Risk-adapted therapy for AML in first complete remission generally calls for allogeneic BMT for patients with poor risk cytogenetics. However, poor risk cytogenetics is defined differently in commonly-applied classification schemes. We hypothesized that differences in cytogenetic classification might result in differences in survival after BMT. From September 1991 to December 2003, we treated 47 patients with AML in first complete remission with high-dose busulfan-containing preparative regimens and an HLA-matched sibling BMT. The median age was 42 years (range 18 to 60). At the time of diagnosis, 35 patients had either a normal karyotype or at least one clonal abnormality. Cytogenetic analysis was unavailable for the other 12 patients. The 47 patients were then classified according to the SWOG/ECOG (

Blood 96:4075, 2000
), MRC (
Blood 92:2322, 1998
), or CALGB (
Blood 100:4325, 2002
) cytogenetic classification systems and analyzed for overall survival. With a median follow-up of 4.9 years, 3 year overall survival for various cytogenetic risk groups are displayed in the table:

SWOG/ECOGCALGBMRC
Intermediate risk 69% 63% 74% 
Poor risk 43% 42% 18% 
Unknown risk 45% 52% 45% 
SWOG/ECOGCALGBMRC
Intermediate risk 69% 63% 74% 
Poor risk 43% 42% 18% 
Unknown risk 45% 52% 45% 

Concordance indices were 0.58 for CALGB, 0.60 for SWOG, and 0.66 for MRC, demonstrating a slight superiority of the MRC system. Consistent with reports of large, prospective randomized trials of BMT for AML in first remission, patients with poor risk cytogenetics have worse overall survival compared to patients with intermediate risk cytogenetics. Overall survival appears to be different for patients classified as poor risk by MRC criteria. Larger, prospective studies are needed to confirm this observation, but our results suggest that standardized, international cytogenetic risk criteria are needed to develop risk-adapted strategies for the treatment of AML.

Author notes

Corresponding author

Sign in via your Institution