Sabatolimab Demonstrates Promising Efficacy in Acute Myeloid Leukemia, Myelodysplastic Syndrome

Andrew M. Brunner, MD

In a phase 1b clinical trial (NCT03066648), the results of which were presented during the 2020 American Society of Hematology (ASH) Annual Meeting, investigators sought to determine the efficacy and safety of the novel TIM-3–targeting IgG4 antibody sabatolimab in combination with hypomethylating agents (HMAs) as treatment of patients with acute myeloid leukemia (AML), as well as high-risk myelodysplastic syndrome (MDS). The study also included patients with chronic myelomonocytic leukemia (CMML). For the HMA therapy, patients received either azacitidine or decitabine.

Overall, the study determined sabatolimab could be safely combined with either HMA. The combination regimen showed promising objective response rates of 64.1% among those with high-risk MDS, 41.0% among those with newly diagnosed AML, and 54.5% among those with CMML. The estimated 6-month duration of response rate was 83.9% in the high-risk MDS group and 78.8% in the newly diagnosed AML group, suggesting encouraging durability in these 2 populations.

The most common treatment-emergent adverse events (AEs) with the combination included thrombocytopenia, neutropenia, nausea, constipation, anemia, febrile neutropenia, and fatigue, which was consistent with what’s been observed for HMAs alone. These were predominantly low-grade, and discontinuation due to AEs was not common, occurring in 0% of patients in the MDS arm and 6.3% of the AML arm. Clinically significant treatment-related immune-mediated AEs were also observed infrequently, with the majority being low grade.

These data support the use of TIM-3 as a novel target for patients with newly diagnosed AML, high-risk MDS, and CMML, and serve as a basis for the future development of this therapeutic regimen. At this time, 3 clinical trials remain ongoing under the STIMULUS Clinical Trial Program (NCT03946670, NCT04266301, NCT04150029), exploring the role of sabatolimab in combination with HMAs in patients with higher-risk MDS and AML.

In an interview with Targeted Oncology, Andrew M. Brunner, MD, medical oncologist, Massachusetts General Hospital Cancer Center, discussed the updated findings from the phase 1b study of sabatolimab in combination with HMAs as treatment of patients with AML or high-risk MDS.

TARGETED ONCOLOGY: What was the rationale for this clinical trial?

Brunner: There are 2 main motivations to exploring this kind of a regimen. In part, we have been interested in how we can bring new immuno-oncology therapeutics into different cancers, including blood cancers, and I think that we have had some challenges in bringing traditional immuno-oncology therapies, like those against PD-1 or CTLA-4, into the treatment of our of malignancies like AML or MDS. We have had a few successes, and we know through transplant that immunotherapy can have a major role in controlling leukemia long term, but there's an ongoing search for new targets in this patient population. Targeting TIM-3, in part, is very appealing, related to what we understand about TIM-3 expression. For instance, in stem cell transplant, patients who relapse after transplant, their T cells upregulate TIM-3, and there's an exhaustion phenotype in those patients, suggesting that if we can mitigate that, that might be a way to maintain a graft-versus-leukemia or host-versus-leukemia effect in the future, as well as trying to understand how to use non–transplant-based immunotherapy for patients that might not otherwise be eligible.

I think the second rationale for exploring TIM-3–based therapies in AML, MDS, and other myeloid neoplasms relates to work that's been done to try to identify leukemic progenitor cells and markers that might distinguish leukemic progenitors from other hematopoietic progenitors. There have been a number of such expression profiles that have been suggested to distinguish what are the so-called leukemic stem cells that are the early leukemic cells that give rise to blood cancers, and how do we distinguish those from healthy hematopoietic stem cells? One such marker has been TIM-3, and I think that we are starting to increasingly understand or explore the biology of [how] TIM-3 expression is seen on both normal and malignant tissues, how they may play a role, and how that cell surface marker might play a role in maintaining a leukemic blast population, suggesting that if we can interfere with that, we may be able to advance the therapeutic options for our patients.

TARGETED ONCOLOGY: How was this study designed, and what did the patient population look like?

Brunner: We had a number of arms that were involved. There were arms both for patients in the upfront phase, as well as for the relapsed/refractory setting. In the abstract I presented, it was primarily [focused on] the upfront phase where people had not had prior HMA therapy. They could have had de novo AML. They could have had MDS that was high risk or very high risk, as well as CMML. They were treated either with standard backbone of HMA therapy, either 5 days of decitabine therapy or 7 days of azacitidine therapy, and then the study drug, sabatolimab, was administered on days 8 and 22 of the cycle for the every-2-week schedule, and then for the patients who were receiving the every-4-week schedule of drug, it was given on day 8 alone.

The patients were older, as you might expect for a study that's looking at patients with higher-risk MDS, AML, or CMML. The average age was around 70, and there was a slight male predominance. They were a pretty typical trial population. Most had an ECOG performance status of 1, and, as mentioned, they tended to have higher risk molecular and stratification features. Twenty-five out of 41 had high-risk MDS, and 16 out of 41 had very-high-risk MDS. For patients who had AML, they were split roughly, 44% with intermediate risk and 56% with adverse risk. If you looked at the mutation profiles, we don't have mutation data for every patient on study. We had it for 31 patients with MDS and 31 patients with AML with more to come, but among the patients with MDS, 12 of the 31 had TP53 mutations, and 7 of the 31 had ASXL1 mutations. This is a population that's enriched for higher-risk mutations that we hope to be able to explore more in the future.

TARGETED ONCOLOGY: Was there anything that surprised you or was interesting to see?

Brunner: There is 1 other interesting thing that we have seen with this agent, and I think it's particular to MDS, where 1 of the challenges with MDS is that patients have low blood counts. Low blood counts for patients with MDS bring them into clinic more often. Transfusions take time, and they are limiting in their quality of life. It seems like a small thing, but it really is a burden for the patient. I think that looking at hematologic parameters increasingly matters, and you can see that in AML. You can see that in our MDS trials, looking at transfusion-independence. You can see that in how the community as a whole is looking more at quality-of-life metrics for patients during treatment, and I think that in the experience with sabatolimab, it doesn't add a lot of severe myelosuppression. That's important when starting a patient, especially individuals who might be older or more frail. Those don't always go together, of course, but getting through the first couple months of therapy until they get a response is such a critical time. Having a therapy that can do that is important.

I also am encouraged because we've reported specifically on hematologic outcomes, and many patients had not just a marrow response, but also had a hematologic response. That is such an important thing for patients with MDS to see their blood counts and prove to limit the number of times they're coming into clinic. I think it's a clinical benefit that's hard to fully capture in trials, but I hope that that becomes a better-defined end point in clinical trials in general because it is so important to judging how their new therapy impacts patient experience.

TARGETED ONCOLOGY: What are the next steps for this research?

Brunner: I think that this study sought to make sure that this combination was safe. Some of the prior studies using immunotherapies in MDS and AML have had certain toxicities that make us cautious in an older patient population, so I think we all want to be cognizant of the toxicities of therapies that we have to offer in the field, in general, as we expand doublets and triplets for disease. We have to make sure that we can safely administer those, and so I think the experience on the trial was that sabatolimab could safely be administered with decitabine or azacitidine. We encountered complications of disease, including neutropenia, neutropenic, fever, anemia, thrombocytopenia, and those were similar to what we might expect, historically with azacitidine. We did see some oncologic toxicity, so immune-related AEs that are possibly related to the sabatolimab. That said, they tended to be low grade or easily treatable, and very few patients had to discontinue treatment for that reason.

Now that we've identified this as a combination that can be administered and really to a typical population that would receive HMA therapy in the outpatient setting. Their goal next is to try to validate the responses that we saw, which were encouraging, but we're using a single-arm study. There are a number of studies that are ongoing right now that are randomizing patients to receive azacitidine with or without sabatolimab. The enrollment on that's already gone rather briskly, so we're fairly encouraged that we'll have some results from that randomized study within the next year or so. I'm excited to see the results of that.

The other main thrust of a lot of our efforts are to better understand that we have a novel agent here. It's a novel mechanism of action. There are several ways that it could have a significant impact in leukemia care and in outcomes, and so I think that a lot of our next steps are to try to better characterize what those mechanisms are. How is sabatolimab inhibiting TIM-3 on T-cell populations? Is that distinct from how it interacts with leukemic precursor or blast cell populations? When it interacts with the blast cell population, was it actually due to the blast? Similarly, are we getting a T-cell response? Is that causing itself directed antileukemic effect?

TARGETED ONCOLOGY: What's your key take-home message?

Brunner: When I look at sabatolimab and think about its role, and in particular, when I think about MDS in general, I think that we have an opportunity to create new combinations that have a higher chance of getting a patient to remission. When you meet someone with MDS or with AML, most of the time, you're looking at monotherapy, which has a chance of getting a remission that's very low, and I think suddenly, in the field, we have a number of agents that all are showing promise when combined with azacitidine. More importantly, it seems like as these trials get larger, that studies are actually able to keep people on trial, and so with some of the earlier combinations, the toxicity of the combination really limited the ability to administer both together. However, suddenly we have therapeutics that we can combine safely for patients, including older patients, or more frail patients. I think that that's going to be critical because if we want to advance the care of patients, we have to be able to get drugs safely to them.

My takeaway from this is that this combination is safe, and I think that sabatolimab as a partner for therapeutics really is encouraging because we don't see a lot of severe immune-related toxicity. We're able to fairly safely add it on to an azacitidine backbone, and so as other doublet backbones are developed in AML, MDS, and CMML, I think the addition of sabatolimab at different phases of therapy is really promising. Hopefully, it allows us to explore the treatment of MDS and AML in a different way, so maybe instead of just starting somebody on a therapy and continuing it forever until they stop, we can actually use some more risk- or response-modified treatment algorithms. Maybe we frontload treatment in the very beginning to get a good deep response, and we can follow that by [minimal residual disease], then once patients have a response, maybe we switch to more of a maintenance type of regimen. Maybe we can track patients over time, so that if we start to see loss of that response, rather than wait for a full relapse, we can rotate our regimen or add a new targeted agent or rotate how we think about maintaining that response. There are a lot of ways that the field is about to adapt to a number of active agents in this setting. I'm excited to see if sabatolimab can play a role in that.

_Reference

_{Brunner AM, Esteve J, Porkka K, et al. Efficacy and Safety of Sabatolimab (MBG453) in Combination with Hypomethylating Agents (HMAs) in Patients with Acute Myeloid Leukemia (AML) and High-Risk Myelodysplastic Syndrome (HR-MDS): Updated Results from a Phase 1b Study. Presented at: 2020 ASH Annual Meeting & Exposition; December 5-8, 2020; Virtual. Abstract 657.}