Evaluating predictive factors for post-HSCT survival in secondary AML

Patients with secondary acute myeloid leukemia (sAML) have poor clinical outcomes, even after intensive chemotherapy, and allogeneic hematopoietic stem cell transplantation (HSCT) remains the only curative treatment option for these high-risk patients. It is unclear whether disease burden prior to HSCT holds similar prognostic value for sAML as for de novo AML, and the influence of cytogenetic and molecular risk at diagnosis on outcomes for patients with sAML, compared with de novo patients, is also unknown.

Data evaluating the impact of pre-HSCT disease burden and European LeukemiaNet (ELN) risk classification at diagnosis on survival for patients with sAML and de novo AML were presented by Claudia Nuñez-Tórrez in two abstracts during the 47th Annual Meeting of the European Society for Blood and Marrow Transplantation (EBMT).^1,2 Here, we summarize their findings.

Prognostic impact of pre-HSCT disease burden

A retrospective analysis was performed on 108 patients with de novo AML (n = 49) or sAML (n = 59) to assess:

• post-HSCT event-free survival (EFS) and overall survival (OS) for sAML compared with de novo AML;
• the impact of achieving a complete remission (CR) before HSCT on EFS and OS; and
• the impact of achieving CR and measurable residual disease (MRD) by flow cytometry on EFS and OS.

Just over half of patients with de novo AML had recurrent cytogenetic abnormalities. Most sAML cases (63%) were classified as AML with myelodysplasia-related changes; Ph-negative blast phase myeloproliferative neoplasms and therapy-related AML accounted for 29% and 8% of cases, respectively.

Selected patient, disease, and treatment characteristics are shown in Table 1. Of note, patients with sAML were older, more frequently had an adverse cytogenetic risk profile, and were more likely to be chemorefractory, than patients with de novo AML.

Table 1. Selected patient, disease, and treatment characteristics*

At a median follow-up of 12 months, EFS and OS after HSCT were significantly inferior for patients with sAML compared with patients with de novo AML; the median EFS was 7 months vs not reached (hazard ratio [HR], 2.6; p < 0.001) and the median OS was 10 months vs not reached (HR, 3; p < 0.001).

• When assessing the impact of pre-HSCT disease state, patients with active disease had worse EFS and OS than those with a CR (EFS: HR, 3.5; p < 0.001; OS: HR, 3.2; p < 0.001).

• In a subanalysis according to AML type, patients with de novo AML who achieved a CR before transplant had an improved EFS and OS compared with those transplanted with active disease (EFS: HR, 6.2; p < 0.001; OS: HR, 6.5; p < 0.001).
• However, the improvement in EFS was less significant for patients with sAML who achieved CR (HR, 2.5; p = 0.003), and there was no significant difference in OS for patients with sAML in CR compared with those with active disease (HR, 1.9; p = 0.05).

The variables identified as significant predictors of EFS and OS in multivariate analysis were AML type (sAML vs de novo AML) and pre-HSCT disease state (CR vs active disease) (Table 2).

Table 2. Factors associated with EFS and OS in univariate and multivariate analysis*

Of 89 patients in CR pre-HSCT, 86 had MRD status determined by flow cytometry; 63 were MRD-negative and 23 were MRD-positive.

• Patients in CR with MRD-negative status trended towards improved 2-year EFS, compared with MRD-positive patients (p = 0.09).
• Prospective studies are required to determine whether MRD status has equivalent prognostic impact in sAML and de novo AML.

Prognostic impact of cytogenetic and molecular risk

The influence of ELN classification at diagnosis on survival outcomes was assessed in a retrospective analysis of a slightly modified patient cohort to the one presented above, with the addition of three patients with de novo AML (N = 111; de novo AML n = 52, sAML n = 59).

Proportional ELN classification at diagnosis for the de novo and sAML patient cohorts is shown in Figure 1.

Figure 1. ELN risk classification at diagnosis*

Median EFS and OS for patients with de novo AML and sAML according to ELN risk classification are shown in Table 3. Whilst patients with de novo AML and a favorable risk profile had improved outcomes compared with patients with adverse risk, the ELN risk classification did not influence survival for patients with sAML.

Table 3. EFS and OS according to ELN risk classification*

Summary

Patients with sAML are more likely to be chemorefractory and have inferior survival compared with patients with de novo AML. Unlike for patients with de novo AML, the achievement of a CR prior to transplantation was not so strongly predictive of EFS and had no significant association with OS for patients with sAML. Moreover, transplant results for patients with sAML were not influenced by ELN risk classification at diagnosis.