Abstract 2265

Poster Board II-242

While reduced intensity conditioning (RIC) haematopoietic stem cell transplantation (HSCT) regimens have expanded the scope of transplantation, use of these regimens are associated with a higher relapse rate. Disease relapse following following allo-HSCT in acute myeloid leukemia(AML) and high-risk myelodysplastic syndromes(MDS) is associated with a particularly poor prognosis. We performed a retrospective analysis on 157 consecutive patients with high risk (AML/MDS) who received a uniform fludarabine-busulphan based RIC HSCT at our centre between Jul 2002 and Oct 2006 and identified 45(28%) patients who had subsequent cytogenetic or morphological relapse. We proceeded to analyse the patient/disease characteristics of the 45 patients to predict response to salvage therapy. The patient diagnoses were: 19 MDS RAEB I/II, 17 AML, 6 CMML, 3 CML-blast crisis. All patients had received intensive chemotherapy prior to transplantation (median 2 courses, range 1-6). 7 patients were receiving a second allo-HSCT. At time of transplantation, 34 patients had <5% bone marrow blasts, with 11(24.4%) having >5% blasts. 81% of patients had intermediate/poor risk cytogenetics. The median patient age at time of relapse was 52yrs (range:24-72). The median time to relapse was 246 days(range:34-1034). The median survival following relapse was 203 days, with a 2-year OS from time of relapse of 24%+/-6%. On univariate analysis, the time to relapse post-HSCT was the most significant predictive factor of OS post-relapse. All patients who had relapsed <180 days post-HSCT (n=22) died of disease progression within 6 months of relapse. 2-yr OS post-relapse for patients who had relapsed >180 post-HSCT was 43%+/-10% (p<0.01). There was no significant difference in disease type, stage at transplant, cytogenetics between early and late relapsing patients. In addition, patients with >5% blasts at time of allograft had an inferior post-relapse 2-yrs OS when compared with those in remission (9% vs 29%, p=0.1). Patients were further analysed based on the salvage treatment received; no therapy: 5, donor lymphocyte infusions(DLI) only:6, chemotherapy only:12, intensive chemotherapy (defined as a regimen containing high dose arabinose cytarabine) followed by DLI:17, second allograft:5. There was a significant difference in therapy given dependant on the time-post relapse, with 18/22 patients who had early relapse receiving either no therapy, or DLI/chemotherapy alone, while 18/23 late relapsing patients received intensive chemotherapy+DLI or a 2nd allograft. At one year post-relapse, only 1 patient who received DLI only (out of those who received no therapy/DLI only/chemotherapy only) was alive. In contrast, 10/22 patients who received either intensive chemotherapy+DLI or a 2nd allograft were alive and in durable remission with a median survival post-relapse of 1767 days (range:1123-2581), with no significant difference in outcomes between either form of therapy (2 year post-relapse OS: chemo/DLI 47% vs 2nd allograft 40%).

In conclusion, the disease kinetics and time to relapse following RIC HSCT play a significant role in determining subsequent outcomes. The prognosis of patients who relapse within 6 months remains extremely poor. In contrast, we demonstrate that of patients who relapse after 6 months, treatment with chemotherapy followed by cellular therapy (either in the form of intensive chemotherapy followed by DLI, or a 2nd RIC followed by HSCT), can effectively salvage a significant proportion of patients with attainment of durable long-term remissions.


Off Label Use: fludarabine, alemtuzumab- used off-label for transplant conditioning. Marsh:Genzyme: Consultancy, Honoraria.

Author notes


Asterisk with author names denotes non-ASH members.