Poor Outcomes Among Pediatric Non-Down Syndrome Acute Megakaryocytic Leukemia Patients Irrespective of Receipt of Hematopoietic Stem Cell Transplant

Background: Pediatric Acute Megakaryocytic Leukemia (AMKL) is a subtype (FAB-M7) of Acute Myeloid Leukemia (AML) that is common in children with Down syndrome (DS). Non-Down Syndrome (non-DS) AMKL is rare, accounting for only 3-10% of childhood AML. This disease is characterized by complex cytogenetics and recurrent oncogene translocations that carry prognostic implications, such as monosomy 7, loss of 7q and CBFA2T3-GLIS2, NUP98-KDM5A, RBM15-MKL1 and KMT2A gene fusions. Children with non-DS AMKL have substantially inferior outcomes when compared to their DS counterparts, due to a high incidence of relapse despite intensive upfront chemotherapy. Due to the frequently high-risk nature of this disease and grim outcomes, cases are often referred for allogeneic hematopoietic stem cell transplant (HSCT) in first complete remission (CR1), with the hope of utilizing the graft-versus-leukemia effect to consolidate remissions and achieve cure. However, due to the rarity of this diagnosis, there are insufficient data to substantiate a benefit from HSCT that would support its use as a standard of care for these patients. We conducted a review of local AMKL cases to investigate outcome differences by treatment with HSCT vs. chemotherapy alone.

Methods: We retrospectively reviewed medical records of pediatric non-DS AMKL patients treated at Texas Children's Hospital from 2000-2021. Patient demographics, immunophenotype, cytogenetics, molecular testing, treatment, and clinical outcomes were abstracted from the electronic health record. Only patients with confirmed non-DS AMKL diagnosis after pathology review were included. Patients who received a 2^nd HSCT were included, but censored at the date of 2^nd HSCT.

Results: Thirty-five patients were included in the cohort, 24 of whom received both chemotherapy and HSCT. Median age at diagnosis was 2 years (range: 0.6-19 years), with 74% of patients under 5 years. Most patients were female (n=23, 66%), and 40% (n=14) identified as Hispanic-White. Central nervous system disease was relatively infrequent (17%). Monosomy 7/loss of 7q (n=4), KMT2A-rearrangements (n=6) and CBAF2T3-GLIS2 fusions (n=3) were the most common recurrent genetic alterations. Most patients were stratified as high risk (n=24, 70%) according to the AAML1031 criteria. Six out of eleven patients (55%) that were treated with chemotherapy alone developed relapsed/refractory progressive disease and were ineligible for HSCT due to disease-related death. Two of 11 patients died from infection during induction chemotherapy, and the remaining 3 patients lacked high risk features, were treated with chemotherapy alone, and remain alive and disease-free. For those treated with HSCT, the median age at HSCT was 3 years (range: 0.8-19 years). Most patients received bone marrow-sourced stem cells (n=15, 62%) from matched-unrelated or haploidentical donors (n=18, 75%) and received busulfan-based conditioning regimens (n=16, 67%). Sixty-three percent (n=15) of patients were in minimal residual disease (MRD) negative CR1 prior to HSCT, while 17% (n=3) had active disease. Relapse occurred in 71% of those treated with HSCT (n=17). Seven patients (29%) are alive after HSCT, 2 of whom required a 2^nd HSCT. Five-year overall survival (OS) and disease-free survival (DFS) for the entire cohort was 27.1% and 20%, respectively. One-year OS was significantly higher for patients receiving chemotherapy plus HSCT (78.5% vs 45.5%; p=0.039). However, 5-year OS and DFS were similar between groups (OS: 40.7% vs 22.2%, p=0.61; DFS: 26% vs 20.2%, p=0.93). Within the HSCT group, 5-year DFS was higher for patients in CR1 (36.9%) vs. CR2 or active disease prior to HSCT (0% for both, p=0.01). However, 5-year OS was not affected by pre-HSCT disease status (p=0.11). Relapse was the leading cause of death, with 5-year cumulative incidence of 0.81 vs 0.78 (p=0.67). Treatment-related mortality was exclusively due to infection (cumulative incidence 0.17 at 5-years).

Conclusion: Non-DS AMKL is an aggressive AML subtype with poor outcomes despite the inclusion of HSCT. HSCT did not improve long-term DFS or OS due to a high incidence of relapse, even for children who achieved an MRD negative CR1 prior to HSCT. Dismal outcomes from this disease support a critical need for the development and incorporation of novel targeted agents in upfront therapy and/or in the post-transplant period.

No relevant conflicts of interest to declare.