Ovarian Cancer Therapies in Development

Ovarian cancer has been historically difficult to treat, in part because the cancer is often only diagnosed at a late stage in the disease. Ovarian cancer often has few symptoms that are unique to the disease, and there are currently no reliable screening tests for the general population. Ovarian cancer is also diverse, and sequencing studies continue to show that the cancer is genomically complex. Ovarian cancer is most often treated with surgery and/or chemotherapy. Recently, several new agents have shown promise in early-stage clinical trials. Today we speak with Dr. Robert Coleman, of the department of gynecologic oncology and reproductive medicine at the University of Texas MD Anderson Cancer Center in Houston. Dr. Coleman treats women with ovarian cancer and is active in clinical trial research to develop new therapies for women with ovarian cancer.

-Interviewed by Anna Azvolinsky, PhD

Cancer Network: Dr. Coleman, I mentioned some of the reasons why ovarian cancer is difficult to treat, but could you discuss this in more detail?

[[{"type":"media","view_mode":"media_crop","fid":"18004","attributes":{"alt":"","class":"media-image media-image-right","id":"media_crop_4263786671612","media_crop_h":"0","media_crop_image_style":"-1","media_crop_instance":"1202","media_crop_rotate":"0","media_crop_scale_h":"0","media_crop_scale_w":"0","media_crop_w":"0","media_crop_x":"0","media_crop_y":"0","style":"float: right;","title":"Robert Coleman, MD","typeof":"foaf:Image"}}]]Dr. Coleman: One of the things that we have known for a long time and that now has been really well documented is that ovarian cancer is really many cancers; it is a really heterogeneous disease. Patients that have the same histology under a microscope have vastly different, genomically driven profiles that dictate the natural history. That is really where our first problem lies-that we have been designing our trials to take all-comers with a certain type of histology and treat them with the same thing. That combined with advanced stage of presentation and large, bulky disease really hampers our ability to treat this. It is really clear that we are talking about many different diseases when we break it down. One of the other issues that is really a problem is that although we do get responses in the majority of patients, we see frequent recurrence even when it looks like the disease is all gone. Because of this, we don’t have any reliable strategy, at the moment, to really impact long-term overall cures. One node of hope is that the median survival of this disease is actually increasing quite substantially, and that is largely due to the increasing portfolio that we have available to us to choose treatment options.

Cancer Network: Are there any promising genetic targets for any ovarian cancer patients or any subpopulations that have been defined?

Dr. Coleman: One of the things that came out of the Atlas Genome Project was that, as I mentioned, many diseases that are wrapped up into the term ovarian cancer-even among the high-grade serous ovarian cancer, which is the most common type. We know that this is a genomically unstable cancer, and some of the other histologies that we deal with-mucinous, clear-cell, endometrioid, and low-grade serous-have started to be identified with similar types of aberrations that appear to be somewhat targetable, or at least somewhat more than we have seen with the routinely seen high-grade serous cancer.

Let me just profile a couple because I think they are illustrative of where targeted therapies are starting to make an inroad in this disease. The most visible of these is the low-grade serous ovarian cancer, which although it shares the serous histology name with the more common high-grade cancer, it is really a different disease. It does not seem to be as bizarre from a genomic standpoint as high-grade serous ovarian cancer. One of the pathways that was identified in this tumor type early on was that it had frequent aberrations in the MAP kinase and PI3 kinase pathways. So, with that in mind, we launched a trial within the Gynecologic Oncology Group (GOG) to look at a targeted agent to MEK, a drug called selumetinib. We were surprised to see that not only does this drug produce responses, which are very uncommon in this particular low-grade histology subtype, but that it also delayed progression for substantially long periods of time. Because of the activity of this drug, which was really remarkable compared with what our experience has been with hormonal therapy and chemotherapy, this has launched an investigative effort in a phase III setting for at least two other compounds. There have also been new protocols that are looking at blocking the other targets in this pathway, so combinatorial trials are also in development and underway.

Another tumor type for which there is a targeted approach is a clear-cell endometrial tumor, which frequently has aberrations in the same pathways, particularly in the PI3 kinase pathways. This is important because we have many drugs being developed across the spectrum of cancers that target PI3 kinase, mTOR, AKT, and some of the downstream targets in this PI3 kinase pathway. I think with these specific subtypes, we are coming into a realm where we are able to carve out specific tumor types and, hopefully in the future, specific pathways that are present in any specific tumor for which we would have actionable events that we could target with therapies.

Cancer Network: As far as other therapies in development, there is an anti-angiogenesis oral agent, trebananib, which is also known as AMG 386. There have been positive early results from a phase III trial in women who are partially resistant or resistant to platinum-based chemotherapy. Could you describe the rationale for this drug in ovarian cancer and what do the results tell us so far?

Dr. Coleman: We are excited to see this new kind of therapy introduced into the treatment domain for patients with recurrent disease. Trebananib is a drug that was reported on a few years ago at ASCO and has been published, and it is a unique peptibody that blocks the interaction between angiopoietin (ang) and its receptors, the Tie receptors. We have come to understand that the ang-1 and ang-2 axis have different functions on the quality and the quantity of blood vessels that are recruited and developed in the tumor microenvironment. It is an alternative pathway to the more well-understood anti–vascular endothelial growth factor (VEGF)-based therapies that are working on the VEGF/VEGFR axis. We had information from the randomized phase II trial, which suggested that there may be activity with this drug, particularly in combination with paclitaxel.

The drug went into robust clinical development after those reports became available, looking at its combination with both paclitaxel in the TRINOVA-1 study and with liposomal doxorubicin in the TRINOVA-2 study. We heard just recently the top-line results of the TRINOVA-1 study, which was the randomized trial of over 900 patients between paclitaxel and paclitaxel plus trebananib. What we saw was that in this group of patients, defined as recurrent as you mentioned, they were both partially and fully resistant to platinum-based therapies. They allowed patients with a progression-free survival of less than 12 months into the trial. What they showed was that the progression-free survival hit the target that the trial was shooting for, which was a difference of about 2 months-7.2 months in the experimental arm compared with 5.4 months in the control arm, with a hazard ratio of about 0.66, which means that the reduction in the hazard for progression was reduced by about one-third. So, this is a significant report. Now the rest of the data are still coming, and we hope to hear a more formal presentation in the fall and next year when the overall survival will be reported. What will be important is to understand the toxicity that this impact in positive progression-free survival was associated with. The toxicity and efficacy always have to be balanced, and once we see the more mature data, hopefully we can make a decision about where this particular combination may fit in the treatment program for women with recurrent ovarian cancer.

Cancer Network: Another oral drug, pazopanib, is being studied as a maintenance therapy. Are there currently approved maintenance therapies for ovarian cancer and where would pazopanib fit in the treatment paradigm?

Dr. Coleman: That is a good question, but unfortunately it is a pretty complex question. Certainly when we talk to patients about what to do after frontline or after subsequent lines of therapy, we discuss how to sustain their best response. In terms of whether there have been any approved strategies, there are no formally approved strategies in the United States for maintenance. We do have efficacy data with regard to progression-free survival with paclitaxel that was obtained from an older trial that looked at a year’s worth of paclitaxel. Those particular data are currently being confirmed in a more current trial that is still accruing, the GOG 212 trial. This is looking at no treatment vs two different taxanes, for one year’s worth of exposure to confirm that previous positive result. But, it is not like we haven’t tried to find a suitable maintenance therapy.

There have been more than a dozen randomized trials in the maintenance setting that have, unfortunately, not shown any benefit. The latest before the report from OVAR16 involving pazopanib, was erlotinib in a large European Organisation for Research and Treatment of Cancer (EORTC) trial that showed no benefit of the long-term exposure to that drug (link). With that background, we look at this new data, presented at this past ASCO meeting, on pazopanib. This is an oral tyrosine kinase inhibitor that blocks the VEGF/VEGFR interaction, as well as other pathways that are relevant to the tumor microenvironment. What we saw in the trial, at least in the top-line results, was that there was a substantial lengthening in progression-free survival of over 5 months; 17.9 months compared with 12.3 months and a reduction in the hazard for events of about a quarter or so, a hazard ratio of 0.77. So, very exciting data. The overall survival data are still immature, and under these circumstances, it may be difficult to show an overall survival benefit because patients are living so much longer. One of the things that is interesting about this trial is that it goes back to its design. What the trial designers tried to do was to continue exposure to this drug through the time point where most patients would ultimately recur.

One of the criticisms of previous trials has been that the maintenance exposure has been too short. We saw that with the bevacizumab trials, where bevacizumab was given to frontline patients and then continued in the maintenance setting. It appeared from the analysis of the two frontline trials, GOG 218 and ICON7, that potentially the benefit was lost when the drug was withdrawn. This particular trial allowed patients to stay on pazopanib for up to 2 years, so, as you can tell from the control group, this appeared to cover the exposure time when most patients would have recurred. Many of these patients were on treatment until progression, and again, when we look at these trials with regard to progression-free survival, we always need to balance the efficacy with the toxicity. We know that there were patients who came off of this trial because of toxicity. Ultimately, we will need to compare the survivorship result against the potential toxicities that were observed in the trial.

Cancer Network: Just briefly, are there any other drugs that are in either early or late stage of development that you see as particularly promising that you can highlight?

Dr. Coleman: I think there are several new drugs that are worth taking through the system. I think that one area that was exciting this year was to look at new approaches towards immunotherapy-based strategies. Right now many of these are going down different paths. One is a way to figure out how to use the immune system to vaccinate against the ovarian cancer on a per-patient basis. There is a study that is ongoing right now that is trying to show whether that is beneficial. The other path is to use immuno-targets as a way to bring drugs into cells, and that has brought a whole class of drugs called the antibody-drug conjugates or the antigen-drug conjugates, and both of these classes of compounds have drugs that are being delivered in a more selective manner to tumors. Probably the most mature of these is vintafolide, which is an antigen-drug conjugate to folate, and that’s in a phase III trial right now (link). I think these-along with the new Toll-like receptor agonists and those that are targeting programmed death receptor-1 (PD-1) and its ligand (PD-L1), seen in melanoma treatment-represent new avenues of really important drug development going forward as we learn about how immune surveillance happens in patients with this disease.

Cancer Network: Thank you so much for joining us, Dr. Coleman.

Dr. Coleman: A pleasure, thank you for having me.