Cytogenetic Patterns of Relapse Following Stem Cell Transplantation in Morphologic Complete Remission in Patients with Cytogenetically Abnormal Acute Myeloid Leukemia

Objectives: Up to one third of patients with acute myeloid leukemia (AML) relapse following allogeneic hematopoietic stem cell transplantation (SCT) in morphologic complete remission (mCR). We hypothesized that that the cytogenetic pattern of relapse is related to cytogenetic risk group at diagnosis. Methods: We analyzed the data from a cohort of patients (n = 45) with AML and abnormal cytogenetics who relapsed following SCT in mCR. Cytogenetic abnormalities were identified by fluorescence in situ hybridization (FISH) and metaphase cytogenetics. Results: All transplants were performed using peripheral blood stem cells. The mean (standard deviation) age of patients was 53 (15) years, and 47% were male. AML was therapy-related in 8 (18%) patients. The most frequent FAB subtypes were M0/M1/M2 (38%) and M4/M5 (38%). Favorable/Intermediate and unfavorable cytogenetic risk disease was present in 11 (24%) and 34 (76%) patients, respectively. Persistent cytogenetic abnormalities despite mCR were present before SCT in 49% of patients. 84% of patients were in CR1. Conditioning was myeloablative in 49% of patients. The donor was a matched sibling, matched unrelated donor, or haploidentical relative in 47%, 51%, and 2% of patients, respectively. Primary graft failure did not occur in any patient. 29% and 4% of patients had CMV reactivation and disease, respectively. Acute and chronic GvHD occurred in 27% and 13% of patients, respectively. The median time to relapse and overall survival from the time of transplant were 3.5 (interquartile range: 1.7-5.3) and 10 (8.9-11.2) months, respectively. Three patterns of relapse were identified: same clone (n = 23, 51%), subclone (n = 17, 38%), and new clone (n = 5, 11%). The latter two were grouped together during analysis because both represented clonal evolution. The only difference in the baseline and transplant characteristics between patients who relapsed with the same clone versus those who relapsed with clonal evolution was cytogenetic risk group. Specifically, patients with high risk cytogenetics at diagnosis were more likely to relapse with the same clone than those with low/intermediate risk cytogenetics (62% vs. 18%, respectively; P = 0.017). There was no difference in median time to relapse (same clone, 3.2 months vs. clonal evolution, 3.5 months, P = 0.67) or overall survival (9.4 vs 10.0 months, respectively; P = 0.88) based the type of clone at the time of relapse. Conclusions: In patients with AML and abnormal cytogenetics who undergo SCT in mCR, the same clone is responsible for relapse in a significantly higher proportion of those with unfavorable risk cytogenetics at the time of AML diagnosis compared to those with favorable/intermediate risk disease. This may be due to higher rates of minimal residual disease after SCT in patients with unfavorable risk disease. Next-generation sequencing may add finer resolution to our findings.