Abstract
Abstract 3079
Patients with acute myeloid leukaemia secondary to myelodysplastic syndromes (sAML) are often considered to have inferior outcomes to patients with de novo AML. This may be attributed in part to the fact that sAML is associated with an older age of onset and more frequent presence of adverse cytogenetics. While reduced intensity conditioning haematopoietic stem cell transplantation (RIC-HSCT) offers a potential cure for patients with sAML, long term outcomes in this subgroup of patients is limited. Here we present the single centre 5-year outcomes of patients with sAML receiving an alemtuzumab based RIC-HSCT.
Patients were considered to have sAML defined as per WHO 2008 criteria. Patients were included based upon the presence of tri-lineage dysplasia at diagnosis (TLD-AML) or a documented history of preceeding MDS, prior cytopenias, or MDS related cytogenetic abnormalities (prior MDS to AML, MDS-AML). Patients with therapy-related AML were excluded. Cytogenetic risk groups were classified as per European Leukaemia Net criteria.
84 patients were identified, 38 female and 46 male. The median age of the cohort was 58 years (range: 21–71). Diagnoses included 23 patients with a history of antecedent MDS and 60 patients who presented as TLD-AML. Cytogenetic abnormalities were intermediate risk in 58 (69%) patients and poor risk in 23 (27%) patients, data was unavailable for 3 patients. All patients received a uniform RIC protocol consisting of fludarabine, busulphan and alemtuzumab from sibling (n= 18) or volunteer unrelated donors (VUD, n=66). 20 VUD had a 1-antigen mm donor. GVHD prophylaxis was with cyclosporine. Patients received PBSC (n=66) or BM (n=22) with a respective median CD34+ cell dose of 6.28 × 10 ^6/kg and 3.02 ×10 ^6/kg. All patients received prior induction chemotherapy with 59 patients in CR1, and 20 patients in CR2 at time of transplantation. Only 4 patients had persistent disease (>5% blasts) at time of transplant.
Median follow-up for survivors was 63.5 months (range :1.6–128.2). Neutrophil engraftment occurred at a median of 13 days(range 8–31). Primary graft failure occurred in only 1 patient (1%). 5-yr OS, and DFS for the entire cohort was 47% and 37% respectively. TRM was 4% at Day 100 and 21% at 5yrs. Relapse rate at one year was 32% and at five years 56%. Acute graft versus host disease occurred in 32% of patients with grade III-IV GVHD occurring in 19%. Rates of de-novo chronic GVHD were low with extensive chronic de-novo GVHD occurring in only 9% of patients. DLI was administered to 30 (36%) patients, 22 patients for low or falling CD3 chimerism and 8 for disease relapse.
On univariate analysis, patients with HCTCI <2 had improved 5-yr OS and DFS when compared to those with HCTCI >/= 2; 61% vs 35% (p=0.03) and 50% vs 25% (p=0.02) respectively. MDS-AML was associated with significantly worse outcomes when compared with TLD-AML, (5-yr OS; 35% vs 53%, p=0.02 and 5-yr DFS; 19% vs 45%, p=0.01). Outcomes for younger patients are very good with 5yr OS of 58% for age <60 yrs (n=50), compared to 30% for age >60yrs (n=34) (p = 0.03). Likewise 5-yr DFS by age group was 49% vs 22% respectively (p <0.01). In contrast, donor type (sibling vs unrelated), stem cell source (PBSC vs BM), the presence of HLA mismatch or cytogenetic abnormalities (intermediate vs poor risk) did not have a significant effect on OS or DFS.
On multivariate analysis, advanced age (>60 yrs) was the only independent variable associated with an inferior TRM. Advanced age and presence of MDS-AML were the only factors independently associated with an inferior 5-yr OS (OS: Advanced Age: HR 1.82 95%CI :1.15–2.89, p=0.11, MDS-AML: HR :2.04 95%CI :1.08–3.87, p=0.03), and 5-yr DFS (OS: Advanced Age: HR 1.88 95% CI :1.24–2.87, p<0.01, MDS-AML: HR :2.09 95%CI :1.17–3.73, p=0.01).
This analysis demonstrates excellent long-term outcomes for younger patients with sAML receiving RIC-HSCT with high engraftment and low rates of GVHD. Younger patients with TLD-AML have very good 5 yr OS and DFS of 60% and 53% respectively compared with older patients with MDS-AML who have a corresponding 5yr OS and DFS of 13% and 0% respectively. Novel strategies are required to improve outcomes in older age groups and for those patients with a prior history of MDS.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.