Magrolimab Development Continues for the Treatment of AML and MDS

Naval G. Daver, MD

There are 4 types of CD47 antibodies currently available in the clinic for patients with acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS). According to Naval G. Daver, MD, a novel option magrolimab, could be the fifth CD47 antibody once data supporting the drug mature.

“[Magrolimab] is a CD47 monoclonal antibody that has been combined with azacitidine [Vidaza] in the frontline setting and in a now completed a phase 1b study, which had 2 parts. One was for high-risk MDS patients….The other part was for TP53-mutated frontline AML,” stated Daver, an associate professor in the Department of Leukemia at The University of Texas MD Anderson Cancer Center, in an interview with Targeted Oncology^TM.

Daver explained how first, a prospective, phase 1 trial (NCT03248479) evaluated single-agent magrolimab in patients with relapsed/refractory AML. Due to the overall positive findings from this trial, a new phase 3 ENHANCE trial (NCT04313881) has begun enrolling patients.

In the ENHANCE trial, the combination of magrolimab plus azacitidine will be examined vs azacitidine plus placebo in treatment-naïve patients with higher-risk MDS. Investigators aim to evaluate the efficacy of magrolimab in combination with azacitidine in this patient population as measured by complete remission and overall survival.

During a panel at the 10th Annual Society of Hematologic Malignancies (SOHO), experts, including Daver, discussed on how to optimize frontline hypomethylating agent (HMA) and venetoclax (Venclexta).

Further, though the combination has only been approved for 4 years, research aims to examine the different ways to move beyond this treatment approach for this patient population as more work is always needed to improve better outcomes for patients.

In the interview, Daver discussed the approved and investigation CD47 monoclonal antibodies for patients with AML and MDS. He also provided a prediction on where research for AML and MDS is headed.

Targeted Oncology: Based on data from preclinical and clinical studies, which in type CD47 agents have shown promise as targeted therapies?

There are 4 different CD47 antibodies in clinic for AML and MDS. The one that has the most advanced clinical data, most mature so far, is a drug called magrolimab. This is a CD47 monoclonal antibody that has been combined with azacitidine in the frontline setting and in a now completed a phase 1b study, which had 2 parts. One was for high-risk MDS patients, including intermediate, high, or very high IPSS patients. The other part was for TP53-mutated frontline AML.

Overall, the data for the MDS patients was that we had 95 patients with this higher risk MDS, and the response rate is encouraging, 75% with a true CR rate of about 33% or 34%. What's interesting is in the subsets that included TP53 mutated, which made up about 25 of those 95 patients, the median survival is quite promising at greater than 16 months. Historically, pretty much in any other study that we have seen in MDS TP53, the survival has been between 9 and 12 months.

This is single arm, but this data looks quite promising. In the non-TP53 subgroup, the median survival has not been reached, with a median follow-up of about 18 months. Overall, it's a single-arm study, so it's hard to know for sure, but this data does look encouraging. This has led to a randomized phase 3 study now ongoing, called the ENHANCE study, which in fact just completed accrual a couple of weeks ago. We're hoping to start seeing some of the primary end point data early next year.

Then the AML study was looking at the most difficult population of AML called TP53-mutated AML patients. In these patients, historically, the response rate is about 30%-40%, with survival about 6-8 months. This is with azacitidine alone, azacitidine/ venetoclax, intensive chemotherapy, low dose cytarabine. With the azacitidine and magrolimab doublet, we're seeing a response rate of close to about 49% or 50% and a true complete response [CR] rate of 34%, which are better than what we have seen in the past with other therapies. Not hugely higher, but they looked to be slightly better response rates.

More importantly, the median survival was around 11 months. Even though that's not very high on its own, it's better than what we have ever seen with TP53. This has also led to a randomized study of azacitidine and magrolimab vs investigator choice, either intensive chemotherapy or azacitidine and venetoclax, and that study is ongoing.

What other CD47 inhibitors are available in this space?

There are other CD47’s which are very early and ALX148 [evorpacept] is one of them. One of the potential benefits of evorpacept is it may cause less anemia. There was very early clinical data presented of about 12 patients. Some responses we're seeing, and it is very hard to know at this time if this is going to be in the same efficacy range as with the azacitidine and magrolimab, or different, and we hope for updates.

Then there are others that we don't have much clinical data yet they're ongoing in phase 1 studies such as lemzoparlimab [also known as TJ011133 or TJC4], TTI622, but hopefully, either by the end of this year or early next year, we will see more information from those. It's a very active field and a major target and pathway for clinical development in acute myeloid leukemia and MDS.

What were the key takeaways from your presentation given this year at 10th Annual Meeting of the Society of Hematologic Oncology?

There was a lot of discussion and focus on how to optimize frontline hypomethylating agent and venetoclax and moving beyond HMA and venetoclax. HMA and venetoclax has not been approved that long, it's only been around 4 years. But at MD Anderson Cancer Center, we were involved in and enrolled a large number of patients and led the studies. We realize that even though it's a good breakthrough, at the end of the day, the 3-year survival is about 35%. That is better than 10%, which it was before that for older unfit patients. But still a lot of work has to be done.

What has research focused on over the past few years?

Our effort over the last 2 or 3 years has been to identify molecular biomarkers of resistance. We have found that certain molecular groups like FLT3, TP53, RAS, KIT mutation, don't seem to have the same durability and survival with HMA and venetoclax as other molecular groups like NPM1, IDH1, IDH2, which do well.

Now, we've been trying to design ways to add targeted therapies in patients who don't have as good outcomes but have a particular targetable mutation. Sort of a rationalized personalized improvement of the HMA event. This is where several combos of HMA/venetoclax with second generation FLT3 inhibitor or HMA/venetoclax with an IDH inhibitor, HMA/venetoclax with CD47 for TP53, and coming soon, HMA/venetoclax with menin inhibitors.

Of course, the inherent issue with combining drugs is you have to be aware of and adjust for myelosuppression toxicities. With a lot of these combinations, the response rates are high with HMA and FLT3 and IDH, but we do see that there's an additional myelosuppression. Now we're cutting down the duration of them. FLT3, because of their response being so high, the synergy is good, and you may not need full doses of them. Now we're getting to a point where we are seeing synergy with less myelosuppression. But again, this has to be more mature before it can be adopted widely, but there was a lot of discussion about how we can optimize these doublets and triplets, and what is the benefit of using them upfront in combination vs sequencing. They were all great questions for clinical trials that are starting and will be coming.

What other panels stood out to you at SOHO 2022?

There was a very nice debate on MRD. The question was, is MRD the most critical marker for selecting subsequent therapy or prognosis? Is it the baseline molecular cytogenetic profile of the patient and is that information actually sufficient? Once you have that, do you benefit more from MRD data or not? It's quite a provocative question.

Going into the debate, I think a lot of us would say that MRD is the king and trumps all but both presenters gave very astute and useful arguments from both sides. What is true is that although MRD negativity is a good factor and is usually just associated with sensitivity or not resistance to chemotherapy, it is not really the eventual decision maker for most patients. Case in point, if you have a TP53-mutated AML, and even if they achieve MRD negativity, we would still consider a patient high-risk and go to transplant because we know that the TP53 biology is present in the leukemia stem cells and may be present at a very low level below the detection threshold. We don't feel comfortable that MRD eradication is sufficient to avoid it.

I think the point is that MRD may be useful in certain situations, NPM1, maybe FLT3 where we have effective therapies and presence of MRD. Despite that, effective therapy is poor. Maybe you need to do a transplant earlier. I escalate therapy. But for a large majority, the baseline molecular cytogenetic features, if you can really do them in a comprehensive fashion, will give you the similar predictive value as you would have gotten from MRD. And the advantages, you could get that upfront in the beginning. I think it really speaks to the need to enhance our comprehensive genomic cytogenetic potentially the near future RNA expression data. And hopefully then we can really segregate and triage the appropriate prognostic and therapeutic selection for AML.

What unmet needs still exist in this space?

The whole space is an unmet need because in general for AML, the frontline survival is about 35%-40% if you take all patients. It's a little better in the younger patients, those who are below 65, but the 5-year survival is closer to about 60%-65%. For those who are above 65, it's poor at about 20%-25%. In general, we are looking for new strategies for all of them. The most difficult group is probably in AML is TP53. With these, the median survival is 6-8 months across pretty much every therapy we have tried in the last decade, and the 5-year survival is less than 10%. There's usually a small subset of patients who if they can make a stem cell transplant with some degree of Disease Control, we could hope for 30%-35% long term survival.

Our goal is to get these people quickly to transplant if we can. The other big unmet need, both from a diagnostic perspective and therapeutic perspective is MRD [minimal residual disease]. We are getting better at identifying MRD with more sensitive techniques, especially molecular techniques including In single cell sequencing, potential duplex sequencing, and we're now able to probably get down to 10 raise to minus five or so routinely with some of these technologies. These have not yet been validated in prospective trials, hopefully they will be and then could be used prospectively potentially as Registration Status strategies.

The other big problem is we don't have great MRD eradicating strategies yet in the acute myeloid leukemia like we have in acute myeloid leukemia, lymphoma, and others. There has not yet been a chimeric antigen receptor T or bispecific that has shown that degree of activity, but there are many of these now in early stage or going into clinic. Hopefully, we will have some MRD based therapies. These are probably the 2 biggest areas of unmet need.

Then of course, in any relapse disease, the outcome is still poor. We look for targetable mutations, FLT3 IDH1, IDH2, MLL rearrangement, and for this, we do have targeted therapies with FLT3 inhibitors, IDH inhibitors, Menin inhibitors, but for the others, we still are looking at clinical trial options and nothing has really made a breakthrough yet.