Allotransplants for Patients 65 Years or Older with High-Risk Acute Myeloid Leukemia

Outcome of persons >65 years with acute myeloid leukemia (AML) is poor. Allogeneic transplant can cure some of these patients but is associated with substantial transplant-related mortality (TRM) and high relapse risk.

We analyzed 185 consecutive patients >65 years with high-risk AML between 2010 and 2015 who had measurable residual disease (MRD) and molecular risk assessments and who received myeloablative conditioning (MAC, n=42) or reduced-intensity conditioning (RIC, n=143) and a transplant from an HLA-identical sibling (n=66) or a 10/10 loci HLA-matched unrelated donor (MUD, n=119) in order to identify patients who would benefit from allotransplant. MRD was assessed by flow cytometry in bone marrow samples. AML cytogenetic and molecular risk were defined using the 2017 European Leukemia Net genetic risk stratification.

The patients who were MRD-negative at time of transplant had a better 2-year cumulative incidence of relapse (CIR 17.6%) and a 2-year overall survival (OS, 69.4%) as compared to the patients in remission but MRD-positive (55.6% CIR, 21.5% OS) or the patients who had morphological evidence of leukemia prior to transplant (48.7% CIR, 19.9% OS), (p<0.0001). Multivariate analysis for 2-year CIR showed that having detectable leukemia at time of SCT (defined as MRD-positive or morphological evidence of leukemia) (HR=14.5, CI=3.4-61.4, p<0.0001), having received <3 cycles of chemotherapy prior to SCT (HR=1.8, CI=1.1-2.8 p=0.01), and high-risk genetics were independent predictors of relapse. However, high-risk genetics only had a deleterious effect on outcomes for MRD-negative patients (HR=9.4, CI=2.0-43.6, p=0.004), and did not affect the outcome of patients with detectable leukemia (HR=1.1, CI=0.7-1.9, p=0.61).

Patients who received RIC (Flu-Mel-RIC or Bu-Flu-RIC) or MAC (Bu-Flu-MAC) had similar proportion of MRD-negative patients (p=1.0). However, MRD-negative patients who received Flu-Mel-RIC had a superior OS than patients who received Bu-Flu-RIC (adjusted HR=1 vs 5.3, p=0.02) or patients who received Bu-Flu-MAC (HR=1 vs 5.093, p=0.04), which is explained by a significantly lower relapse rate in patients receiving Flu-Mel-RIC (adjusted HR=1 vs 4.8, p=0.03) and a significantly lower TRM as compared to patients receiving Bu-Flu-MAC (adjusted HR= 5.1, p=0.01).

Patients who had major medical complications (MMC), defined as a medical event requiring admission to the intensive care unit for ventilatory or inotropic support or a medical event that prolonged the patient hospitalization for more than 2 weeks, during induction or consolidation chemotherapy preceding the transplant, had a higher day +100 mortality (30.6% vs 6.0%, p<0.0001), 2-year TRM (55.6% vs 16.8%, p<0.0001) and lower 2-year OS (8.3% vs 44.6%, p<0.0001).

Multivariate analysis for 2-year OS showed that history of delayed hematological recovery during induction or consolidation chemotherapy prior to transplantation (HR=1.5, CI=1.0-2.3, p=0.04), high risk genetics (HR=1.8 CI:1.2-2.6, p=0.006), donor-recipient HLA-DRβ3/4/5-DP mismatch (HR=2.2, CI=1.3-3.6, p=0.001), history of cardiovascular disease (HR=1.7, CI=1.1-2.6, p=0.02) were independent predictors for OS. Other variables such as secondary leukemia, CMV sero-status, FEV1 or creatinine clearance prior to SCT, sex mismatch, ABO group mismatch, donor type, or stem cell source did not have a significant impact on OS or TRM. Outcomes were also similar between patients transplanted in CR1 or in ≥ CR2 or in CR or CRi.

We sought to identify those patients who may clearly benefit from a SCT. To that end, we classified patients according to the MRD status prior to transplantation and the presence or absence of the other prognostic factors identified in the multivariate analysis. As seen in the Figure, the High risk group which includes patients with detectable leukemia and >1 additional adverse prognostic factors (or a MMC), had a 2-year OS of 7.7% (CI=3.1-17.8). In comparison, the Low risk group which includes MRD-negative patients with ≤3 other prognostic factors (and no MMC), had a 2-year OS of 76.2% (CI=63.3-85.6). Finally, the Intermediate risk group which constituted the remaining patients, had a 2-year OS of 32.2% (CI=22.1-44.3, p<0.00001).

These data indicate the possibility to identify persons >65 years with high-risk AML likely to benefit from an allotransplant.

Figure.

View largeDownload slide

Figure.

View largeDownload slide

Close modal