Allogeneic transplantation for patients with advanced acute leukemia: a single center retrospective study of 92...

Allogeneic transplantation is a well recognized treatment strategy of leukemia. However, its use in advanced leukemia patients is a subject of some debate especially when donors are not HLA-identical siblings because of the toxicity and cost of the procedure. We reviewed retrospectively the outcome of patients (pts) who received allogeneic transplantation for advanced acute leukemia in our center between 09/86 and 11/97. Thirty-six pts (study group) who lacked a matched sibling donor received partially matched related donor (n=14: PMRD group) or matched unrelated donor transplantation (n=22: MUD group). Fifteen pts had AML and 21 ALL. Seventeen pts (47%) were in CR>1, 13 pts (36%) had refractory disease and six pts (17.7%) were in untreated relapse. The outcome was compared to that of 56 patients (AML: 45.5 %, ALL: 55.5 %, CR>1: 49.9 %, refractory disease: 37.5 %, untreated relapse 19.6 %) who received allogeneic transplantation from a matched sibling donor (control group). Various conditioning regimens and GVHD prophylaxis were used. The actuarial incidence of grade II to IV acute GVHD was significantly higher in the study group (57%) than in the control group (34%) (p=0.047). The actuarial risk of relapse at three years was 21% +/- 22% in the study group versus 65% +/- 16% in the control group (p= 0.04). The actuarial probability of transplant-related mortality at 3 years is 64 +/- 16% for the study group and 25 +/- 11% for the control group (p=0.001). The leading cause of death in the study group was infection (30%) followed by acute GVHD and relapse. Relapse was the major cause of death in the control group (54%), followed by infection, interstitial pneumonia, veno-occlusive disease and GVHD. The OS and probability of leukemia-free survival at 3 years were 28 % +/- 15% (95% CI) and 27% +/- 15% (95% CI) in the study group. The overall survival and probability of LFS at 3 years were respectively 28 +/- 12% (95% CI) and 23 +/- 12% (95% CI) in the control group (p = 0.08 and p=0.11 respectively). In multivariate analysis, transplant-related mortality was higher in the study group (p=0.04) and lower if both donor and recipient were seronegative for CMV (p=0.007). OS was significantly higher for seronegative couples (p=0.0001), and when CR was achieved before BMT (p=0.0022). These results suggest that all efforts in this field should be directed on lowering the transplant related mortality for non geno-identical transplants and the relapse rate in geno-identical transplants.