Phase IB/II Study of Lirilumab in Combination with Azacytidine (AZA) in Patients (pts) with Relapsed Acute Myeloid Leukemia (AML)

Background: Inhibitory killer-cell immunoglobulin receptors (KIRs) negatively regulate NK cell-mediated killing of HLA class I-expressing tumors. Lack of KIR-HLA class I interactions has been associated with potent NK-mediated antitumor efficacy in AML patients in remission upon haploidentical stem cell transplantation (SCT) from KIR-mismatched donors (Ruggeri L et al., Science 2002). Blockade of KIR2DL1, 2 and 3 receptors induced augmented NK-cell mediated lysis of tumor cells (Romagne F et al., Blood 2009). Hypomethylating agents possess anti-leukemia activity and concomitantly alter immune regulation. Blockade of the KIR-receptors by lirilumab may improve response rates and abrogate immune-mediated resistance to hypomethylating agents.

Patients and Methods: Pts are eligible if they have AML and failed prior therapy (including prior therapy with a hypomethylating agent), have adequate performance status (ECOG ≤ 2), and organ function. AZA was given at the dosage of 75mg/m² on days 1-7; lirilumab was given on Day 8 at the dosage of 1 and 3 mg/kg in 2 consecutive cohorts of 6 pts each. Courses were repeated approximately every 4-5 weeks. No dose-limiting toxicities were observed and lirilumab 3mg/kg was established as the recommended phase 2-dose (RP2D) in combination with AZA. 9 additional pts have been treated at the RP2D. Responses were evaluated at the end of 3 courses of therapy.

Results: To date, 21 pts (11 de novo, 10 secondary AML), median age 60 years (range, 33 - 89), 48% with adverse cytogenetics, median prior therapies of 3 (range, 1-8), and prior allogeneic SCT in 6 (28%) have been enrolled. All 21 pts had baseline next generation sequencing for 28-genes and frequently identified mutations included TP53 (n=6), TET2 (N=6), ASXL1 (n=5), RUNX1 (n=5), EZH2 (n=3), DNMT3A (N=2), and CEBPA (n=2). 12 pts are ≥ 3 months into therapy and evaluable for response at this time: 1 achieved complete remission (CR), 1 achieved complete remission with insufficient count recovery (CRi), 2 (17%) had ≥50% bone marrow (BM) blast reduction, 2 (17%) had hematologic improvement (HI) > 6 months, and 6 (50%) had progressed. Nine pts were too early for response assessment. The 4- and 8-week mortality is 0 and 5%, respectively. The median duration of response, overall survival (OS) and event-free survival for the 12 evaluable pts were 2.5 months (range, 1.1 - 3.0) 4.4 months and 3.2 months, respectively.

Grade 3/4 toxicities irrespective of causality were similar to those seen with AZA based therapies in salvage patients included 15 episodes of neutropenic infections, 6 pneumonia, 1 UTI, 1 skin infection, 2 abdominal pain, and 1 mucositis. Immune mediated toxicities were observed in 3 (14%) pts (1 pneumonitis Grade 3, 1 colitis Grade 2, 1 infusion reaction Grade 2), respectively. The immune mediated toxicities responded rapidly to steroids and all 3 pts could be rechallenged safely with lirilumab. Six pts were postSCT and no Grade 3/4 GVHD flares were noted. No pts have come off study due to toxicities and the azacytidine or lirilumab were not discontinued in any pts due to toxicities.

Multicolor flow-cytometry studies and Mass-cytometry (CyTOF) studies are being conducted by the Immunotherapy Platform on baseline and on-treatment BM aspirate (end of cycle 1, 2, 4, 8) and peripheral blood to assess NK- and T-cell costimulatory markers.

Conclusion: Full doses of AZA and lirilumab were well tolerated in heavily pretreated pts with relapsed AML with poor risk features, including pts with post-allogeneic SCT relapse. The efficacy data are still preliminary. AZA with lirilumab is being investigated in earlier salvage in AML and frontline and salvage settings in myelodysplastic syndrome.