BACKGROUND Relapse after allogeneic hematopoietic stem cell transplantation (allo-HSCT) remains the main cause of treatment failure for patients with acute myeloid leukemia (AML). Conventional treatments such as donor lymphocyte infusions (DLI) or secondary allo-HSCT as a monotherapy or in combination with novel agents or immunotherapy showed limited improvements on long-term survival. Our previous studies have shown that a rare subset of CD4- and CD8- double negative T cells (DNTs) expanded from periphery of healthy donors are effective in targeting AML cells in vitro and in xenograft models. Also, allogeneic DNTs can be used as an off-the-shelf adoptive cellular therapy (ACT) for AML. We undertook a first in-human phase I/IIa trial to assess the safety and efficacy of escalating doses of DNTs for the treatment of AML patients relapsed post allo-HSCT. This study protocol was approved by Chinese Ethics Committee of Registering Clinical Trials. Informed consent was obtained in accordance with the Declaration of Helsinki. The trial was registered on the Chinese Clinical Trial Registry (ChiCTR-IPR-1900022795). METHODS From June 2019 to June 2020, 12 patients with relapsed AML after allo-HSCT were enrolled in the trials. Three patients received HLA-matched sibling stem cell transplant, and 9 patients received umbilical cord blood transplantation (UCBT). Hematological relapse was found in 9 cases, and molecular relapsed in 3 cases. None of the patients were able to obtain DLI from previous donors for various reasons. Clinical-grade DNTs from peripheral blood of healthy donors were expanded. After lymphodepleting chemotherapy with fludarabine and cyclophosphamide, each patient received three infusions of HLA-mismatched DNTs expanded from healthy volunteers at escalating doses of 5×107, 1×108, or 2-4×108 per kilogram of body weight at one week interval. Numbers of total DNTs and donor-derived DNTs, inflammatory cytokines levels in vivo were measured according to the study protocol. Post-remission therapy was permitted 14 days after the last DNTs infusion according to treating physician's discretion. RESULTS All surviving patients were followed until August 1st, 2020. At a median follow-up of 4.3 months (range 1.7 to 12.2 months), 1 patient withdrew from the trial after the second DNT infusion for a personal reason. 6/11 patients (54.5%) achieved complete response (CR) with 5 of these patients negative for (45.4%) minimal/measurable residual disease (MRD). 4/11 patients (36.3%) achieved partial remission. Up to the last follow-up, 4/11 patients remain CR. Increase in the levels of serum inflammatory cytokines, including interleukin (IL)-6, TNF-alpha, MCP-1, and MIP-1-beta over baseline were observed without signs of cytokine release syndrome, neurotoxicity, or graft-versus-host disease. The most common adverse events related to DNTs treatment were fever and headache. None of the patients showed >grade 2 adverse event (AE). The maximum tolerated dose was not reached. Donor DNT cell expansion was detected as early as 6 hours after infusion and reached the peak at 24 hours after the 1st infusion in most patients, slightly increased after the 2nd and 3rd infusions, and lasted for 2 weeks after 3rd infusion. However, total DNTs increased and reached the peak at 2 weeks after 3rd infusion. CONCLUSIONS Our results show that allogeneic DNTs therapy is safe, well tolerated, and potentially effective in treating AML patients who relapsed after allo-HSCT. The total number of DNTs increased after infusion. Compared to other agents, DNTs more effectively induced complete responses in this patient population. Our on-going study is to elucidate the mechanism of DNTs-mediated leukemia clearance, and explore the use of DNTs as carrier for Chimeric Antigen Receptor or transgenic TCR to further enhance the efficacy targeting resistant AML cells. Disclosures Zhang: AbbVie: Current equity holder in publicly-traded company; University Health Network: Consultancy, Current equity holder in private company, Other: Principal Investigator, Patents & Royalties.

Intravenous busulfan (IV-Bu) or total body irradiation (TBI) based regimens are currently the most widely used myeloablative conditioning regimens for patients with hematologic malignancies undergoing allogeneic stem-cell transplantation(allo-SCT). Numerous trials have been undertaken on the clinical outcomes between IV Bu and TBI, but there are no comparative data for cord blood transplantation(CBT). We conducted a prospective registry-based study to analysis the outcomes of IV Bu and TBI in CBT patients with hematologic malignancies. From May 1, 2008 to Mar 31, 2016, a total of 331 consecutive patients with hematologic malignancies recieved singe unrelated CBT were involved in the study. Eligibility criteria for this analysis included:(1)Weigh ≧35 kilograms and age ≦ 60 years; (2)All patients recieved a single unit CBT but not a double units CBT; (3)Consensus criteria preparative regimens were based on full dose IV Bu(total 12.8 mg/kg, 0.8mg/kg every 6 h for 4 days) or TBI(total 12 Gy, 4 fractions) combined with Cy(60mg/kg × 2d); (4)GVHD prophylaxis regimens include cyclosporine(CSA) and mycophenolate mofetil(MMF) without Antithymocyte Globulin(ATG). Patients who has recieved a previous autologous or allogeneic transplantation was excluded in the study. The cumulative incidence of neutrophil engraftment were 91.6% in IV Bu/Cy cohort and 98.0% in Cy/TBI cohort(P < .001), respectively. The median follow-up time in IV Bu/Cy and Cy/TBI cohorts was 28.7(range, 12.2 to 91.3) months and 55.5(range, 13.1 to 117.1) months, respectively(P<.001). Of them, 200 patients received Cy/TBI, and 131 patients received IV Bu/Cy. The median dose of infused total nucleated cells were 4.51(range, 2.19 to 12.06 )× 107/kg and 3.77(range, 2.14 to 9.05)(P < .001), respectively. And the median number of CD34+ cells was 2.61(range, 0.91 to 9.64) ×105/kg and 2.01(range, 1.12 to 8.71)(P < .001), respectively. The median times to neutrophil engraftment were 16 days and 19 days(P < .001). The estimated DFS at 3 years were not statistically different between the conditioning regimens in univariate analysis. 63.9%±4.21% in patients undergoing conditioning with IV Bu/Cy and 54.4%±3.57% in patients undergoing conditioning with Cy/TBI(P = .21).These results were similar in multivariable analysis (HR, 1.03; 95% CI, 0.68 to 1.54; P = .905). Univariate analysis and multivariable analysis demonstrated that estimated OS and GRFS at 3 years were also not different between those two conditioning regimens. Our resluts demonstrates that, compared with TBI, IV Bu regimen was associated with a higher incidence of graft rejection in CBT. But there was no difference in survival with no increased risk for NRM or relapse between two regimens. For those centers lack of radiation facilities, IV Bu may be a valid and efficient alternative to TBI. With the restriction of a retrospective registry analysis and limited patient munbers, rigorously designed prospective randomized controlled trials are needed to further investigated the availability of IV Bu and TBI for CBT. Disclosures No relevant conflicts of interest to declare.