Comparison of allo-HSCT outcomes in younger versus older patients with AML: Findings from two prospective studies

Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is a potentially curative treatment for intermediate and high-risk acute myeloid leukemia (AML). Until recently, the use of allo-HSCT was limited to younger patients (<60 years) with high-risk disease and was dependent on the availability of donors. Older patients with AML who achieve complete remission (CR) with intensive induction therapy have higher relapse rates than younger patients and are also more likely to experience higher complications rates following allo-HSCT. However, with advances in conditioning regimens and donor selection, nonrelapse mortality (NRM) post-allo-HSCT has reduced, leading to an expanded role of allo-HSCT in older patients with AML. The existing evidence surrounding the efficacy of allo-HSCT in older patients with AML is mostly derived from retrospective and single-centered studies. In addition, meta-analyses in younger patients with AML suggest that allo-HSCT may be more beneficial in patients with intermediate-risk AML. Therefore, prospective studies are warranted to investigate the role of allo-HSCT in younger and older patients with AML.

Here, we provide a comparison of the key findings from two abstracts investigating the role of allo-HSCT in younger and older patients with AML, presented during the 63rd American Society of Hematology (ASH) Annual Meeting and Exposition. These abstracts discussed findings from the Etal-1 trial (NCT01246752) of allo-HSCT in younger patients (≤60 years) with intermediate-risk AML in first remission¹, and from the ECOG-ACRIN (E2906) trial (NCT02085408) in fit older patients.²

Study designs

Etal-1¹

A prospective cohort study using data from a multicenter, randomized controlled trial in 16 centers in Germany between 2010 and 2018. Eligible patients had AML with intermediate-risk cytogenetics, were aged ≥ 18 and ≤ 60 years, in first CR (CR1) with normal karyotype, fit for transplant, and availability of a human leukocyte antigen (HLA)-matched sibling or unrelated donor.

Patients were stratified according to

• Age: ≤ 40 vs > 40 years
• Donor type: Unrelated vs HLA-matched sibling donor
• NPM1/FLT3 and CEBPA mutational status

ECOG-ACRIN (E2906)²

A prospective cohort analysis using data from the ECOG-ACRIN (E2906) study; the inclusion criteria and trial schema have been previously reported by the AML Hub. Included patients with AML were aged 60–73 years, and those who achieved CR1 were eligible for allo-HSCT.

In both studies:

• Primary endpoints were overall survival (OS), and disease-free survival (DFS) defined as time from allo-HSCT to death from any cause or to relapse.
• Secondary endpoints included cumulative incidence of relapse and NRM.

Results

Baseline characteristics

Etal-1¹

Data from 143 patients in CR1 were included, of which 76 were randomized to allo-HSCT and 67 to conventional consolidation chemotherapy. The trial was halted prematurely due to slow patient accrual.

• There was a slightly higher proportion of males and patients aged 41–60 years in both allo-HSCT and consolidation arms (Table 1).

ECOG-ACRIN (E2906)²

A total of 166 patients who received allo-HSCT were included in the cohort, of which 105 were in CR, CR with incomplete hematologic recovery (CRi), or leukemia-free stage (LFS) (Table 1).

• Median follow-up was 33.6 and 29.1 months from diagnosis and transplantation, respectively.

Table 1. Baseline characteristics*

Clinical outcomes

Etal-1¹

• Although DFS was improved in the allo-HSCT arm (71%) compared to the consolidation arm (41%; p = 0.010), OS rates were similar (allo-HSCT arm, 76%; consolidation arm, 83%; p = 0.495).
• The cumulative incidence of relapse at 2 years was higher in the consolidation arm (57%) compared to the allo-HSCT arm (20%; p < 0.001).
• At 2 years, NRM was low in both allo-HSCT and consolidation arms (9% vs 2%; p = 0.017).
• Notably, all 38 patients in the consolidation arm who relapsed (33 hematologic, 4 molecular, and 1 extramedullary) proceeded to allo-HSCT as salvage therapy.

ECOG-ACRIN (E2906)²

In patients undergoing allo-HSCT who had achieved CR/CRi/LFS:

• Median DFS was 25.8 months and median OS was 35.3 months
• 2-year DFS was 53.6% and OS was 56.4%
• 3-year DFS was 45.6% and OS was 49.4%
• 4-year DFS was 39% and OS was 42.9%

Univariate analysis revealed no impact of age on OS (HR, 1.03; 95% CI, 0.63 – 1.69; p = 0.90). In patients aged >65 years, NRM was 8.4% at 6 months and 24% at 2 years, whereas in patients aged ≤65 years, NRM was 4.4% at 6 months and 15.6% at 2 years).

Conclusion

The analysis of the Etal-1 cohort found no OS benefit in younger patients with intermediate-risk AML following allo-HSCT, although relapse risk was significantly reduced. On the other hand, data from the ECOG-ACRIN (E2906) cohort showed improved OS in fit older patients who underwent allo-HSCT in CR1 with an acceptable NRM rate, with comparable NRM rates between older and younger patients with AML.

One of the major barriers to prolonged survival after allo-HSCT in the older population is the occurrence of relapse, reflecting the biology of AML in these patients. Therefore, further research into post-remission strategies that are complementary or competitive to allo-HSCT is warranted. In addition, investigations into delayed allo-HSCT approaches in patients with CR1 intermediate-risk AML, and studies with longer follow-up periods and larger sample sizes would be helpful to understand any effects of cumulative toxicities. Such studies should consider MRD status in younger patients undergoing allo-HSCT, and algorithms to integrate donor availability, transplant-related risk, disease biology, and treatment response.