Senior Member, Vice Chair
Leukemia and MDS
H. Lee Moffitt Cancer Center & Research Institute
Department of Oncologic Sciences
University of South Florida
Rami S. Komrokji, MD, has disclosed that he has received consulting fees from Celgene, Daiichi Sankyo, Jazz, Novartis, and Pfizer.
Several studies presented at the 2019 ASH annual meeting highlighted notable emerging therapeutic strategies for patients with myelodysplastic syndrome (MDS) or myeloproliferative neoplasms (MPNs). In this commentary, I briefly review data and share my thoughts on select studies of interest.
Luspatercept for Myelofibrosis-Associated Anemia
Management of anemia in patients with myelofibrosis is a clear area of unmet need. Anemia affects approximately 40% of patients with myelofibrosis, and 25% of anemic patients are transfusion dependent at the time of myelofibrosis diagnosis. In addition, standard therapies like JAK inhibitors can worsen anemia.
Luspatercept is a novel agent that was recently approved by the FDA for treating red blood cell (RBC) transfusion dependency in beta thalassemia. The agent is currently under review for the setting of myelofibrosis, with a decision expected early this year. Luspatercept is a recombinant fusion protein that binds the ligand that activates the TGF-beta family members involved in terminal erythroid differentiation.
The study presented at ASH was a first look at luspatercept activity in myelofibrosis. This was an open-label phase II trial in which multiple cohorts of patients with myelofibrosis and anemia—including cohorts receiving treatment with the JAK inhibitor ruxolitinib and others not on ruxolitinib—received luspatercept every 21 days for 168 days (N = 76). Transfusion-dependent and nontransfusion-dependent individuals were included in both cohorts; among transfusion-dependent patients, the transfusion burden was approximately 2-3 units per month, making them heavily transfusion dependent. The primary endpoints were transfusion independency (for those who were RBC transfusion dependent) and a hemoglobin increase of ≥ 1.5 g/dL from baseline over any 12-week period (for those who were RBC transfusion independent).
Of the patients who were RBC transfusion dependent, 32% who were on ruxolitinib and 10% who were not on ruxolitinib became RBC transfusion independent with luspatercept treatment for at least 12 consecutive weeks. The median duration of response for individuals receiving ruxolitinib was 39 weeks. I think this is a very encouraging result. Among patients who were not transfusion dependent at baseline, a ≥ 1.5 g/dL increase in hemoglobin was seen at every assessment in 21% of patients on ruxolitinib and 14% of those not on ruxolitinib.
Treatment was well tolerated, with no major adverse events unrelated to ruxolitinib. In my view, I think these data demonstrate an encouraging signal of anemia response with luspatercept in myelofibrosis, particularly among patients who were on ruxolitinib. Moving forward, studies that evaluate luspatercept for anemic patients on ruxolitinib who are transfusion dependent or transfusion independent will be of great value.
MEDALIST: Luspatercept for Anemia in Lower-Risk RBC Transfusion-Dependent MDS With Ring Sideroblasts
Anemia is the most common manifestation of MDS, especially in patients with lower-risk MDS, and many can become transfusion dependent. Treatment options are limited, with low response rates and duration of response. This unmet need is particularly clear among patients with ring sideroblasts; although the overall prognosis is good, transfusion dependence is common.
The MEDALIST trial was a randomized, double-blind, placebo-controlled phase III study that evaluated luspatercept for treating anemia in patients with lower-risk MDS with ring sideroblasts who were transfusion dependent. Patients were randomized to receive luspatercept or placebo every 3 weeks for at least 24 weeks. Transfusion independence at Weeks 24 and 48 was previously presented and the current analysis looked at extended benefits of treatment.
Previous analysis showed that this study met its primary endpoint (RBC transfusion independence for ≥ 8 weeks in treatment Weeks 1-24: luspatercept group, 37.9%; placebo group, 13.2%; P < .0001). The updated analysis reported rates of transfusion independency for periods of ≥ 8 weeks over the entire treatment period. This rate reached 47.7% with luspatercept vs 15.8% with placebo (P < .0001; OR: 5.978; 95% CI: 2.840-12.581). The investigators also evaluated transfusion independence with luspatercept according to baseline transfusion requirements. Luspatercept was associated with higher rates of transfusion independence vs placebo across the tiers of transfusion burden, although among individuals who were very heavily transfusion dependent (at least 6 units per 8 weeks) this comparison did not reach statistical significance. Of interest, many patients had more than one ≥ 8-week period of response, which is typically not seen with other agents. In total, 69.9% of luspatercept-treated patients had at least 2 separate periods of being transfusion independent for ≥ 8 weeks.
In terms of safety, no concerns emerged with longer follow up. Fatigue led to treatment discontinuation for 2 patients, although it is sometimes very difficult to tease out whether this is from disease background or is treatment related. Of importance, there was no signal of increased risk for disease progression to higher-risk MDS or acute myeloid leukemia (AML) with the treatment compared with placebo (approximately 5% in both arms). Typically, patients with ring sideroblasts are at lower risk for progression, and these numbers are consistent with what is known historically about this population.
In conclusion, the treatment benefit over longer follow up with luspatercept is clear in terms of transfusion independency rates and multiple responses. This drug is currently being reviewed by the FDA for approval in patients with MDS, and the phase III COMMANDS study in erythropoiesis-stimulating agent–naive, RBC transfusion–dependent patients with lower-risk MDS is ongoing.
APR-246 Plus Azacitidine in TP53-Mutated MDS and AML
At the ASH meeting, 2 studies with very similar designs evaluated the combination of APR-246 plus azacitidine in the setting of TP53-mutated MDS and AML—one from the United States and another from France.
As background, TP53 mutation in MDS and AML is associated with poor outcomes, including short duration of response to hypomethylating agents, poor response to intensive chemotherapy, a high risk of relapse after transplant, and a median survival in the range of 6 months. Therefore, there is a huge unmet need for improved treatment strategies for patients with TP53-mutated cancers.
APR-246 is a novel small molecule that can reactivate mutant p53, restoring its functionality and triggering apoptosis in cancer cells. Phase I studies successfully evaluated the combination of APR-246 plus azacitidine, and the phase II studies presented at ASH 2019 expanded upon those results.
In the French study, patients had intermediate-risk, high-risk, or very high–risk disease by IPSS (including patients with more than 30% bone marrow blasts). Of the 53 patients in the study, 19 had AML and, as expected, most patients had a complex karyotype and a large proportion were monosomal. Patients were treated with APR-246 for the first 4 days, followed by azacitidine on Days 4-10 of a 28-day cycle. A subset of patients were offered allogeneic stem cell transplantation after 3 or 6 cycles, whereas the others continued on the combination treatment.
The ORR in the intention-to-treat population was 55% (66% among evaluable patients). The CR rate was 49% among evaluable patients, with 9% having hematologic improvement. In this study, no responses were seen among patients with AML and more than 30% blasts, suggesting responses were concentrated among the patients with lower-risk MDS.
TP53 mutations were sequenced in 35 evaluable patients, and most were found to have a missense mutation. For patients in CR, many achieved complete cytogenetic response and negative TP53 by next-generation sequencing status. Overall, the median follow-up for this study was short and the median OS was not reached. Among the responders, all remained alive. Although it is early, these results are encouraging in terms of survival.
The adverse events profile was neurologically diverse, including ataxia, confusion, and facial numbness. All effects were reversible, but some patients required a dose reduction. The primary reason for discontinuation was related to disease progression rather than adverse events.
The US study was very comparable to the French study, with the combination of APR-246 and azacytidine on 28-day cycles for patients with higher-risk, TP53-mutated MDS and AML (N = 55). Of note, however, the study design allowed for up to 30% blasts, and patients with AML were required to be oligoblastic. Baseline characteristics were as expected, with 85% having a complex karyotype. Most patients had only one TP53 mutation; however, 38% had another somatic mutation.
The ORR in the US study was 71% among all patients, with 44% achieving a CR. When we look at the evaluable patients, the ORR was 87%, 53% achieved a CR, and 7% showed hematologic improvement. In terms of cytogenetic response, 59% had a PR or CR, and 44% of patients became TP53 negative by next-generation sequencing, where the cutoff is 5% to detect TP53.
Of importance, 22 patients discontinued treatment to proceed to allogenic stem cell transplant. The median duration of CR was > 7 months, with the understanding that many patients discontinued and went to transplant while in response. The investigators also found that isolated TP53 mutations predicted better responses. In addition, having more than 10% p53 immunohistochemistry at baseline was predictive of a response.
Similar to the French study, the adverse event profile included gastrointestinal effects, nausea and vomiting, and reversible neurologic toxicity. Looking at the 30-day and 60-day mortality, rates were low at 2% and 5%, respectively. Only 5% of the patients discontinued due to adverse events; none were APR-246 related.
OR by intention-to-treat analysis was 10.8 months. In subgroup analyses, responders and individuals that proceeded to transplant had better survival outcomes. Like the French study, the follow-up was short, but these survival data are encouraging.
Overall, these studies showed promising high responses with APR-246 plus azacitidine in patients with TP53 mutations, with encouraging early survival signals. These results have led to the initiation of a phase III clinical trial, where the patients will be randomized to receive azacitidine with or without APR-246 (NCT03745716).
Venetoclax With or Without Azacitidine for Relapsed/Refractory MDS
The next selection from the ASH meeting is a phase I study of venetoclax with or without azacitidine for the treatment of relapsed/refractory MDS. As background, venetoclax has been approved by the FDA for the treatment of newly diagnosed AML in patients who are ineligible for intensive chemotherapy. The next step is to evaluate venetoclax in higher-risk MDS, where typically only one half of patients respond to azacitidine alone and the average duration of response is 1 year.
The current study was an open-label, nonrandomized phase Ib study in which patients with relapsed/refractory MDS were treated with venetoclax with or without azacytidine. Venetoclax was administered at escalating doses of 100 mg/day, 200 mg/day, or 400 mg/day on Days 1-14. Primary endpoints included safety, maximum-tolerated dose, recommended phase II dose, and pharmacokinetics. The study included patients with relapsed/refractory MDS following initial CR, PR, or hematologic improvement with at least 4 cycles of azacitidine or decitabine within the last 5 years, bone marrow blasts < 20%, and Eastern Cooperative Oncology Group performance score ≤ 2.
After a median time on study of 8 months, deaths occurred in 33% of patients. It should be noted that the median survival for relapsed/refractory patients is expected to be 4-6 months after azanucleoside failure, with no current known approach to treatment with improvement in outcome. Intensive chemotherapy also offers low chances of response, in the range of 20% to 30%.
In terms of adverse events with venetoclax treatment, there were no dose-limiting toxicities. This is very similar to what was seen in AML. Based on this outcome, the 400-mg venetoclax dosing administered in a 2 weeks on, 2 weeks off fashion is expected to move forward in clinical trials.
As a single agent, the marrow CR rate with venetoclax alone was 8%, with no CRs observed. With combination therapy, the marrow CR rate was 32% and CR was achieved in 8% of patients; this was an interesting result among individuals who had been pretreated with hypomethylating agents and shows that adding venetoclax to azacitidine restores sensitivity in those patients. Blast reductions were also seen in the overall population, including complete responses. This endpoint can be meaningful for patients who proceed to allogenic stem cell transplant.
The median PFS for patients on venetoclax plus azacitidine was 9.1 months and the median OS was not reached. Although these results are encouraging, this is an early-phase trial, so they must be framed as such. Overall, the study does show the feasibility of this combination in patients with MDS.
Magrolimab Plus Azacitidine in MDS and AML Patients
Finally, I wish to highlight a phase Ib study evaluating the first-in-class anti-CD47 antibody, magrolimab. This ongoing study administered magrolimab plus azacitidine to patients with untreated intermediate-risk to very high–risk MDS or patients with untreated AML. Magrolimab was given as a priming/intrapatient dose escalation regimen together with azacitidine in a 28-day cycle. In this cohort, 19% had intermediate cytogenetic risk and 63% had poor risk, and TP53 mutations were present in 28% of individuals.
The results were promising, with ORRs for combination therapy in patients with MDS and AML of 100% and 69%, respectively. After a median follow-up of 5-6 months, no median duration of response or OS was reached for either patient population.
What are your thoughts on the new data in MDS and MPNs presented at ASH 2019? Answer the polling question on the screen and join the conversation in the discussion box below!
You are accessing CCO's educational content today as a Guest user.
If you would like to continue with free, full access to the CCO Web sites, including free CME/CE credits, please click the button below.