Comparable outcomes among unmanipulated haploidentical, matched unrelated, and matched sibling donors in BU-based myeloablative hematopoietic stem cell transplantation for intermediate and adverse risk acute myeloid leukemia in complete remission: a single-center study

There are a limited number of studies comparing outcomes of busulfan (BU)-based myeloablative hematopoietic stem cell transplantation using unmanipulated haploidentical donors (HIDs), HLA-matched unrelated donors (MUDs), and HLA-matched sibling related donors (MSDs) in acute myeloid leukemia (AML) patients with complete remission (CR) status. With this background, we compared outcomes among 377 cases of CR following consecutive HID-HSCT for AML (CR) to 86 MUD and 92 MSD-HSCT cases. All patients received BU-based myeloablative conditioning and an unmanipulated graft within the same period. The median patient age was 23 years (range 1.1 to 65 years), and 230 patients (41.4%) were under age18. Among the 555 patients, 432 (77.8%) were of intermediate cytogenetic risk and 123 (22.2%) were of adverse risk. A total of 113 patients (20.5%) had FLT3-ITD+ AML, 425 patients (76.6%) were in first complete remission (CR1) post-transplant, and 130 (23.4%) patients were in second CR (CR2). GVHD prophylaxis included mycophenolate mofetil (MMF), cyclosporine-A (CSA) with short-term methotrexate (MTX) for HID, and MUD-HSCT. MMF is not used for MSD-HSCT. The median survival follow-up time was 42 months (range 18-91 months). The 3-year leukemia-free survival (LFS) among the HID, MUD, and MSD cohorts was 73.8% ± 4.8%, 66.4% ± 8.5%, 74.5% ± 2.4%, respectively (P = 0.637). Three-year overall survival (OS) was 74.9% ± 2.4%, 81.8% ± 4.3%, and 77.5% ± 4.5% among the HID, MUD, and MSD cohorts, respectively (P = 0.322). There were no difference among the relapse rate among the HID, MUD, and MSD donor cohorts (14.3% ± 4.0% vs 20.3% ± 6.4% vs 14.5% ± 2.2, respectively; P = 0.851) or the non-relapse mortality (NRM) (12.3% ± 3.5% vs 9.5% ± 3.2% vs 14.0% ± 1.8%, respectively; P = 0.441). Multivariate analyses showed that MRD-positive pre-HSCT was the only risk factor associated with a lower OS and LFS and higher risk of relapse among all 555 patients. Compared with the use of a MUD or MSD, an HID for HSCT had similar outcomes among AML patients with CR states who underwent an allo-HSCT with BU-based myeloablative conditioning. MFC-MRD-positive pre-HSCT was an independent negative factor impact on outcomes for AML patients in CR. We conclude that for AML patients who do not have a MSD or if an urgent transplant is required, HSCT from an HID is a valid option.