study a little remote because using checkpoint after checkpoint is not felt to have any true benefit. But this study has received approval for use of cemiplimab in Canada, Brazil, the European Union, all 27 countries of the European Union, and most recently, in Japan.

Tisotumab vedotin plus pembrolizumab in first-line recurrent or metastatic cervical cancer. So the innovaTV 205 study is an important study. We learned 2 years earlier with the accelerated approval of tisotumab vedotin for all-comers in the second-line setting, that tisotumab may be an agent that we want to bring into the front-line setting.

So, what the innovaTV 205 study has done is something very interesting. They’re looking at combining pembrolizumab with first-line tisotumab, and you can see the objective response rates are in excess of 40%, with 15% complete responses. Median duration of response not reached. But in addition to first-line tisotumab plus pembrolizumab, we also have second-line and third-line tisotumab plus pembrolizumab. Here you can see objective responses reduces a little bit, but still is very favorable, 38%. Remember, pembrolizumab by itself is an objective response rate in the second-line of 14%, TV by itself in the second-line is an objective response rate of 23%. So the combination increases it.

And in this study they also looked at TV plus carboplatin in the first -line. So what you can almost envision is the possibility of switching out paclitaxel in front-line with tisotumab. Because the front-line regimen is carboplatin, paclitaxel and pembrolizumab. And in this novel study, and you can see 54.5% objective response with TV plus carbo, and it’s also active when it’s combined with pembro, perhaps moving forward this may provide an opportunity to study tisotumab vedotin in the front-line setting, maybe switching it out with paclitaxel.

So, conclusion. In the front-line TV plus pembro, 41%, median duration of response not reached; 16% complete responses. Second-line and third-line tisotumab vedotin plus pembrolizumab, 38%. And then first-line tisotumab vedotin plus carboplatin, 55%.

So watch this space. I think that there is going to be an opportunity to possibly move tisotumab vedotin up into the front-line, switching it out for paclitaxel.

This is a really important paper. One of the concerns with tisotumab vedotin is the ocular toxicity. And the FDA has made some very specific requirements for patients being treated with this and we’ll go over them. And this paper, published in Gynecologic Oncology, reviews some of the mitigation and management strategies to prevent ocular toxicity.

I think it’s first important to talk about what are the ocular toxicities that we observe with tisotumab vedotin. Most of it is conjunctivitis, dry eyes, keratitis. The vast majority of these patients have Grade 1/Grade 2, but there were 3 Grade 3 ocular toxicities, 1 of which led to a patient requiring a corneal transplant, which is why the FDA has taken this very seriously.

This is the grading of the different ocular toxicities that we see. And in the clinical trial innovaTV 204, which was the study that led to accelerated approval for tisotumab in the second-line, you can see the different types of ocular toxicities that occurred in 100 patients. And as I said, there were only 3 cases of Grade 3, unfortunately 1 of these patients — one of these Grade 3s, did lead to a corneal transplant.

So what is the mitigation strategy? The mitigation strategy is as follows: patients, before every single treatment with tisotumab vedotin, which is given every 3 weeks, need to be seen by an eye specialist. It doesn’t have to be an ophthalmologist; it could be an optometrist, or anyone that can perform a slit lamp examination and a visual acuity test. So they need that. Then patients are advised not to wear contact lenses. And during their infusion, they get 2 types of steroids — they get vasoconstricting steroids on day 1, as well as corticosteroids on day 1.

They’re given cold packs to place on the eye pads of their eyes. And in post-infusion, for another 48 hours, they continue with the steroid eye drops. So vasoconstricting eyedrops day 1, steroid eye drops days 1, 2 and 3. And then for days 4 through 21, before they get retreated, they take lubricating eyedrops.

So it’s a really important mitigation strategy. My own personal experience, I’ve not seen any ocular toxicity other than Grade 1 dry eyes. And patients are able to follow this. We’ve partnered up with our ophthalmologist at UCI, and they’re more than happy to help us by seeing these patients with every meeting — prior to every infusion.

These are the dose modifications that appear in this chart — in this paper, depending on what type of adverse event — ocular adverse event occurs. I want to point out for ulceration, if it’s the first occurrence, withhold the dose until it’s completely resolved and then you can reduce the dose. The doses are 2.0 mg/kg, and then it drops down to 1.3. There’s a total of 4 dose reductions available before you have to discontinue the therapy.