http://www.onclive.com/web-exclusives/targeting-tumor-vasculature-in-ovarian-cancer?p=3
Targeting Tumor Vasculature in Ovarian Cancer

Angelica Welch

Robert L. Coleman, MD
Robert L. Coleman, MD
Angiogenesis inhibition is a process for tumor survival that has been known for many years, says Robert L. Coleman, MD. In ovarian cancer, antiangiogenesis agents have shown particular benefit, as they target tumor vascular biology.

Specifically, the antiangiogenesis agent bevacizumab (Avastin) has been a storied component of the treatment paradigm of ovarian cancer since its approval. More recently, combinations with bevacizumab have been studied in this tumor type, specifically with PARP inhibitors—which have exploded onto the scene these past 2 years.

In an interview with OncLive during the 35th annual CFS®, Coleman, professor, Department of Gynecologic Oncology and Reproductive Medicine, The University of Texas MD Anderson Cancer Center, discussed targeting tumor vasculature and angiogenesis in ovarian cancer.

OncLive: Please provide an overview of your lecture on targeting tumor vasculature in ovarian cancer.

Coleman: I started out with a reference from 1907 in Lancet, where there was a question posed looking at the vasculature around cancers—asking why they were there, what role they play, and how to take care of it. I used that as an opening because it was more than a century ago that we recognized there was some relationship, and it was only about half of a century since we started to understand why those processes happen. Even today, it’s a well-acknowledged factor of tumor biology, and how we are approaching treatment is developing science.

Dr Judah Folkman, who many of us think of as a pioneer [of angiogenesis], identified that there were circulating factors in the tumor microenvironment that were actually regulating and controlling the expression and development of blood vessels, and how that related to metastases and invasion. Since then, we have learned that there are multiple factors along the way, and there are different types of angiogenesis. One of the strategies that we have gone after is sort of the lowest-hanging fruit—the growth factor and its receptor. Drugs such as bevacizumab and the tyrosine kinase inhibitors (TKIs) that are targeted at VEGF really emerged on the scene when they showed single-agent activity in multiple different disease types.

In ovarian cancer, we at the Gynecologic Oncology Group (GOG) launched a series of phase II trials that were based on similar eligibility, exclusion criteria, and statistical design. We were looking for that “sweet spot” between activity from response and delay in progression. When you look at the drugs that we have studied, only bevacizumab actually hit both of those endpoints. It was then rapidly developed into the recurrent and frontline setting, and since then, we have had phase III trials that look at bevacizumab, and 9 trials that have looked at other VEGF-targeted therapies that have shown a progression-free survival (PFS) benefit in the frontline and maintenance settings, and in platinum-sensitive and platinum resistant/recurrent disease.

What about other VEGF agents, such as ramucirumab (Cyramza)?

[Ramucirumab] is not being looked at in ovarian yet, but the idea is the same. This is a drug that is an antibody to VEGFR2, which is an important receptor, particularly in gynecologic cancers. In endometrial, ovarian, and cervix cancers, that receptor seems to be important, and we have studied them. Nintedanib (Ofev, Vargatef) and pazopanib (Votrient) are TKIs that have been studied in the phase III setting that also target this receptor.  We have looked at other growth factors such as angiopoietins, which have been investigated. These seem to be important for the endothelial remodeling that happens in the angiogenesis process. It has a similar story about improvement in PFS.

Are there any unanswered questions with bevacizumab in ovarian cancer?

Great question; what do we do next? I always say that we either have to add to it, look at it in sequence, or deal with its adaptive resistance mechanisms to make this strategy better. One of the areas of research that I have been working on for many years now is to try to understand what happens in the microenvironment where the tumors seem to learn how to escape.

In my talk, I show this banana—and it is kind of a joke—but if you think about the shape of a banana, this is what we are seeing in the PFS curves. At the beginning, everyone is the same, then there is a split while they are on the treatment, and then the split comes back together. The tails of the curves collapse, but the middle of the banana is still there; there is an effect that gets lost over time. We hypothesized in the past that maybe it was just because we pulled the drug away; some of the studies were limited exposure to the antiangiogenesis drug. However, even when we were allowed to use those drugs to natural progression, we still saw the same banana effect. The GOG-262 trial is a very good example of that kind of strategy in the frontline setting. 

Why is this happening? We created this animal model in immuno-competent animals to basically recapitulate that. We treated all of these animals, and some of them developed resistance to the targeting while others were sensitive and stayed sensitive. We looked at the tumor at the beginning, middle, and at the time of resistance, and we found that one of the factors driving resistance has to do with the infiltration of an immune cell called macrophage. We identified that these patients, who have an emergence of adaptive resistance, are actually trafficking macrophages from the bone marrow into the microenvironment where there are stimulating factors of angiogenesis escape. 

We have an immune response to an antiangiogenesis strategy. That is one of our focuses in a clinical trial that we have launched, which is actually targeting these macrophages at the point when we think they are trafficking into the tumor. The way that the trial is set up is that everyone receives standard chemotherapy plus bevacizumab, but the patients who are not responding after the first 2 cycles of therapy are actually then randomized to continuation of treatment versus continuation of treatment plus a CSF1R-inhibiting drug. Hopefully, we will identify if this particular mechanism can be exploited.

However, there are many others. We understand that there are regulatory factors on natural angiogenesis. For instance, vasohibin-1 (VASH1) is a microenvironment molecule that is controlled by the transcription factors and by EZH2. It just so happens that EZH2 inhibitors are starting to emerge on the scene. Therefore, EZH2 is an inhibitor of VASH, which is a natural angiogenesis inhibitor. By targeted EZH2, we release the restriction on VASH1, and potentially induce a resensitization to VEGF therapy. How do we make bevacizumab better? This is a way. Identifying the patients who are ultimately developing these resistance or escape mechanisms will help us learn how to target.

Is there potential to combine bevacizumab with immunotherapy agents or PARP inhibitors? 

Right now, our excitement is around the immunotherapy and PARP inhibitor spaces. Both of those additional drugs are being used in combination with anti–VEGF-based strategies. Those are currently ongoing. In fact, even the National Cancer Institute is running 2 trials; one is in platinum-resistant patients and one in platinum-sensitive patients and is using the combination of an antiangiogenesis drug with a PARP inhibitor. The idea is that the antiangiogenesis inhibitor induces local hypoxia and hypoxia can then lead to homologous recombination (HR) deficiency. You can take an HR-competent tumor and make it incompetent via hypoxia, and then treat it with a PARP inhibitor.

Is there anything else you would like to add?

As we have learned more and more about what is going on in the microenvironment with angiogenesis, we have learned that there are many growth factors. That is what led to the development of things such as FGF inhibitors. We know that there are other elements within the microenvironment such as pericytes, which are supportive cells for endothelial cells, and those can be targeted. Therefore, we are looking at the endothelial cells as potential primary targets with vascular-disrupting agents and anti-VEGF therapies; we are looking for cytokines, we are looking at targeting the pericytes that may make the endothelial cells more vulnerable, and we are looking at the immune escape mechanisms.

One of the things that has always been problematic for us is that VEGF is a natural growth factor that is responsible for our own wound healing, so one of the side effects is the impact on normal processes. We have never been able to completely document why we see these catastrophic complications, such as bowel perforation, in our patients with ovarian cancer who are treated with this class of drug. One of the hypotheses is that microtrauma happens in us all the time. If you inhibit microtrauma in compromised patients, that microtrauma can lead to localized infection, and ultimately, to viscous disruption and catastrophic bowel perforation. We see them, so we know that they occur.

We were really interested in seeing if there was a way to distinguish tumor angiogenesis and normal angiogenesis. We did a series of experiments with my colleague Dr Anil Sood from The University of Texas MD Anderson Cancer Center, and we tried to distinguish factors that were driving tumor angiogenesis in patients, but were missing in the normal angiogenesis.

We did a series of experiments, including wound-disruption studies. We targeted the endothelial cells and we got these from our patients intraoperatively. What we found was that there was a series of factors that seemed to drive tumor angiogenesis, which are missing from normal angiogenesis. We have been in the process of trying to come up with a therapeutic targeting strategy for that specific factor. We know that these are relevant, and we are really excited at the possibility of coming up with a tumor angiogenesis drug that won't affect wound healing. 
Printer Printing...