Prospective Phase II Study of Prophylactic Azacitidine and Donor Lymphocyte Infusions Following Allogeneic Hematopoietic Stem Cell Transplantation for High Risk Acute Myeloid Leukemia and Myelodysplastic Syndrome

Although allogeneic hematopoietic stem cell transplantation is presently the only curative option for many patients with AML or MDS, relapse remains the main cause of morbidity and mortality. Strategies are therefore developed to prevent relapse. Post-transplant immune intervention with administration of prophylactic or pre-emptive donor lymphocyte infusion (DLI) and/or chemotherapy maintenance using DNA-demethylating agents such as azacitidine (aza) is being investigated. The exact mechanism of action of aza remains obscure and might be due in part to tumor antigen upregulation or other gene induction by tumor cells causing an immune response. We enrolled, in a clinical trial (ClinicalTrials.gov Identifier:NCT01541280), patients (pts) with high risk AML or MDS and candidates for allo-HSCT transplantation to receive azacitidine (aza) and DLI post-transplant as prophylactic treatment with the primary objective to reduce the relapse rate at 2 years following allo-HSCT and secondary objectives to increase disease free survival at 2 years post-transplant, increase overall survival at 2 years, investigate feasibility and safety of maintenance strategy combining chemotherapy and immunotherapy and follow the incidence and severity of acute and chronic GVHD. High risk AML was defined as AML in CR1 with unfavorable cytogenetics, AML in CR2 or greater remission, refractory AML or in relapse prior allogeneic transplantation. High risk MDS was defined as MDS with intermediate-2 group and higher risk group according to IPSS criteria. Aza was scheduled to begin between d+56 and d+112 post-transplant at the doses of 32 mg/m²/d sc for 5 days every 28 days for up to a total of 12 cycles if the pt had not acute GVHD >1 or severe infection. The first DLI was started following 3 cycles of aza and discontinuation of immunosuppressive prophylaxis, and if the pt had no clinical signs of GVHD, uncontrolled infection or a recent history of gr>2 acute GVHD. Two other DLI were scheduled every 8 weeks after the 5^th and 7^th cycle of aza. The doses of DLI 1, 2 and 3 were respectively 5x10⁶, 1x10⁷, 5x10⁷ CD3+cells/kg for related donor, and 1x10⁶, 5x10⁶, 1x10⁷ CD3+cells/kg for unrelated donor. Sixty-four patients were pre-included prior transplantation, 30 pts were subsequently included, 20 pts with AML and 10 pts with MDS, median age 58 y (22-70). The status at transplantation was: CR1 = 16 pts (53%), CR2 = 6 pts (20%), refractory = 5 pts (16%), upfront transplantation for MDS = 3 pts (10%). Cytogenetics was normal or intermediate for 15 pts and unfavorable for 15 pts (namely 8 pts with complex caryotype). Conditioning was myeloablative for 11 pts, reduced for 19 pts (including 2 sequential). Donors were unrelated volunteers in 18 pts (60%).The time between allografting and first aza cycle was 66 days (38-93). The median number of cycles of aza administered was 5 (1-12) with 10 pts (33%) completing the 12 cycles. Forty one DLI were injected in 17 pts: 5 pts received one DLI, 2 pts received 2 DLI, 8 pts received 3 DLI. Two additional pts received 4 and 5 DLI because of a mixed chimerism. The first DLI was given at a median of 142 d (129-221) post transplantation. Aza was well tolerated, but was discontinued in 20 pts: because of GVHD (n=11), relapse (n=5), GVHD/infection (1pt), sudden death due to heart failure (n=1), withdrawal of consent (n=2, one after 1 cycle and another after 5 cycles). Four months following transplantation, 24 (80%) demonstrated full donor chimerism (>95%) in CD3+ cells. Nine patients developed grade 1 to 3 acute GVHD (CI 29.8±9%), 6 who did not receive DLI and 3 following DLI (grade 1 n=2, grade 2 n=3, grade 3 n=4). No grade 4 acute GVHD was observed. Nine pts developed chronic GVHD (2 limited, 7 extensive), 3 who did not receive DLI, 6 following DLI. Twenty patients are alive. With a median follow-up from the allotransplant for those alive of 36 months (range 12 - 46 months), the survival at three years is 66%. Causes of death were infection (n=1), relapse (n=8), sudden death due to heart failure (n=1). The median time to relapse was 5 months (2.5-9) and the cumulative incidence of relapse at 3 years 28.1±8.5%. These results confirm that aza is well tolerated as a prophylactic treatment to reduce the risk of post-transplantation relapse. The incidence of GVHD following aza + DLI was not overwhelming. Analysis of T cell population and immune response as well as comparison to matched-pair control are currently performed and will be presented.