Checkpoint Inhibitors Advancing for AML, MDS

“We have a lot of work going on in this area, to see what the expression of these checkpoint are and how we can clinically use them,” said Daver, assistant professor, Department of Leukemia, MD Anderson Cancer Center. “The next step, like with most leukemias, is how can we combine them with approved agents for these disease.”

Early research into the viability of checkpoint inhibitors in hematologic malignancies focused heavily on identifying which checkpoints might have the greatest activity. In these studies, DNA methyltransferase (DNMT) was found to upregulate PD-L1 and PD-1, which was linked to resistance to therapy and worse outcomes.

Clinical activity, specifically for those with AML, was first demonstrated in a phase I/Ib study assessing the CTLA-4 inhibitor ipilimumab (Yervoy) for relapsed hematologic malignancies following allogeneic stem cell transplantation. Across 12 patients with AML treated with ipilimumab at 10 mg/kg, the complete response (CR) rate was 42%.

Additionally, a separate phase I study assessed the combination of the hypomethylating agent azacitidine with the PD-1 inhibitor nivolumab (Opdivo) for patients with relapsed/refractory AML. After a median follow-up of 8.3 months across 35 patients, the CR and CR with incomplete hematologic recovery (CRi) rate was 18%. The overall response rate (ORR) was 32%.

Although the data were still immature and the patient population was small, there was a strong trend toward improved overall survival (OS), according to Daver. The median OS with azacitidine plus nivolumab was 9.3 months, which compared favorably with a historical OS of 4.7 months for this population (P= .02).

Overall, 12 patients in the study experienced a grade 2 to 4 adverse event (AE), which included pneumonitis (n = 11), nephritis (n = 5), skin rash (n = 2), and colitis (n = 2). A majority of the AEs occurred within the first 6 weeks of treatment and all toxicities responded rapidly to treatment with steroids.

“The toxicity has been a learning curve for me and for my group, since leukemia doctors have not yet been using these drugs,” said Daver. “The pattern and profile of toxicities seems to be different than what was seen for solid tumors. The majority of toxicities have been pneumonitis, and the others have been colitis and skin related.”

When looking at T cell subpopulations before and after nivolumab/azacitidine, the number of CD3+ cells and the ratio of CD8+ cells to Tregs predicted response to treatment. In these early observations, PD-L1 expression did not appear to impact results. “There were some hints that baseline immune infiltrate is able to predict for response,” added Daver.

A study assessing immunotherapies for MDS demonstrated similar promise. In this trial, patients received single-agent nivolumab (n = 15), azacitidine plus nivolumab (n = 13), or ipilimumab (n = 13). The combination demonstrated an ORR of 68% and responses were still pending for ipilimumab. Monotherapy with nivolumab did not elicit a response.

A second study is looking at the combination of the killer-cell immunoglobulin-like receptors (KIR) inhibitor lirilumab and azacitidine. In early data from 3 patients treated with the combination, the ORR was 100% and there were no unexpected toxicities.