The addition of nivolumab to azacitidine led to significantly better overall response rate and overall survival compared with previous hypomethylating agent-based regimens in patients with relapsed acute myeloid leukemia (AML), according to the results of a recent phase 2 study.
The addition of nivolumab to azacitidine led to significantly better overall response rate (ORR) and overall survival (OS) compared with previous hypomethylating agent (HMA)-based regimens in patients with relapsed acute myeloid leukemia (AML), according to the results of a recent phase 2 study.
Since the introduction of immunotherapy, many patients experienced longer progression-free survival and OS. Immune checkpoint inhibitors, in particular, have been associated with high clinical efficacy in many types of solid and hematologic tumors. In relapsed AML, the use of checkpoint inhibitors alone offers minimal benefits; however, information regarding checkpoint inhibitors combined with chemotherapy is lacking.
Azacitidine is an HMA approved for use in myelodysplastic syndrome and commonly used for the treatment of AML. Although azacitidine promotes anti-tumor signaling through upregulating interferon-gamma pathway genes and HLA class 1 antigens, it also increases the expression of PD-1 and PD-L1 on these tumors. Nivolumab is a PD-1 inhibitor that can potentially counteract the mechanism of resistance that azacitidine causes. Researchers in a phase 2 study analyzed the safety and efficacy of nivolumab in combination with azacitidine in patients with relapsed AML.
In this study, 70 patients with relapsed AML received azacitidine 75 mg/m2 and nivolumab 3 mg/kg. ORR was 33% with 15 complete remission (CR)/CR with incomplete recovery of counts (CRi), 1 partial response, and 1 hematologic improvement. Factors that were associated with improved response rate were patients having received no prior HMA-based therapy, pretherapy bone marrow blast <20%, circulating white blood cells ≤10,000/μL, presence of ASXL1 mutation, and pretherapy bone marrow aspirate CD3-positive.
Median OS was 6.3 months for all patients. Patients who responded to the treatment had significantly better OS compared with non-responders (16.16 months vs 4.1 months; P < .0001). OS was similar among patients who had CR/CRi/partial response, hematologic improvement, and stable disease (17.1 months vs 11.9 months vs 16.2 months; P = .8). In comparison to other HMA-based regimens, HMA with immunotherapy had better ORR (33% vs 20%), OS (6.3 months vs 4.6 months), and event-free survival (EFS, 4.2 months vs 2.2 months). Patients who were defined as “first salvage” had even greater OS and EFS improvements on an immunotherapy-based regimen compared with other HMA-based regimens (OS, 10.6 months vs 5.3 months; EFS, 6.8 months vs 2.7 months).
One of the identified predictive factors of response was CD3-postitive and CD8-postitive cells in pretherapy bone marrow aspirates, with cutoffs at 13.2% and 4.01%, respectively. For CD3-postitive, ORR was 56% for patients with bone marrow CD3-positive ≥13.2%, compared with 23% in patients with bone marrow CD3-postitive <13.2%. Cutoff at >13.2% for CD3-postitive had a sensitivity of 74% and a specificity at 65%. Similar sensitivity and specificity was found for CD8-postitive at a cutoff of >4.01%.
Sixteen patients developed immune-related adverse events (irAEs); 9 were pneumonitis, 6 were nephritis, 3 were skin rashes, and 2 were transaminitis. Two irAEs resulted in deaths, one from pneumonitis and the other from hemorrhage. Most irAEs occurred within 8 weeks after starting nivolumab treatment.
Nivolumab in combination with azacitidine was able to significantly improve ORR and OS in patients with relapsed AML. Further phase 2 and phase 3 studies revolving around these regimens are currently being conducted (NCT03092674, NCT02775903).
Daver N, Garcia-Manero G, Basu S, et al. Efficacy, safety, and biomarkers of response to azacitidine and nivolumab in relapsed/refractory acute myeloid leukemia: a randomized, open-label, phase 2 study [published online November 8, 2018]. Cancer Discov. doi: 0.1158/2159-8290.CD-18-0774.