Clinical Trial Highlights in MDS

Clinical Commentaries published on August 10, 2017
Download Transcript Download Audio
Guillermo Garcia-Manero, MD
Dr. Kenneth B. McCredie Chair in Clinical Leukemia Research
Chief, Section of Myelodysplastic Syndromes
Department of Leukemia
The University of Texas MD Anderson Cancer Center
Houston, Texas
Clinical Trial Highlights in MDS

To view additional highlights from the 14th International Symposium on MDS, with the option to earn continuing education credit, visit our Annual Meeting Highlights page.

Welcome to Managing MDS. I am Dr. Guillermo Garcia-Manero, and today, I will be providing highlights from the 14th International Symposium on MDS in Valencia, Spain. This MDS meeting happens every other year and it is one of the key meetings for clinical, translational, and basic investigators working in MDS. Today, I am going to summarize two different activities. I am going to summarize one of the talks that I gave regarding the landscape of clinical trials in North America, and then I am going to give you a summary of one of the other presentations pertaining to a compound known as CC-486 for patients with hypomethylating agent failure MDS.

With that, I am going to start first with the summary of the talk that I gave on U.S. MDS clinical trials. First of all, let me start by saying that we are very excited that the portfolio of clinical trials is increasing in number and quality. I think this is going to result, hopefully in a few years, in multiple drugs approved for our patients with MDS. This is, of course, a great need for our patients and ourselves.

One key slide shows how I see the different needs for our patients. As you can see, I think of four different buckets, or situations. One is patients with low-risk disease that have not received any prior therapy. The second is patients with low-risk disease that have already gone through a hypomethylating agent; we call this low-risk HMA failure. The third one is patients with high-risk disease that have not received any prior therapy, and the last one is patients with high-risk HMA failure myelodysplastic syndrome.

I divide patients like that because, first of all, these are the true clinical needs for our patients. Second, they have very distinct biological and clinical characteristics that make their treatment quite different. Therefore, we are thinking about treatments in those different subsets with a goal to, at some point, get approval indications for each one of the subsets of patients. Let's start with low-risk MDS. This is a disease that is very heterogenous. Some of the patients may or may not need therapy. A lot of investigators have thought about these patients in terms of basically ameliorating transfusion requirements. Some others, like myself, have thought more from the perspective of trying to develop disease-modifying interventions.

In the first slide that I presented at this meeting, there was basically a menu, if you will, of different clinical trials that are already ongoing in the United States and also in Europe. For instance, we discussed a very important phase 3 trial of CC-486. This is oral azacitidine in patients with low-risk disease that are transfusion dependent of red cells and have a low platelet count. This is a subset of patients with poor prognosis. Even if they are at lower risk, the disease entity is a very urgent need, and this clinical trial is open basically all over the world. It is not accruing very fast due to the specificity of the eligibility criteria in the subset of these patients, but it is moving along both in North America and in Europe. I think that the data that we have generated in the past will prove that low-dose oral hypomethylating agents will be very effective for our patients. We also discussed in a little bit of detail the use of new TGF beta inhibitors. One of these compounds is known as ACE-536. It is in a phase 3 trial that completed accrual the week before the MDS Valencia symposium, so we are very excited about this. This compound - known as luspatercept - has been shown in prior trials to have an excellent toxicity profile and very significant activity in patients with low-risk MDS anemic transfusion dependent, particularly if they have refractory anemia with ring sideroblasts. The study, as I mentioned, just completed accrual. Now, we are going to wait to see the results of this trial, but we are very happy with how fast and how well the study was conducted. We have to wait to see what type of clinical output this provides us. A very similar compound was tested in a trial in North America known as ACE-11, or sotatercept, and this gave us really excellent results in terms of improving transfusion requirements in our patients. During this presentation, we talked about other more investigational types of approaches We do not have a lot of data, but some of these patients are characterized by splicing mutations. There is already an ongoing phase 1 trial with drugs like H3 biocompound that inhibit patients with this type of splicing mutation. We talked about new monoclonal antibodies, for instance, targeting CD38. The studies are starting to happen both in the United States and in Europe. We talked about other modalities, like immune checkpoint inhibitors, that I will discuss a little bit later.

One of the key aspects of my presentation was focusing on the use of lower doses of hypomethylating agents, either azacitidine or decitabine. To me, this is very important because in Europe they do not have these compounds approved for patients with lower risk disease, so they are not used and I do not think they are really globally understood. Data from the original oral azacitidine trial (and then the studies with lower doses of decitabine and lower doses of azacitidine that I updated at this meeting) clearly show that attenuated schedules of these compounds in patients with lower-risk MDS are actually very safe and very, very active. We showed a complete response rate close to 30% with a quarter of these patients having complete cytogenetic responses. Most importantly, what this data is showing us is that the survival of these patients may be longer than what we would expect from prior historical data. I made a plea to my European colleagues to start thinking about the hypomethylating agents, not only in high-risk disease but also in lower-risk disease.

I mentioned a study that is going through the North American MDS Clinical Research Consortium (CRC) that I currently lead, where we are doing a trial of early intervention with lower doses of hypomethylating agents for patients with lower-risk MDS. We think that this is going to be a very important initiative. The study has accrued almost 50% of the 250 patients that we plan to enroll, and we are very excited about this initiative.

That said, the use of hypomethylating agents has led to a new kind of clinical subset of patients that is what we call HMA failure low-risk MDS. Again, this may be something that we only see in the United States or in countries where the hypomethylating agents are used for lower-risk disease.

The first question is: What is their outcome when azacitidine or decitabine stops benefiting our patients? We and others have published survival for about a year and a half. It is not an acute situation but one associated with poor prognosis.

Our data indicates that stem cell transplantation is very important in this context, and best outcomes are probably seen in those patients that are still transplant candidates.

We also showed some data with a toll-like receptor 2 inhibitor known as OPN-305, which targets the pathway that we call innate immunity. Interestingly, in this meeting, Dr. List gave a very interesting talk about how important innate immunity is from the pathophysiology of myelodysplastic syndrome, but also as a potential therapeutic target. We have pioneered these targeting TLR2, and a group from Moffitt with Dr. List is looking at targeting TLR4 and molecules associated with that particular receptor. Again, we will be talking about immune checkpoint inhibitors a little bit later.

When we move to the front-line high-risk subset of patients, here is where azacitidine has given us the best results, not only in terms of response but also in survival, and remains the standard of care with a 7- or 5-day schedule. This is basically our building block.

We discussed new second-generation hypomethylating agents like guadecitabine or SGI-110. These are studies that are ongoing at our center and also through major phase 3 trials in acute myelogenous leukemia, and I will be discussing a little bit later some of the newer combinations.

I think one of the key needs is the group of patients with hypomethylating agent failure disease in the high-risk situation. This is something that now all of us are very familiar with. These are patients with high-risk MDS that we have treated with azacitidine or decitabine that have not responded, or that now are not benefiting anymore from the therapy: we call that failure. We know that the prognosis of these patients is poor, with a survival of around 4 to 6 months. This is a difficult group of patients because, in general, they do not really respond well to AML-like chemotherapy type strategies, and it is an intense focus of our research and research in multiple centers.

I presented data from a phase 3 clinical trial with a compound known as rigosertib that was kind of agnostic to any HMA failure. I discussed the design of a trial known as INSPIRE that is now ongoing, trying to focus on patients with hypomethylating agent primary failure; in order words, patients that have not responded to azacitidine or decitabine. On the original study that was published by us in Lancet Oncology last year, we showed that this group of patients actually had a significant benefit in terms of survival. We are trying to reiterate these data, hopefully with the goal of obtaining an approval indication for this compound for these patients with HMA failure disease. We talked about some data with SGI-110 in this HMA failure situation. This is a very interesting concept in terms of giving a second-generation hypomethylating agent for patients with HMA failure. I will be discussing this in a little bit more detail when I go through the CC-486 trial.

We also discussed the potential for some forms of chemotherapy. We modeled these with low-dose cloforabine and low-dose cytarabine, and what we found is that, in general, this type of approach tends to be toxic and not very effective. However, we identified a subset of patients that are diploid, meaning a normal cytogenetic karyotype, who actually may benefit from a little bit more intense/more traditional type of chemotherapy strategies. This does not need to be done with cloforabine. This could probably be replaced with either fludarabine or cladribine at very low dosage with low-dose cytarabine, so there may be a subset of patients that benefit from chemotherapy.

Finally, we all know about the very important data from multiple groups looking at the genomic heterogeneity at the molecular level of this disease. I reminded the group at the meeting that a fraction of these patients, when these hypomethylating agents stop working, acquire a FLT3 mutation, or maybe now this mutation is detectable for the first time. Data has suggested that, for instance, adding a FLT3 inhibitor like sorafenib to azacitidine may result in a response. We all know now that a couple of weeks ago midostaurin, also a FLT3 inhibitor, was approved for patients with acute myelogenous leukemia.

Thinking about FLT3 targeted interventions in combination with a hypomethylating agent may be important for the group of maybe 15% to 20% of our patients with HMA failure disease. To end this part of the presentation, I want to summarize a couple of things. First of all, we discussed results with IDH2 and potentially IDH1 inhibition in myelodysplastic syndrome.

At the meeting, Dr. Stein showed the data from the group of patients with MDS that were treated in a large study of IDH2 inhibitor that we think may be an intervention that may become available to our patients pretty soon.

The data – very similar to what we saw in acute myelogenous leukemia with this compound known as AG-221 – shows not only an excellent toxicity profile, but also significant activity and prolonged duration of this response. We think that targeting IDH2, potentially IDH1, is going to be of benefit for our patients. There are now ongoing clinical trials targeting specifically the small but very important subset of patients with MDS with an IDH mutation. I think that is going to be one of the examples of targeted therapy for our patients.

We also discussed potential use of BCL-2 inhibitors with ABT-199. We discussed the potential trials that are now ongoing or ready to start combining azacitidine with ABT-199, both in the front line and in the relapsed context. We are very excited about this because we think that this could be a very powerful doublet that could have a role both in the front line and in the relapsed setting in combination with hypomethylating agents.

Finally, we discussed the potential role for incorporating drugs that inhibit immune checkpoints like PD-1, PD-L1, or CTLA-4 for patients with MDS.

I showed data from an initial trial with pembrolizumab presented at ASH last year. Single-agent pembrolizumab gave us a low response rate in patients with MDS. Interestingly, in the small exploratory trial, there was a subset of these patients with very prolonged survival, suggesting that potentially there is a role for these types of compounds in patients with MDS.

We also showed data with nivolumab and ipilimumab that target PD-1 and CTLA-4 in combination with azacitidine. Data from basket trial here at MD Anderson is showing clinical activity and potentially prolongation of survival with this combination. That said, these drugs have some toxicities that we are not very familiar with from the leukemia perspective. Of course, solid tumor oncologists are very familiar with this type of intervention. I think the main issue here is to understand who is the subset of patients that may benefit from this type of combination, and if there is a potential biomarker for activity, or toxicity, or both.

I concluded my talk with a lot of hope. I think the summary is that the development of this genomic data that we have been working on now for a decade is really helping us understand this disease and is going to allow us to better target our patients. I highlighted the data from the four phase 3 trials that are now ongoing for patients with myelodysplastic syndromes.  This included the CC-486 for front-line low-risk disease that I started the conversation with. I discussed the rigosertib Onconova INSPIRE trial for HMA failure, SGI-110 for HMA failure, and the study that just completed accrual with luspatercept for patients with refractory anemia and ring sideroblasts who are transfusion dependent after ESAs. This is very exciting that now, after a decade of no new drugs for patients with MDS, we have now four phase 3 trials, two of them almost completed. This is of great importance. We summarized the importance of oral hypomethylating agents, the lower attenuated dosing of either azacitidine or decitabine, and potential new strategies for low-risk HMA failure like OPN-305.We also summarized the new targeted intervention against FLT-3, IDH, potential use of BCL-2 inhibitors and new forms of hypomethylating agents that could include guadecitabine or SGI-110. I did not mention this in the talk but there is very exciting data with an oral decitabine form known as E7727.

That summary is quite intense and comprehensive, and I would like to finish my presentation today in terms of the CC-486 oral abstract presented at the Valencia meeting. This is data that we had already presented at ASH, and I think it is a very interesting concept per se. We have been working with CC-486 now for over a decade. First with the phase 0, 1 trials, we showed that CC-486 (or oral azacitidine) is clinically active in MDS.

What is interesting is that the pharmacokinetic profile of this compound tells us that we do not need very high concentrations of these drugs to give us clinical responses, and that may translate then into a better toxicity profile.

While we were discussing clinical development of this particular compound in MDS, a question arose of whether a drug like CC-486 could have a role in HMA failure. Of course, at the beginning we were a little bit hesitant because our experience says that if a patient has been treated with either azacitidine or decitabine, it is not very common that they will respond or benefit from switching to decitabine or azacitidine later on, if they have failed the contralateral agent at the beginning, because our data basically has not been very positive. With the help of Celgene, we went back in a retrospective fashion to three studies that we have conducted with oral azacitidine or CC-486. A very interesting result was that, first of all, CC-486 is well-tolerated. We expected that because we have quite a bit of data from a very long phase 1 trial. We also saw that, both in MDS and also integrally in patients with acute myelogenous leukemia, single-agent oral azacitidine or CC-486 had clinical activity in both entities in patients that had been previously treated with either decitabine, azacitidine, or both.

Again, this is not a prospective trial. What we did here was to mine the data from three clinical trials where patients had been treated with oral azacitidine that had previously failed either azacitidine, decitabine, or both. Again, the data clearly says that this intervention is safe, but most importantly, what we are seeing are both hematological and morphological responses with single-agent CC-486. In patients with myelodysplastic syndrome, we saw an average of maybe a third of the patients having a benefit in terms of hematological improvement or transfusion improvement. Interestingly, in patients with acute myelogenous leukemia, we also saw evidence of true responses with one patient with complete remission. Now, you may think this is anecdotal, but I do not think so. I think the CC-486 in this context as a single agent is giving us a hint that there may be a role for second-line hypomethylating agent type of approach in this context. I think the question is, why would this be the case? If you did not benefit from azacitidine or decitabine, why would you now all of the sudden benefit from an oral compound?

Well, the detail is that CC-486 is not given just for a few days every month. What we have noticed is that if we prolong exposure with oral azacitidine, then we get the results that we could see with standard azacitidine or decitabine, and I think potentially even higher response rates. What we are doing here is giving smaller concentrations by the oral route, but on a prolonged course, let's say 14 to 21 days.

This actually could be done once a day, twice a day, and this is the data that we presented on the trial. Perhaps a longer exposure, even if it is more concentrated, may result in clinical responses in these patients that are now not benefiting from azacitidine or decitabine.

I think this is very important not only from a clinical perspective, but also from a scientific perspective: trying to design what will be the best rational approach to use these compounds in second line. My conclusion is number one, they are safe, second, there is a hint of activity, but most importantly this serves as a platform for combination. For instance, one of the most exciting protocols that we have right now in this context is a prospective trial of oral azacitidine or CC-486 where we can add back an inhibitor of PD-L1. This, in some way, can confirm the data that we have generated as I mentioned earlier with immune checkpoint inhibitors with either pembrolizumab or with the other PD-1 and CTLA-4 inhibitors. This data in the HMA failure context goes in parallel to a very important randomized study (also sponsored by Celgene) where azacitidine is combined in a randomized way with an inhibitor of PD-L1. This study is now going on in both North America and the United States. In summary, I think that this is a very rich research clinical trial portfolio for our patients. We are trying to cover every spectrum of this disease, and I think that this effort in the next few years is going to really change the natural history of this disease. Again, I am very grateful for this opportunity to share this knowledge from the Valencia meeting. Thank you very much.

Last modified: August 10, 2017
Related Items by Author
Highlight from ASCO 2018: Omacetaxine Mepesuccinate after HMA Failure in High-risk MDS
Guillermo Garcia-Manero, MD
Clinical Commentaries published on July 31, 2018