Prognostic Factors for Outcome of Patients Undergoing Allogeneic HSCT from Unrelated and Related Donors for AML and MDS: A Retrospective Analysis.

PURPOSE: Allogeneic hematopoietic stem cell transplantation (HSCT) is the only curative therapy for many patients with myeloid leukemia, especially in refractory or relapsed disease. To determine prognostic factors that correlate with outcomes after HSCT in adults with AML and MDS we analysed several clinical and haematological parameters.

PATIENTS AND METHODS: In this unicenter retrospective study, we studied 120 adults who received allogeneic HSCT for AML (n=88) or MDS (n=32) between 1993 and 2003. The median age was 44 years (range 21–72), included were 59 male and 61 female patients. 42% of the patients were transplanted in first complete remission (CR1), 17% were transplanted in CR after salvage therapy and 41% were in relapse or refractory to induction or salvage therapy. The grafts consisted in 88 cases (73%) of PBSCT and in 32 cases of bone marrow, in 34% unrelated donors were used. The medium number of infused CD34+ cells was 5.51x106/kg. As conditioning regimen standard myeloablative therapies without (53%) or with (25%) TBI, or a non-myeloablative regimen including TBI (22%) were used. The prognostic impact for several groups with different clinical and hematological parameters were assessed by univariate and multivariate analyses using the Cox regression model.

RESULTS: One-year and three-year probabilities of overall survival (OS) were 52% (48% to 57%) and 38% (36% to 43%). In univariate analysis, remission status and cGvHD were significantly associated with OS (CR vs. non-CR p=0.00001; cGvHD vs. non-cGvHD p=0.00001). Median OS for patients who underwent transplantation in CR were 2374 vs 180 days and for chronic GvHD 2374 vs 294 days. One-year and three-year probabilities of relapse were 30% (31% to 17%) and 31% (34% to 24%). CR, cGvHD and ablative regimens resulted in a lower relapse rate (p=0.00001; p=0.0002, p=0.028). One-year and three-year probabilities of transplant-related mortality (TRM) were 27% (28% to 30%) and 34% (32% to 40%). Extensive cGvHD, transplantion before 1997 and unrelated donor were significantly associated with higher TRM (p=0.03; p=0.04; p=0.03). Multivariate analysis was performed to test for the prognostic role of those factors that had achived a p-value of 0.25 or less in the Log-Rank test. For OS, sex, donor, tbi, remission status, and cGvHD and time from diagnosis to transplantation were included. For a better OS and lower RR especially the remission status and the occurrence of cGvHD were significant prognostic factors for the outcome after allogeneic transplantation (OS: p=0.0001, p=0.001; RR: p=0.0001, p= 0.001); sex and time from diagnosis to transplantation were also predictive for a better OS (p=0.007, p= 0.045) and the use of non-tbi based regimens resulted in a lower RR (p=0.006); for TRM, transplantions before 1997 and unrelated donors were detect as significant prognostic factors (p=0.027, p=0.033).

CONCLUSION: Remission status at transplantation and the occurrence of cGvHD are the major prognostic factors for the outcome after allogeneic HSCT in AML and MDS. We therefore conclude that allogeneic stem cell transplantation should be performed early in haematological remission in patients with high-risk AML and MDS, if a suitable donor is available.