Nivolumab Combo Shows Early Activity in Relapsed/Refractory AML

Partner | Cancer Centers | <b>MD Anderson</b>

The combination of nivolumab and azacitidine demonstrated favorable responses and overall survival in patients with relapsed/refractory acute myeloid leukemia.

Naval Daver, MD

The combination of nivolumab (Opdivo) and azacitidine demonstrated favorable responses and overall survival (OS) in patients with relapsed/refractory acute myeloid leukemia (AML), according to preliminary findings of a phase II study published in Cancer Discovery.1,2

Results showed that the overall response rate (ORR) was 33%, including a complete remission (CR)/CR with insufficient recovery of counts (CRi) of 22% (4 CR and 11 CRi), 1 partial response (PR), and 7 patients with hematologic improvement (HI) >6 months. The ORRs were 58% and 22% in hypomethylating agent (HMA)—naïve and HMA-pretreated patients, respectively. Six (9%) patients had stable disease (SD) >6 months.

“Over the last decade, 6 PD-1, PD-L2 and CTLA-4 antibodies have been approved for over 25 indications in 10 tumor types in the United States and Europe,” lead author Naval Daver, MD, an associate professor of leukemia at The University of Texas MD Anderson Cancer Center, said in a statement. “However, single-agent PD-1 antibodies have shown little effect in patients with relapsed AML or high-risk [myelodysplastic syndromes]. This study was designed to assess whether the addition of nivolumab to azacitidine was safe and effective.”

Preclinical data have shown that blocking PD-1/PD-L1 pathways enhance anti-leukemic responses, and azacitidine upregulates PD-1 as well as interferon-γ signaling.

In the single-center, open-label, nonrandomized study, 70 patients with relapsed/refractory AML were treated with azacitidine at 75 mg/m2 intravenously (IV) or subcutaneously on days 1 to 7 plus nivolumab IV at 3 mg/kg on days 1 and 14 every 4 to 6 weeks. Patients received a median of 2 prior therapies, and 45 (65%) patients received prior treatment with an HMA. The median age was 70.

To be eligible for enrollment, patients ≥18 years of age had to have failed prior therapy; had an ECOG performance status ≤2; and specified serum creatinine, bilirubin, and transaminase levels. Patients were excluded if they had a history of a systemic autoimmune condition, severe interstitial lung disease or active pneumonitis, prior solid organ allograft, symptomatic central nervous system leukemia, and any other uncontrolled disease.

Safety and ORR served as the study’s primary endpoints; secondary endpoints included OS, event-free-survival (EFS), and duration of response (DOR).

Additional data showed that the median OS was 6.3 months and 10.6 months in the all-savage and first-salvage settings, respectively. The median EFS was 4.5 months among responders and those with SD (n = 29); the DOR among responders was 5.2 months.

There was a significant improvement in OS in patients who achieved a CR/CRi/PR/HI/SD (n = 29; 42%) versus those who did not respond (n = 41; 58%), without censoring for autologous stem cell transplant (16.16-4.1; P <.0001). After censoring for transplant, OS was 17.1 months, 11.9 months, and 16.2 months in patients who achieved a CR/CRi/PR, HI, and SD, respectively (P = .8).

“In addition, bone marrow samples taken prior to treatment indicated a higher frequency of pretherapy bone marrow (BM) CD3 and CD8 cells predicted for response to therapy,” said Daver. “In particular, CD3 appeared to have a high sensitivity and specificity rate for predicting response, indicating it might serve as a reliable biomarker for selecting patients for this combination therapy.”

Univariate analysis revealed a greater trend toward improved ORR in patients with no prior HMA-based therapy, pretherapy BM blast <20%, circulating white blood cells <10,000/L, the presence of an ASXL1 mutation, and pretherapy BM aspirate CD3 positivity. Factors that contributed to an OS improvement were response or SD to therapy, salvage 1 status, and the presence of an ASXL1 mutation.

Multivariate analysis was done in 47 patients with evaluable pretherapy BM CD3-positivity flow cytometry data. No prior HMA (P = .059), higher pretherapy BM aspirate CD3-positivity (P = .065), and the presence of an ASXL1 mutation (P = .053) also showed a trend toward improved ORR, though this was not statistically significant.

Grade 3/4 immune-related adverse events (irAEs) occurred in 11% of patients (n = 8). The most common grade 2 through 4 AEs were pneumonitis (n = 9), nephritis (n = 6), skin rash (n = 3), and transaminitis (n = 2), most of which were effectively managed with steroids. Treatment discontinuation with nivolumab occurred in 13% of patients as a result of grade 3/4 irAEs. There were 2 irAE-related deaths.

At a median follow-up of 21.4 months, 57 patients (81%) had died (95% CI, 14.8—not estimated). Sixteen patients died on therapy, attributable to 8-week mortality (n = 8), relapsed/refractory AML (n = 1), death in CR/CRi/PR/HI from sepsis (n = 6), or hemorrhage (n = 1).

“We believe that implementation of clinical and immune biomarkers to select patients are likely to yield further improved outcomes with these types of therapies in AML,” said Daver.

Following these findings, a randomized phase III study of azacitidine with or without nivolumab or midostaurin (Rydapt), or decitabine and cytarabine alone in older patients with newly diagnosed AML or high-risk myelodysplastic syndromes is ongoing (NCT03092674). An ongoing phase II trial is evaluating nivolumab in eliminating minimal residual disease and preventing relapse in patients with AML in remission after chemotherapy (NCT02275533).


  1. Combination Chemotherapy and Immunotherapy Effective in Phase II Leukemia Study. MD Anderson News Release. Published November 8, 2018. Accessed November 27, 2018.
  2. 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 non-randomized, open-label, phase 2 study [published online ahead of print November 8, 2018]. Cancer Discov. doi: 10.1158/2159-8290.CD-18-0774.