Impact of Pre-Transplant Minimal Residual Disease Assessed by Flow Cytometry on Outcome Following Myeloablative Hematopoietic Cell Transplantation for Patients with AML-CR1.

Abstract

Allogeneic hematopoietic cell transplantation (HCT) often offers the best and sometimes only chance for cure for acute myeloid leukemia (AML) patients. However, the decision of whether to transplant a patient with AML while in first remission remains a difficult one, and a great deal of attention has been given to attempting to identify factors that will predict HCT outcome. While many patient- and disease-specific factors have been studied, little attention has been given to the impact of the presence of MRD as determined by multiparametric flow cytometry (MFC) at the time of presentation for transplant on outcome. We have therefore evaluated outcome among patients transplanted for AML in CR1, comparing those with evidence of MRD as determined by MFC compared to those with no evidence of MRD prior to myeloablative HCT. Between 1992 and 2004, 140 patients greater than 18 years of age with AML-CR1 received a myeloablative HCT at our Center. Immediately prior to HCT, 23 CR1 patients had evidence of MRD (MRD+) whereas 117 CR1 patients had no evidence of MRD (MRD-) by flow cytometric analysis. Four (17%) MRD+ patients and 23 (20%) MRD- patients had an unrelated donor. Median age was 48.1 and 40.7 years for MRD+ and MRD- patients, respectively. Conditioning included use of total body irradiation (n=8 MRD+; n=37 MRD-), busulfan and cyclophosphamide (n=8 MRD+; n=42 MRD-), an anti-CD45 radiolabeled antibody (n=6 MRD+; n=34 MRD-), or other regimens (n=1 MRD+; n=4 MRD-). Most received cyclosporine and methotrexate for GvHD prophylaxis (96% MRD+; 97% MRD-). Source of stem cells was bone marrow in 39% MRD+ and 58% MRD- patients. Patients were further classified by cytogenetic risk as favorable (0% MRD+; 3% MRD-), intermediate (61% MRD+; 74% MRD-), or unfavorable (30% MRD+; 18% MRD-) (cytogenetics were missing for 2 MRD+ patients and 5 MRD- patients). At last follow-up, 16 (70%) MRD+ patients had died compared to 46 (39%) MRD- patients. Five-year estimates of overall survival were 35% and 64%, respectively. After adjusting for age, source of stem cells, donor, and unfavorable cytogenetics, the hazard ratio (HR) for mortality was 2.13 (95% CI 1.11–4.08, p=.02). Nine (39%) MRD+ patients relapsed after HCT compared to 25 (21%) MRD- patients and the 5-year estimates of relapse were 39% and 21%, respectively. The adjusted HR or relapse was 2.83 (1.19–6.73, p=.02). Thirty percent (n=7) of patients in the MRD+ group experienced transplant-related mortality (TRM) compared to 21% (n=25) in the MRD- group [adjusted HR=1.50 (0.54–4.19, p=.44)]. Among those who were MRD+, there was a positive correlation between the percentage of flow blasts and the risk of mortality and TRM. However, there was no suggestion of an association with relapse in this group. When modeling flow blasts as a continuous linear variable, each increase in flow blasts of 10% was associated with an increase in the hazard of mortality [HR=1.20 (1.00–1.45, p=.05)] and TRM [HR=1.39 (1.10–1.76, p=.006)], but not relapse [HR=0.77 (0.35–1.70, p=.51)]. While these data need to be confirmed in a larger number of patients, they suggest that any evidence of MRD+ at HCT by flow cytometry in AML-CR1 patients increases the risk of relapse and death after HCT relative to MRD- patients.