Most available data on the relative value of allogeneic stem cell transplantation (allo-SCT) for acute myeloid leukemia (AML) in first complete remission (CR1) in the past, were generated from donor versus no-donor comparisons, focusing on documented tissue-typed patients and their siblings. The inclusion of patients who are not HLA-typed, for instance all those without siblings, causes bias of unknown magnitude. Additionally, concerns about the equivalence of related and unrelated donors should no longer be a problem in contemporary evaluation of allo-SCT. Using data of the prospective AMLCG 1999 trial, we performed a matched-pair analysis, to evaluate outcome in patients with AML according to post-remission allo-SCT or conventional postremission chemotherapy (PRT). 165 patients pairs in CR1 were identified, who matched for the following criteria: AML type (de novo AML, s-AML, t-AML, high-risk MDS); cytogenetic risk group [unfavorable (UNF-CG), intermediate (INT-CG), and favorable with the exclusion of t(15;17)]; age (± 5 years); and time in CR1 to account for the time to transplant in allo-SCT patients. If possible, patients were also matched for sex and assigned induction treatment (TAD-HAM versus HAM-HAM). 34 patient pairs had an UNF-CG, 122 pairs INT-CG, and 9 pairs had favorable cytogenetics. Median patients age at diagnosis was 45 years (range: 16–59). In the allo-SCT cohort, 105 patients had a related donor (matched related donor [MRD] 104, haploidentical 1) and 60 a matched unrelated donor (MUD). Median follow-up of surviving patients after first diagnosis of CR1 was 7.5 years. Projected 7-year relapse-free survival (RFS) was 56% in the allo-SCT group and 39% in the control group (p <.0001, log-rank test). Overall survival (OS) was 58% and 45% (p=.143), respectively. RFS was significantly improved by allo-SCT in patients with UNF-CG (23% vs. 12% at 7 years; p=.005) or INT-CG (58% vs. 37%; p=.001). OS was 31% in allo-SCT patients with UNF-CG versus 18% in matched controls (p=.052) and 64% in INT-CG patients with allo-SCT versus 54% in matched controls (p=.403). Dividing the 330 patients into age groups by decades, revealed an age dependent, increasing risk of relapse for patients receiving conventional post-remission therapy, with cumulative relapse incidences of 51% (<31 years), 47% (31–40 years), 60% (41–50%) and 87% (51–60 years) at 7 years, whereas allo-SCT patients had similar relapse incidences of 32%, 34%, 25% and 34% respectively. The higher relapse incidence in control patients >50 years of age, resulted in a significantly better OS of allo-SCT patients with 27% versus 58% (p=.022) in this age group. In the subset of patients with INT-CG, allo-SCT patients with non-normal karyotype had both a significant better OS and RFS after 7 years compared to control patients, whereas patients with normal karyotype had similar RFS and OS regardless of NPM1 and FLT3 mutational status. Of note, 48 of 99 patients with AML relapse in the control cohort, received an allo-SCT (18 from a MRD, 30 from a MUD) beyond CR1 (9 with UNF-CG, 38 with INT-CG, 1 with favorable CG). Median OS of 48 matched patients receiving an allo-SCT in CR1 was 54%, while it was 39% in paired patients with allo-SCT beyond CR1 (p=.289). We conclude that allo-SCT is the most potent post-remission therapy for AML with UNF-CG and INT-CG. Its impact on OS is difficult to assess, as about a third of patients initially treated with conventional PRT, underwent allo-SCT beyond CR1. In contrast to results from donor versus no-donor comparisons, our data highly suggest a benefit of allo-SCT in CR1, particularly for elderly patients, and in line with such comparisons, for patients with intermediate-II (according to the European LeukemiaNet [ELN] recommendation) or unfavorable ELN cytogenetic risk. Ultimately, the gold standard for the evaluation of allo-SCT in patients with INT-CG in CR1 is a randomized controlled trial, which is now feasible with unrelated donors becoming widely available and is conducted by the German Cooperative Transplant Study Group (ETAL-1 study).
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.