DSP-5336 Is Safe, Generates Responses in KMT2A+ or NPM1+ R/R AML

Naval G. Daver, MD, discusses efficacy and safety data from the first-in-human phase 1/2 study investigating DSP-5336 for relapsed or refractory acute leukemia.

Early efficacy and safety data from the first-in-human phase 1/2 DSP-5336-101 study (NCT04988555) warrant further investigation of the menin inhibitor DSP-5336 for the treatment of patients with relapsed/refractory acute myeloid leukemia harboring KMT2A rearrangements or NPM1 mutations, according to Naval Daver, MD, who added that future studies could examine the agent as monotherapy and as part of combination therapies.

Updated results from the dose-escalation portion of the study presented at the 2024 EHA Congress demonstrated that patients with relapsed/refractory AML harboring KMT2A rearrangements or NPM1 mutations treated with single-agent DSP-5335 at a dose of 140 mg twice per day or higher(n = 21) achieved an objective response rate (ORR) of 57%, including a composite complete remission (CR) rate of 33% and a CR/CR with partial hematologic recovery (CRh) rate of 24%. In the KMT2A-rearranged population (n = 12) and NPM1-mutated population (n = 9), the respective ORRs were 67% and 44%.

Regarding safety, no dose-limiting toxicities, treatment-related deaths, or treatment-related discontinuations were reported. Additionally, no evidence of drug-related cardiac toxicity was observed. Differentiation syndrome occurred in 3 patients (5.7%); however, no deaths or permanent treatment discontinuation occurred due to differentiation syndrome, and no prophylaxis for differentiation syndrome was used during the study.

“So far, the safety profile has been excellent, and we're now getting to doses where we're seeing a good ORR. We're starting to think about when we will decide which dose level should be the recommended phase 2 dose for further expansion,” Daver explained.

In an interview with OncLive®, Daver detailed the background for the menin inhibitor DSP-5336, expanded on the efficacy and safety data reported from DSP-5336-101, and explained future research directions for the agent, including its potential for use in combination therapies.

Daver is a professor and director of the Leukemia Research Alliance Program in the Department of Leukemia at The University of Texas MD Anderson Cancer Center in Houston.

OncLive: What role could menin inhibitors play in the AML treatment paradigm?

Daver: There has been a lot of interest in targeting [the] upregulation of particular gene signatures, such as MEIS1 and HOXA. In patients [whose disease features the upregulation of these gene signatures], we are now [evaluating] therapies called menin inhibitors. There is an epigenetic pathway that causes upregulation of certain metabolic enzymes, such as MEIS1 and HOXA, and when these are upregulated, they'll result in a differentiation blockade, [resulting] in blasts that are not able to mature to normal, healthy neutrophil and monocyte components.

Menin inhibitors seem to disrupt this process with reduction in the production of HOXA and MEIS1, resulting in normal differentiation with the production of normal, healthy neutrophils and monocytes. This is called the differentiation process, which eventually results in achievement of response and remission. There are many menin inhibitors that are now in clinical development.

DSP-5336 is one of these menin inhibitors [under development]. We are seeing differences in menin inhibitors in terms of both safety and efficacy. Even though there are multiple [menin inhibitors being studied], there will be a need for more safe and effective menin inhibitors, both the single agents and more importantly, in combinations, where some of the safety issues [with some agents] could become further exaggerated. This is why we are interested in developing the DSP-5336. We want to better understand its differentiation from other menin inhibitors in terms of safety, efficacy, its potential for [use in] combinations, and its ability to overcome some of the known resistance mutations that have been seen with other menin inhibitors.

What patient population was enrolled in the DSP-5336-101 study?

These were [patients with] relapsed/refractory AML. Initially in the first few dose-escalation [cohorts], we allowed all patients, regardless of their molecular and cytogenic profile to see if there could be activity [with DSP-5336] outside of the known target menin populations, such as those [with] KMT2A-rearranged and NPM1-mutated [disease]. The key thing to highlight is that these were heavily treated patients. The median lines of salvage therapy was 3 [range, 1-9], and more importantly, approximately 90% of patients [received] prior venetoclax [Venclexta]. This is important because [past] studies have shown that exposure to prior venetoclax resulted in a significantly lower ORR to subsequent therapies with poor survival.

Without menin inhibitors, the expected ORR in this [heavily pretreated] population would be approximately 10% to 15% with limited survival. We enriched [the study population] for [patients harboring] NPM1 mutations or KMT2A rearrangements, and this [constituted] approximately 70% of the study population. Moving to the higher dose levels, we're focusing on efficacy and planning expansion, where we're exclusively [enrolling] patients with KMT2A rearrangements or NPM1 mutations rather than all-comers.

What safety data were presented at the 2024 EHA Congress?

Safety is the most interesting and encouraging aspect of the DSP-5336 data. We're familiar that there are multiple menin inhibitors that have had data presented over the past year or so. We have noticed that despite good efficacy [for these other menin inhibitors], there are safety signals emerging, and these [signals] are different between the different menin inhibitors.

For example, with revumenib [SNDX-5613], we see a prolonged QTc. In general, this is manageable, but we have seen some patients with a significantly prolonged QTc, especially if they have concomitant QTc-prolonging medications. This could become a hurdle [in the] broad adoption and use of this [agent], especially when you get outside of academic centers that are good at monitoring for a prolonged QTc.

[Other menin inhibitors] like ziftomenib and JNJ-75276617 do not seem to have a significant cardiac QTc signal, but we've seen significant differentiation syndrome—not just hematological, but clinical differentiation—which in some cases has been severe and proliferative, even with mortalities.

The nice thing with DSP-5336 is we have not seen any significant cardiac signals. Of the 57 patients on the study, there were only 2 who had grade 3 or higher QTc prolongation. For 1 patient, this was taught to have no relation to the DSP-5336, as he had other arrhythmias and electrolyte changes that were causing a prolonged QTc. For [the other] patient, there was a possible relationship of the grade 3 QTc prolongation with the DSP-5336; however, this patient also had hypokalemia and was using medications that could increase it. [This patient’s QTc] reverted back to normal without clinical consequence. [Regarding cardiac-related toxicities], DSP-5336 has a good profile.

When you look at differentiation syndrome, this has been surprising in a positive way [with an any-grade rate of 5.7%]. We have not seen significant clinical differentiation syndrome requiring an intensive care unit stay, drug discontinuation, mortality, or any of the [issues] that we observe with some of the other menin inhibitors. We also are not doing any prophylaxis [for differentiation syndrome], and despite [observing] clinical responses and efficacy, we're not seeing the clinical toxicity that has been seen with severe differentiation syndrome. Again, [the known safety is based off] small numbers [of patients], and as we enroll, this profile may evolve. However, so far, the safety looks good.

What initial efficacy data were reported?

From an efficacy perspective, we focused on the target populations where we expect responses: [patients with] KMT2A rearrangements and NPM1 mutations. We looked at efficacy in the dose range of 140 mg to 300 mg [twice per day], which is the predicted biologically active range for this agent. Overall, we had 21 patients who had a KMT2A rearrangement or NPM1 mutation who were treated at a dose level of 140 mg twice per day or higher. The ORR was 57%, which is quite robust. When you look at the CR/CRh [rate], which is what the FDA often focuses on in patients with relapsed/refractory AML without curative intent, this was 24%, which is quite good and in line with other agents such as IDH inhibitors.

The current ORR is looking quite robust. [These data were with DSP-5336] a single agent, and there are efforts to move it into combinations, where the ORR would likely be further boosted. To put this in context, historically, if you took a patient with relapsed/refractory AML on their third or fourth salvage [regimen], whose disease harbors KMT2A rearrangements and had prior venetoclax exposure, the expectation for a response would be about 15% to 20%. Seeing a 57% ORR with a single agent is encouraging.

What are the potential clinical implications of DSP-5336?

From a regulatory standpoint and getting these drugs available to patients outside of large academic centers currently running trials like ours, we're looking at the potential for developing a larger phase 2 trial evaluating DSP-5336 as a single agent in patients with relapsed/refractory AML. We haven't decided exactly the target population, but it's probably going to be a mix of different aberrations that are known to upregulate HOXA and MEIS1, which are the key biomarkers for menin sensitivity. That will likely be the initial direction to see if we could reproduce the good CR/CRh rate and ORR, and maintain the encouraging safety in terms of both differentiation syndrome and cardiac issues.

Long term, I suspect that more combination approaches will be [evaluated]. We're already looking at starting to develop combinations with azacitidine and venetoclax, given that azacitidine plus venetoclax is a standard frontline therapy for older patients with AML. The question is: can you add DSP-5336 in patients who have target aberration and further improve the ORR? This is also being investigated with FLT3 inhibitors. We're going to see some of these combination trials start and see if these further boost the ORR, depth of response, durability of response, and outcomes.

Reference

  1. Daver N, Zeidner J, Yuda J, et al. First-in-human phase 1/2 study of the menin-MLL inhibitor DSP-5336 in patients with relapsed or refractory acute leukemia: updated results from dose escalation. Presented at: 2024 EHA Congress; June 5-14, 2024; Madrid, Spain. Abstract S132.