A phase 1 study of durvalumab as monotherapy or combined with tremelimumab with or without azacitidine in patients with myelodysplastic syndrome

TL;DR

This phase 1 study tested the safety and effectiveness of durvalumab, alone or combined with other drugs, in patients with myelodysplastic syndrome who had previously been treated with hypomethylating agents.

Contribution

The study is the first to evaluate durvalumab in combination with tremelimumab and azacitidine for MDS in a clinical trial setting.

Findings

Durvalumab monotherapy and combinations were generally safe, with no dose-limiting toxicities in part 1 and 11% in part 2.

Marrow complete response was observed in 15% of patients in both parts of the study.

Hematologic improvement was seen in 35% and 30% of patients in parts 1 and 2, respectively.

Abstract

Upregulation of programmed death ligand-1 (PD-L1) has been observed in patients with MDS, and its expression on myeloblasts is associated with progression to AML. This open-label, phase 1 study evaluated the safety and tolerability of the PD-L1 antibody durvalumab as monotherapy (part 1) and in combination with tremelimumab, with or without azacitidine (part 2), in patients with MDS who progressed following hypomethylating agent treatment. Sixty-seven adults with MDS were enrolled (part 1, 40 with low/intermediate-1 or intermediate-2/high IPSS risk status; part 2, 27 with intermediate-2/high IPSS risk status). Primary safety endpoints included dose-limiting toxicities (DLTs) and treatment-emergent adverse events (TEAEs). Secondary endpoints included evaluation of clinical outcomes, survival, and pharmacokinetics. Dose-limiting toxicities were experienced by no patients in part 1 and 3 patients (11%) in part 2. The most common treatment-emergent adverse events were diarrhea and fatigue (40% each) in part 1 and fatigue (44%) and anemia (37%) in part 2. In parts 1 and 2, 15% of patients experienced marrow complete response as their best overall response, according to IWG criteria. Hematologic improvement was observed in 35% and 30% of patients respectively in part 1 and part 2. The study was terminated early due to limited efficacy.

The online version contains supplementary material available at 10.1007/s00277-024-06081-4.

Figures