Addressing Current Questions and Emerging Considerations with the Use of PARP Inhibitors in the Management of Ovarian Cancer
Addressing Current Questions and Emerging Considerations with the Use of PARP Inhibitors in the Management of Ovarian Cancer
Faculty presentations from an independent satellite symposium during the Society of Gynecologic Oncology’s 2019 Annual Meeting on Women’s Cancer. Featuring perspectives from Dr Robert L Coleman, Prof Jonathan A Ledermann, Dr Stephanie Lheureux and Dr Kathleen Moore.
1 AMA PRA Category 1 Credit™
To complete the Post-test and Educational Assessment and Credit Form, click here. A statement of CME credit will be issued for scores of 80% or better. Your statement of credit will be mailed to you within 3 weeks or may be printed online.
Genomic Testing and Management of Primary Ovarian Cancer DR MOORE: So we’re just going to talk a little bit about mostly SOLO1. So why is it? We are in an era where there are 5 international Phase III randomized studies enrolling all comers, some with biomarker stratifications, some without, onto studies that at least as a component have a PARP inhibitor. So why is that? Why are we so excited about PARP inhibitors? And the reality is that, at least for high-grade serous and high-grade endometrioid, of course we have our BRCA patients, which you can see denoted on the upper right-hand side of this pie, that are germline and somatic and the methylation. BRCA promotor methylation is important and is a predictive biomarker, comprising about 25% of high-grade serous and high-grade endometrioid. So a fair number of patients we already know are going to respond well to a PARP inhibitor or PARP inhibitor maintenance. And then there’s an additional 10% to 15% that have other important mutations in DNA damage response genes, such as RAD51C, which I’ve mentioned several times. That’s a very important one, both mutations and methylation. We’re still learning about the other ones, PALB2, BRIP1, the whole Fanconi gene pathway. There’s genes that have 0.5% prevalence. So again, unless we put patients on trial, we just don’t know if these are predictive or not. They may mean nothing. But they have the potential to respond, and so we should consider them. And then there’s patients that we don’t really know yet. These PTEN loss, EMSY amplification, we’re still figuring out. But really it’s close to 40%, 45% of patients have some kind of molecular derangement that would potentially predict a response to a DNA damage response agent such as a PARP inhibitor. And then we have, unfortunately, the group with cyclin E amplifications that are really categorizing your platinum-resistant patients. And they don’t respond at all. And then this group that has mismatch repair, nucleotide excision repair mutations and other things that we still are trying to figure out. So I wouldn’t say they’re a no, but they’re probably not as robust as the pie chart on the right. So this is an important area of benefit for our patients. And so you can certainly see kind of across the board, there are currently 4 PARP inhibitors. Actually, there should be a fifth one, because veliparib should be on this. I left it off. Four PARP inhibitors that are approved though — veliparib’s not approved yet. So these are the approved ones, in PARP inhibitors in solid tumors, olaparib, rucaparib and niraparib, with indications in ovary cancer. Talazoparib, of course, with an indication only in breast cancer thus far, but clinical trials in ovary are running. And you can see the various indications, both in the US and internationally, with the red box around the most recent indication for olaparib, which is in front-line maintenance for patients with a BRCA mutation. Based, of course, on SOLO1, which I was incredibly privileged to lead the team effort on this. SOLO1, just to remind you, was a randomized Phase III study. It enrolled patients with Stage III or IV ovarian cancer who harbored a BRCA mutation, either germline or somatic. They had to have excellent performance status. They had to have undergone an attempt at a debulking surgery, either primary or interval. And at the end of induction platinum-based chemotherapy, they had to either be in complete or partial response, and the response was the stratification factor for the study. They were randomized to olaparib twice daily or placebo, and that continued until progression. Or if no progression occurred, assigned treatment was stopped at 2 years, unless they came on the study with a partial response. And at the time they hit 2 years, if the provider felt like they were still getting clinical benefit, and with discussion with the medical monitor or me, they could remain on study. The primary endpoint was investigator-assessed progression-free survival. Secondary endpoints were progression-free survival, as assessed by the blinded independent central radiographic review, progression-free survival 2, which is an important regulatory endpoint in the EU, overall survival, time to first, time to second subsequent therapy, and the main primary quality-of-life endpoint was the FACT-O TOI scale. And then there’s a number of other quality of life that will be presented later this year. This was the main primary endpoint that was presented at ESMO last year, with at least 3 years of follow-up for every patient who is enrolled, and on the median we had about 41 months of follow-up. We demonstrated a hazard ratio of 0.3 in favor of maintenance olaparib. So a 70% reduction in the risk of recurrence. And the landmark analysis we kind of look at as based on the Kaplan-Meier estimates at 3 years, we had 60% of patients on olaparib progression free as compared to only 26% progression free on placebo. And you can see the difference in the morphology of the curves. This isn’t just a shifting of the curve to the right. The complete shape of the curve has changed, and it changes early on. And I think that’s a very important point. We’ve never seen a curve that looks like this in ovarian cancer. This is progression-free survival 2, which is measured from the time of randomization to the time of second progression among those patients who have progressed. And so this maintains statistical significance with a hazard ratio of 0.5. And again, this is an important regulatory endpoint in the EU, because it’s believed to be a surrogate for overall survival. And the fact that this is still positive and very positive is even more remarkable, because, appropriately, among the placebo patients, we had over 35% crossover to a PARP-containing regimen as their next line of therapy, as those regimens became FDA approved and available. So this includes quite substantial crossover for the PFS2 endpoint. This is the safety summary for olaparib and placebo, and in red I’ve just outlined the key things. So we did see treatment-emergent adverse events that required dose reduction, mainly anemia in 28%. But only 11% of these treatment-emergent adverse events required dose discontinuation. And that’s fairly low. So patients were not coming off of olaparib due to adverse events. This is just the demonstration of the adverse event profile. The dark boxes reflex those that hit the Grade 3 or 4 grade. The rest of it is low grade. So you can see only really anemia at 21% hits any kind of Grade 3, and it was really mostly Grade 3 in any appreciable frequency. Otherwise, this looks exactly the same as every other PARP inhibitor in every other setting. We did look for specific adverse events of special interest. Of course, MDS and AML is always something we’re paying attention to. It was 1.2%. Again, very consistent with every other PARP inhibitor trial. We did see new primary malignancies in 2% and 2.3% of patients respectively, mainly BRCA-related malignancies, but we actually had a thyroid cancer and a head and neck cancer. These patients develop cancers, and you have to pay attention. And then pneumonitis is not a new risk to PARP inhibitors. It’s been well described, and it’s about this rate, about 2%. So that’s the data from SOLO1 and led to the approval. And really, I think, brought to the forefront — and I’m happy about this — got the conversation kind of urgently going globally about how we test patients, basically mandatory now. I don’t think you should be taking care of an ovarian cancer patient in the front line unless you have some kind of plan in place to test her for a BRCA mutation, germline at least, and then germline/somatic, in some kind of sequence, ideally. And so this has sort of become a mandatory part of care, and I think our patients are going to increasingly expect this from their providers. So this is the next big challenge, is how we do this. Do you start with germline and then reflex to somatic if negative? Or do you start with somatic and then reflex to genetic counseling if you find something that is relevant? And so the latter test is probably the most cost effective, and that’s true, but the issue is that you miss about 5% of germline mutations if you stop with the somatic test alone, because you miss large rearrangements. There’s differences in call coverage, and there’s differences in categorization of variants. So you miss 5%, which, of course, then you lose the ability to do cascade testing on the relatives. And really, don’t we all want to put ourselves out of business if we can prevent this? I mean, I do, and I know you do, too. So is that acceptable, 5%? I think that’s something ethically we have to discuss as a society, but let’s get closer to 100% of testing of some sort, and then we can kind of work out the nuances. And then as I mentioned, there’s 3 PARP studies that are going to result I think this year. I don’t know for sure, but we expect results from VELIA, PAOLA-1 and PRIMA probably later this year. That may change this entire landscape. PRIMA’s primary endpoint is in homologous recombination deficiency patients with a hierarchical subsequent analysis in the intention-to-treat group. PAOLO-1 and VELIA are both intention-to-treat primary analyses. So if those are positive, everyone’s going to get a PARP, I guess. We’ve got to talk about that. And then there’s 5 ongoing studies that combine PARP with IO with anti-angiogenic currently running, which makes sense if we cure a lot more patients, which I hope we do. So here’s what’s running right now. These are the front-line maintenance trials that are both completed and the ones that are pending. So I mentioned VELIA, PRIMA and PAOLA-1. There you can see them on the right-hand side of your graph, as compared to SOLO1, which is a front-line indication as compared to GOG-0218, of course, resulting in a front-line indication. So you can see the differences in the trials here. And there are significant differences that we’re going to have to work through. And then this is just data on homologous recombination, just kind of speaking to the PRIMA issue. We know that BRCA is a biomarker and a very strong biomarker, but how do we identify the rest of the patients that may benefit? So we have this group that we’ve identified. We’re going to give them olaparib and potentially another PARP, pending on approvals. But then how do we kind of figure out this other group? These other mechanisms of HRD? The methylations? The other DNA damage response pathway mutations? Other ways that we lose homologous recombination. There are all of these assays that are good, but imperfect for a number or reasons, and so how do we really develop the biomarkers that are going to identify the patients that benefit most from PARP inhibition so we can use PARP where it makes sense and spare the patients who are not going to benefit and just going to get toxicity from these agents? So we’re a long way from that. But these are some of the things that are being looked at in terms of candidate biomarkers. I’ve mentioned methylation, looking at BRCA protein — much harder than it sounds. It’s not just a simple IHC looking at specific homologous recombination gene mutations. And then maybe a better way of assessing the HRD score than we currently have may help us get closer to a directed biomarker so that we can really personalize therapy for our patients rather than empirically just giving everyone a PARP inhibitor and seeing what happens. That’s probably a less-great plan. But that may be where we end up at the end of the year. Role of PARP Inhibition in the Management of Relapsed Ovarian Cancer DR LEDERMANN: So this is where it all started. All the work in ovarian cancer has been in the management of relapsed disease and where we’ve learned the most about it and actually still great opportunities for treating patients. Although we’ve heard about the very exciting data that’s emerging and I’m sure will continue in. So we really tested PARP inhibitors most extensively in the field in the manner of maintenance therapy. And this arose really from the emerging data that PARP inhibitors were perhaps active not only in patients with BRCA mutations but also in BRCA wild type, and at that time what we had was platinum sensitivity as ourkey indicator for potential sensitivity to a PARP inhibitor. So the basis of the trials was to look at patients with platinum-sensitive disease. Having responded, they were then randomized to a PARP inhibitor or placebo, and then we looked at progression-free survival. And that was the basis of the SOLO2, the NOVA trial, and the ARIEL3 trial, which were the 3 Phase III trials that brought these 3 drugs into widespread clinical practice. But the background to it was a study called Study 19, which was a randomized Phase II trial, which was the one that showed us first that maintenance olaparib can significantly improve the progression-free survival. Here on the left with all comers, that is BRCA-positive and BRCA wild-type patients. And when we broke that down and looked at the subgroups of patients with a BRCA mutation, and those were BRCA wild type, we see a significant benefit in both these subgroups. So the BRCA-mutated group hazard ratio was 0.18. And we see a significant benefit also in the BRCA wild-type group, which has led to this widespread license now of olaparib in all patients who are platinum sensitive. And Study 19 is still the study that has the maturest data in terms of survival. And there, these are the final OS data of Study 19, showing you the long-term survival. And the curves are interesting insofar as you can see after 36 months that there seems to be a divergence of the curves in favor or olaparib. The hazard ratio is actually 0.73 in favor of olaparib. Although that does not meet statistical significance by the criteria that’s used because of multiple preanalyses. But if we focus on that tail of the curve, the flattening of the curve, what we can see is that 15% of the patients in Study 19 continue to receive olaparib for 6 or more years without evidence of any disease recurrence or progression. And that, I think, is unprecedented in the management of recurrent ovarian cancer. So who are these 15 patients? Nine of them have a BRCA mutation, either germline or somatic, but 5 of those patients are BRCA wild type. One of those patients has a RAD51B mutation, but for the others we don’t know why it is that they have such strong homologous recombination deficiency that they continue to benefit from olaparib for more than 6 years. And there’s 1 patient in the control arm, but she had a BRCA mutation. So these are the data that we have at the moment showing improvement in survival. And of course we wait with interest for the mature data of the SOLO2, NOVA and ARIEL3 trials. So how then should we select patients for maintenance therapy with PARP inhibitors? Should we look at patients with BRCA mutation? BRCA wild type? We’ve already discussed that and shown that the drugs are active in both groups. What about somatic versus germline? I haven’t got time to show you all those data, but I can say that from the data that we’ve seen so far, there seems to be an equivalent effect in somatic mutations of BRCA as there are in germline. Histology. Serous or high grade in general. The original trials were with serous tumors, but the more recent trials have included the high-grade endometrioid cancers. And again, the effect is seen in all the high-grade patients. So now the label is for high-grade ovarian cancer, serous or endometrioid. Let’s focus a little bit more on the next few slides on the volume of disease, the age of the patient and HRD evaluation. So if we look across the board at the results of the key trials, SOLO2, NOVA and ARIEL3, with these 3 drugs, what you can see here, ringed, is the hazard ratios showing you the effect size, if you like, in terms of progression-free survival. And we see in those patients with a BRCA mutation very similar effect size. And those patients who have wild type or BRCA, as in Study 19 or in ARIEL3, we see a hazard ratio of 0.35 or 0.36. Again, a very similar effect. And then we have the one study, in Study 19, with a BRCA wild type, have been analyzed separately, showing a hazard ratio of 0.54. So I think there’s very little to distinguish between the 3 drugs in terms of efficacy, and we’ll hear later about side effects. So let us now look at volume of disease. And this is an interesting subgroup analysis of SOLO2 looking at patients who went into CR at study entry, so having had the chemotherapy and those that had a PR at study entry. And what you can see here are the median progression-free survivals in CR patients and in PR patients, and what’s interesting about this is that the median progression-free survival in the CR patients has not yet been reached. And I think that raises questions that we need to think about in terms of when we should be treating our patients with recurrent disease. And that may come up in further discussion in the future, but this is an important observation that I think needs more thought as far as whether patients may benefit more if they’re started early with a low volume of disease and achieve a CR. What about age? I presented these. These are the ARIEL3 data in age cohorts of less than 65, greater than 75 and the 10-year cohort between. And what you can see is, the effect of the drug is similar in terms of hazard ratio across all these age cohorts. And that’s very important, particularly when we think about treating patients who don’t have a BRCA mutation, and that tends to be the older group of patients, that the drugs are effective in that older group of patients and can be used safely. When we look at HRD testing, and there are 2 commercial tests available, what we see here from the niraparib NOVA study and the ARIEL3 trial is that across all the groups of patients tested, whether they’re HRD-positive or HRD-negative as in the NOVA trial or LOH high or LOH low, which is the measure of HRD that was used in the ARIEL3 rucaparib trial, we can see that there are significant benefits of either niraparib or rucaparib. So what that means is that at the moment, we cannot distinguish which patients will or will not benefit on the basis of HRD testing. We cannot refuse a patient treatment on the basis that she’s HRD-negative, because these trials show significant benefits in the HRD-negative groups. So what we have learned so far from all of these analyses is that the drugs are more active in patients with a BRCA mutation, whether it’s somatic or germline. They have a longer progression-free survival, and the long-term benefit may be greater and may not depend on BRCA status. And high-grade tumors, either serous or endometrioid, will respond. Baseline disease does not affect the magnitude of benefit, but when we see the patients coming with a CR at the end of chemotherapy, they seem to do better, at least in that SOLO2 subgroup analysis. The HRD assays at the moment can’t distinguish patients who are going to benefit or not benefit. And platinum sensitivity remains the key biomarker, if you like, for whether or not a patient’s likely to benefit from maintenance of PARP inhibitor. And the choice of PARP inhibitor is largely dependent, I think, on the familiarity you have with the drug and its side effects. We can’t see any differences in activity when we compare across the trials. So turning now briefly to monotherapy. We’ve got olaparib and rucaparib licensed for monotherapy in patients with a BRCA mutation. This is on the basis of Phase II trials and composite Phase II studies seen here on the basis of response rate and duration of response in this setting. And you can see here for the ARIEL2 and Study 10 rucaparib trial, progression-free survival in terms of platinum sensitivity, platinum resistance and platinum refractory, you can see this falls off in relation to the platinum sensitivity. Again, platinum sensitivity being a biomarker in this setting. And lastly, we have the use of single-agent therapy with PARP inhibitors here outside the context of germline BRCA mutations. The QUADRA trial is a Phase II trial in patients who have had multiple previous lines of therapy, and they’re treated with single-agent niraparib. And you can see here, many of these patients are platinum resistant and have had multiple previous lines of therapy. Some of these patients had up to 6 previous lines of therapy. And when we look at the waterfall plot of response, here you can see in the HRD-positive group this very good effect of niraparib here in terms of response rate. Some of those patients had BRCA mutations, but most were BRCA wild type. And if we look overall we can see this benefit, this overall response rate of 10%, with a clinical benefit rate of 35% and a duration of response in these multiply pretreated patients of 9.4 months. And 44% of these responses lasted 12 months. Now, this is the sort of thing we really rarely see at this stage in the pathway of patients treated with chemotherapy. So I think these are very interesting data and I think add more to the drugs that we have available to treat patients in this later stage of their disease. So in summary, maintenance therapy with PARP inhibitors, any of the — olaparib, rucaparib or niraparib, following platinum-based chemotherapy, should now be the standard of care for patients responding to platinum-based therapy, leads to significant delay in tumor progression, delay in further chemotherapy and possibly an increase in survival. And there’s no evidence that one of these drugs is superior to any other. The monotherapy with PARP inhibitors in patients with a BRCA mutation could still be considered, particularly in patients who are not suitable for platinum-based therapy, and emerging data are suggesting that PARP inhibitors, particularly here niraparib in the QUADRA study, may provide useful disease control in later line of therapy. Prevention, Recognition and Management of PARP Inhibitor-Associated Side Effects DR COLEMAN: These are kind of the focus point of what we’re discussing today, the 3 big recurrence trials and SOLO1 in the front-line setting. And although these drugs have a lot of similarity — homology, in terms of response, and it’s very difficulty to tease those — they are different drugs. They have different elements, and I think as time goes on and we start to look at how these drugs actually combine with other drugs, these particular nuances between them will become important. We know that they’re metabolized different. And there’s interactions that we can see with these enzymes, particularly in the renal transporters. And so these are important, not just to understand about the drugs themselves in terms of their metabolism but for people who are on con meds. So concomitant medications, particularly ones that are inducing CYP enzymes, either inhibiting or augmenting them, they will definitely have an effect on the serum blood levels of these drugs. And so you need to be aware of that. When we look across the kind of the major trials where we keep the populations relatively the same, so this is always hard to do. And we put together these slides some time ago to try to at least have some homology between the patient populations. You can see the specific drugs that are listed up there. There are some things that are very common between the trials. Particularly the top three: the nausea, constipation, which may be related to the control of nausea, the vomiting. And then some of the taste changes that happen. I don’t know if patients have actually approached you about that particular feature. But they do have some very different taste changes that are different than what you see with platinum. So it’s not that tinny kind of taste. It’s just a different perception of how their taste is. And so, actually, I’ve got to credit Katie, because she’s got these nice little pictures, which I thought were cute the last time she gave this talk. So I stole them from her. But when you talk about compliance, it’s super important. So while I don’t institute a prophylactic regimen for them to start taking, I instruct them prophylactically about the likelihood that it’s going to happen. As I showed you, most of these patients have some form of GI toxicity. And I, too, try to avoid the 5-HT3s because of the constipation issue, which is also another known side effect. But I don’t know that it even works that well. I tend to use the simpler antihistamines that we used to use in the old days, prochlorperazine, et cetera, as I think that those are nice places to start, particularly since for the first month or two they may be on it a little bit more frequently than you might expect, for instance, with chemotherapy. And so you’ve got to remember, this is a daily regimen with a pill that can remind them that they’re sick. So I think that those types of things are important and are helpful. Fatigue is definitely something to ask the patients about. Obviously if they’re in the recurrent setting and multiply pretreated, this is very much on their radar, and it can have a huge impact on their quality of life. On the other hand, if you don’t ask about it, then they kind of think there’s something wrong with them. And I actually like the nonpharmacological aspect to this first. So although it’s counterintuitive that fatigue can be counteracted by exercise, but in reality, fatigue actually begets more fatigue. We know there’s about a 4:1 ratio for the amount of time that you’re in bed to the amount of time that it’ll take to get out of bed. So I try to counsel them ahead of time and to try to continue this process of keeping themselves active even though they have this fatigue. There is definitely an opportunity here to do dose modification. That can alleviate this symptom quite well. And then there are some other pharmacological approaches. But I tend to try to just get them active as much as possible and then also work on the dose modification to get them over there. Remember, all these drugs we have a big therapeutic window. All of them. So you have an opportunity to adjust the dose to the patient. So in terms of the hematologic toxicity, I think we’ve talked a lot about this. Dose exposure is probably very important here. All of the drugs — actually, if you go back to their Phase I data, you can actually induce a lot of hematologic toxicity based on dose. And they all have the characteristic curves. What’s interesting is, if you look at the Phase I data that informed the Phase II dose, many times they’re picked on the upper limit of the range, so where the toxicity is actually fairly high. So the dose exposure for all these drugs have really been picked at levels that were intended to be on the higher level of the AUC for the pharmacokinetics, which I’m not going to show. But just know that that can happen. And, of course, all of us are very familiar that these factors can happen. And there’s probably some differential between the drugs. And you can see that, for instance, anemia is commonly reported in the olaparib trials — platelet toxicities, at least high-grade platelet toxicities. And we now know about the mitigation of that in the prospective aspect of using an adjustment before going onto therapy, that you actually are able to ameliorate the high-grade thrombocytopenia by about 50%. So it’s quite nice from that standpoint. The anemia is interesting. I was looking at this for some time trying to figure out what the mechanism was, and I ran across this paper that was published in Cell Death & Differentiation in 2015. This was a preclinical experiment looking at the impact of PARP-2 on erythrogenesis. And what you’ll see in this paper, on the curves on the right, there’s actually a little element of hemolysis going on. So that part of the reason that we’re seeing some anemia is this low-grade hemolysis. And it also is probably due to the decreased expectation for the erythrocyte survival. So after they’re being produced and circulated, their overall time in circulation is reduced by a small fraction. And so it may be that PARP-2 activity actually could be driving some of the anemia that we’re seeing in the studies. And you can see anemia and the anemia trends for SOLO2 and for ARIEL3 are presented here, just to give you an idea of that reference over time. So we do see some long term, but with respect to the dose adjustments and some tolerance of the drug, we see that those are normalized and don’t end up in serious issues. Platelets, as I mentioned before, definitely is a dose-effect factor that we saw. These are those platelet toxicities in NOVA and ARIEL3, where we had those nice curves that have been followed serially. And again, many of these, just remember that these are going to be the reflection of your interaction. So how did you respond to the fact that this toxicity as it was seen? So one of the things we learned very early on in NOVA was that we had to follow those values early on a trial when we didn’t do the dose adjustment up front. So we did, we saw it, we made an adjustment and we were able to get patients back onto an acceptable dose that went out for many, many months. And you can see the curves go off into several different cycles, over 25 cycles. So you’ve seen these nice-colored curves before, because they kind of demonstrate what the impact is of your dose modification along the way. And as was mentioned in the case with the now prespecified dose adjustments based on weight and by platelet baseline levels, you can see that these curves now start to end up at basically an optimized dose, after about the fourth month or so when you don’t do a prospective adjustment for dosing. So in terms of trying to keep patients on drug for compliance reasons, to the highest level that we can, these are some general guidelines. And I put them in here just so you guys can have them as takeaway bullets. But essentially we follow these labs weekly in the first month. Frequently we’ll get them more frequently than that, maybe weekly in the first month, just to assess for the side effects. But in general, your response to those things will lead to a sustainable regimen that can be given over time. Now, the nonhematologic toxicities are also quite interesting. This is the creatinine and the ALT that was shown in ARIEL3. And you can see these values, they do bump up, and they stay stable over time. But as is the case and our expectation, the creatinine is not due to kidney injury. And when we do this, I do send them for a nucleotide GFR scan so that we can actually assess whether or not it’s real or not. That’s one of the things we do in addition to looking for concomitant medications, because we want to make sure that there’s not something more serious going on. Fortunately, the high-grade rates are very low, and these do normalize with time. Hypertension is also a side effect that we see. It’s been reported with niraparib. We think that this might be due to the dopamine norepinephrine metabolism disruption for the compound. Again, these things can sometimes be quite obvious, but many times they’re just not. And so we just educate the patient and will do a dose adjustment to address them at least as the immediate first move when it’s present. And then I just put the summary of them here. All of the drugs have this uncommon side effect. Whether or not this is more common than expectation of chemotherapy, we’ll have some information about that at ASCO. So be on the alert for that. But we do know this is an adverse event of special interest, and so we’re constantly trying to make sure that we understand this. This is also something that we have to be careful of if we are just treating with some level of cavalier expectation. So I do want to say that that is something we have to be aware of, and it can be something that will be further evaluated. In terms of dose, treatment discontinuations, this comes up a lot. Fortunately, it’s not common, but it does happen in about 10% to 15% of patients. So PARPs definitely have a class effect. I think what I mentioned shows you that. The quality of life that we’ve seen so far, at least, is not to be impacted negatively, even with increased side effects over placebo, and Katie showed that in SOLO1 very nicely. Some of these drugs do have unique toxicities, but our ability to dose modify and to dose reduce can ameliorate that so patients can stay on. Investigational Applications of Approved PARP Inhibitors; Other PARP Inhibitors in Clinical Development DR LHEUREUX: When we can’t get DNA repair, the big success of targeting DNA repair was first the cisplatin, but then now is a PARP inhibitor treatment. It’s really a success story from the discovery when we knew the BRCA germline status. We have some preclinical data showing some activity. Then the clinical trial with early stage showed some potential benefit. Then we moved to the late stage of the trial, still showing a really good benefit that led to the approval of different PARP inhibitor treatments in the recurrent setting, maintenance. And we moved forward in the first line, and again, olaparib has been recently approved as maintenance in the first-line setting. But where do we go from here? So what is important is, first, what is the target? As we discuss the issue of the biomarker here, just a reminder, we target the DNA repair. PARP inhibitor actually inhibits PARP, which is involved in the single-strand break. And we discuss how to select the patient based on the HRD deficiency and the homologous repair deficiency, which actually is on the other side. So we try to see how to select our patients who respond to the PARP inhibitor. Currently, we have companion diagnostic tests. But it’s not sufficient. So we’re trying to move forward to try to see if we can have some functional test to monitor this homologous repair deficiency. So we need to have functional tests, which is very different than looking at the archival tissue for HRD. And then moving forward, as we heard, more and more of our patients will have PARP inhibitor treatment. The mechanism will be different. We need to monitor this disease. And the only way eventually we can monitor that through the treatment is eventually with circulating tumor DNA. So it’s where? How will we try to define better a patient? By defining the target. But we heard of these different PARP inhibitor treatments. How we’ll choose which PARP inhibitor treatment? We heard the 3 PARP inhibitors that have been approved by FDA, and we talk about PARP inhibitor PK. The pharmacokinetic of the drug is very important. Very important, because there’s a specific half-life and a specific metabolism. And it’s true, when we talk about nausea, the PK was key, because initially, for example, olaparib or rucaparib was a capsule, and they have to take, for example, olaparib 8 capsules in the morning, 8 capsules at night, which actually gives preemptive nausea. Just by taking these 8 capsules. So the PK was very key. And then here you can see here, the picture is not the right size, but how now the patient has to take these pills. So it’s very important in terms of nausea. And the PK is also important because it allows the patient to take these pills, eventually with a snack. That can also improve the tolerance of the drug. There’s also some drugs in development, and veliparib, that we’ll talk about, and also the talazoparib. Any difference? So it’s where the pharmacodynamic of the drug is important to say, what is the difference between these PARP inhibitor treatment? First, in terms of pharmacodynamic is how it can reach the target. As we say, you need to target the PARP1 and PARP2, and the targets differ according to the different PARP inhibitor treatments. There’s also PARP trapping, and the difference between olaparib with talazoparib is very different. Talazoparib binds and traps the PARP more effectively than all the other PARP inhibitor treatments. This was done in preclinical setting. We do not know if it translates in the clinical setting. There’s variation in the clinical setting, because probably there’s patient selection, but also some dose-limiting toxicity. And the assay is very variable. Don’t forget, as well, that when we talk about PARP inhibitor treatment, there’s probably off-target effect that probably varies among the PARP inhibitor treatments as well. Veliparib. We don’t hear about veliparib, so we’ll just go quickly on it. In terms of preclinical data, what we know: It inhibits PARP-1 and -2. And, as most of the PARP inhibitor treatment, they pass the brain barrier. Interestingly, in preclinical data, veliparib didn’t show single-agent activity. It was potentiated with chemotherapy. In terms of the clinical data, in triple-negative breast cancer, veliparib combined to chemotherapy failed to show an improvement in terms of pathological response. But it showed some efficacy in combination with chemo in non-small lung cancer. What about ovarian cancer? We have a Phase II trial with germline BRCA1 and 2 mutation, and the overall response rate was 20% in platinum resistant and 35% in the platinum-sensitive setting. There’s different trials currently still ongoing with veliparib, and one is really the VELIA trial, which is veliparib with chemotherapy followed by maintenance therapy in the first-line setting. So we are awaiting the results. The other one to come back to is talazoparib. Talazoparib is a new PARP inhibitor treatment. What do we know in terms of preclinical data? It’s a highly potent PARP inhibitor treatment, as we saw, with really good PARP trapping. So really traps really well. More than most of the PARP inhibitor treatment. So very efficient for the BRCA patient. And it’s synergistic with other treatment. What do we know in the clinical setting? We have Phase I data showing that the maximum tolerated dose is 1 mg per day. As you see, the half-life is higher than the other PARP inhibitor treatments and is well tolerated. In terms of development, the talazoparib is FDA approved, but in breast cancer, germline BRCA mutation, HER2-negative. Why? It’s because talazoparib shows PFS improvements statistically significant versus chemotherapy and based on early signal in the interim overall survival. And in terms of patient-reported outcome, which is very important to assess the toxicity, it was beneficial to talazoparib. What we do in ovarian cancer is currently developed specifically for the BRCA population, given the fact that it’s good PARP trapping. So we heard that most of our patients should have a PARP inhibitor treatment at one point in her treatment as maintenance for platinum sensitive. So we unfortunately have to overcome the resistance. What is a mechanism that we know currently about resistance? There are 3 boxes. So the first box is related to the mechanism of action. We target DNA repair pathway, the homologous repair deficiency. So actually there’s a restoration of the homologous repair deficiency by reversion mutation, methylation of the HR gene or eventually hypoxia or the upregulation of PI3K pathway. Then the specific box of the replication stress. Why? We look at the DNA repair, but we cannot forget that the DNA repair is linked to the cell cycle. So that is also very important to see and assay the replication stress in the specific cancer. And the third box is, the PARP inhibitor cannot reach actually the target. And it’s all the problem with the drug efflux. What we know the most, and what we heard a lot, is the PARP with the reversion mutation. It’s a dynamic process with de novo cellular adaptation under treatment pressure. The one that we know the best is the BRCA1 and 2 reversion, and what is interesting is that it’s seen more in the resistant setting, and we can see more than one BRCA reversion in the patient. And what we know is that to have a reversion BRCA mutation, as you can see on the graph, is worse for the outcome than not having a reversion BRCA mutation. But that shows that it’s a dynamic process. And really highlights the need to have a functional biomarker that we can follow through disease’s treatment. There’s also secondary BRCA mutation in RAD51C and RAD51D that was described. But globally, the mechanism for different PARP inhibitor treatment will be different and specific to one patient. But globally, all that will follow on the fact that we have the restoration of the DNA repair function. So what we will do? We will have to combine. There’s different ways we can combine. When we look at the cancer cell, we look at the DNA repair pathway, but as we know, we need to look at the cell cycle. So one of the ideas is eventually to combine the PARP inhibitor treatment with a cell cycle inhibitor. We can also look at the microenvironment, and we can definitely try to see if we have anti-angiogenic or immune therapy. In terms of PARP inhibitor and anti-angiogenics, it’s an active area of investigation. We know that bevacizumab is active in ovarian cancer. There’s different trials currently ongoing assessing PARP and bevacizumab. One Phase II trial that has presented at ESMO in 2017 with niraparib and bevacizumab showed it was feasible and safe, and we are currently awaiting the results of the randomized Phase II. There’s other anti-angiogenics, for example cediranib, which has broader activity in terms of anti-angiogenics. And when we compare olaparib to the combination olaparib and cediranib, this is significant in terms of PFS and overall survival, specifically and clinically for the patients BRCA wild type. And as we know, by combination we increase the side effects. And the side effects were mainly fatigue, diarrhea and hypertension. Now, we also move forward with a combination of PARP and immune therapy. Why? Because when you have a DNA repair deficiency, you have intrinsic immunogenicity. How? Because you have this upregulation of the STING NF-kappaB pathway. Moreover, when you treat the patient with a PARP inhibitor treatment, you induce PD-L1. So that is the rationale not to combine the two. And in the specific population of BRCA1 deficiency, we know that they have a higher neoantigen load. And there’s different trials that have been done, and a lot are still ongoing, but I just want to show you two. This trial actually has the same thing, niraparib and pembrolizumab. The Phase I has been done. And there’s some interesting early activity in platinum resistance, with a response rate of about 25%. Another one, MEDIOLA, it was a basket trial assessing olaparib and durvalumab. Again, it seems to be safe, and some early activity that we need to be validated. And the trial is still ongoing with a new arm assessing the triplet — olaparib, durvalumab and bevacizumab. And just to show it’s an exciting area that we need to follow for our patients. |