Survival for patients with acute myeloid leukemia (AML) and myelodysplastic syndrome (MDS) remains poor, and although immunotherapy has been positioned as a holy grail, it would be preemptive to predict its future based on the number of small studies that have been performed to date, according to Amer Zeidan, MD, MHS.
Nonetheless, one of the first studies that showed the potential for checkpoint inhibition in hematologic malignances was a phase 1/1b study that evaluated ipilimumab (Yervoy) after allogeneic stem cell transplant. In the study, ipilimumab was administered at a dose of 3 mg/kg and 10 mg/kg every 3 weeks. The results demonstrated efficacy in patients who received the 10 mg/kg dose. Specifically, 5 of 13 patients with heavily pretreated AML achieved a complete response (CR).
Ipilimumab has also demonstrated activity in patients with relapsed/refractory MDS in a single-arm study of 29 patients. Although the marrow CR rate was only 3.4%, 7 patients experienced prolonged stable disease for 46 weeks or more, including 3 patients with stable disease surpassing 1 year. Moreover, the median survival was 9.8 months (295 days; 95% CI, 240-671+).
In correlative analysis, we observed that patients who had increased expression of the costimulatory marker ICOS seemed to have better disease stabilization, so the direction is clearly headed toward trying to select patients using biomarker-driven strategies, Zeidan, an associate professor of medicine in the Department of Internal Medicine and Section of Hematology at Yale University School of Medicine, Yale Cancer Center, said in a presentation during the5th Annual International Congress on Immunotherapies in Cancer.
In another phase 2 study that evaluated the combination of nivolumab (Opdivo) and azacitidine vs ipilimumab and azacitidine in MDS, a higher response rate was observed with either combination compared with ipilimumab alone, at 70% and 62% vs 30%, respectively. However, the median overall survival (OS) was similar, at 11.8 months, not reached, and 8.5 months, respectively, said Zeiden.
Pembrolizumab (Keytruda) is also being subject to research in myeloid malignancies. Specifically, in a phase 1b study in combination with entinostat in MDS after failure on hypomethylating agents (HMAs). Data from the study have yet to read out, but are highly anticipated, said Zeiden.
Findings from a phase 2 study (NCT02775903) however stunted some of the excitement that had been generated with immunotherapy, putting into perspective the work that had been done to date. When the combination of a checkpoint inhibitor and an HMA, specifically durvalumab (Imfinzi) and azacitidine, was taken into a randomized trial vs azacitidine alone in patients with high-risk MDS and older AML, no difference was seen in progression-free survival (PFS) or OS.
Bispecific antibodies are another treatment class under investigation as a potential avenue forward for immunotherapy. One such antibody is sabatolimab, which targets IgG4 and TIM-3, and is the focus of several ongoing studies in MDS and AML. Specifically, the phase 2 STIMULUS-MDS1 (NCT03946670) and phase 3 STIMULUS-MDS2 (NCT04266301) trials in MDS and phase 2 STIMULUS-AML1 trial (NCT04150029) in AML.
Another path that will be explored is that of checkpoint inhibition plus chemotherapy, explained Zeiden. In a phase 2 study, the combination of pembrolizumab and 7+3 chemotherapy will be evaluated as frontline therapy in fit patients with AML.
The combination of azacitidine and venetoclax (Venclexta) has become the standard of care for older patients with AML, and preclinical evidence suggests that the BCL-2 inhibitor can augment the antitumor response of PD-L1 inhibitors.
As such, investigators have launched the phase 2 BLAST AML 2 study in which unfit patients with AML will be randomized to azacitidine plus venetoclax vs azacitidine/venetoclax plus pembrolizumab as frontline therapy.
Anti-CD47 antibodies are also under study and have shown promising, though early, activity in AML and MDS. For example, in combination with azacitidine, magrolimab has shown objective responses exceeding 60% in untreated AML and 90% in untreated MDS, with CR rates of 41% and 50%, respectively.
Importantly, a lot of the responses seem to occur in patients who have TP53 mutations, which is one of the highest areas of unmet need in AML and MDS, because those patients do very poorly with conventional treatment, said Zeidan.
Although magrolimab will move forward in development, Zeidan cautioned that the antitumor effects of anti-CD47 antibodies may not be class specific. For example, in a phase 1 study, CC-90002 failed to demonstrate any benefit in patients with relapsed/refractory AML and higher-risk MDS.
In conclusion, Zeidan stated, Many of the studies that have been conducted are single-arm trials with small sample sizes. [However,] we are doing more and more randomized studies using novel inhibitors against TIM-3 and CD47. The field is definitely exciting for us, and we are hoping to see some clinical activity for our patients soon.
Zeidan A. Immunotherapy for treatment of myeloid malignancies: will it fill the promise? Presented at: 5thAnnualInternational Congress on Immunotherapies in Cancer; December 12, 2020; virtual. gotoper.com/go/ICIC20Virtual