Length mutations within the FLT3 gene (FLT3-LM) can be found in 23% of acute myeloid leukaemia (AML) and thus are the most frequent mutations in AML. FLT3-LM are highly correlated with AML with normal karyotype and other cytogenetic aberrations of the prognostically intermediate group. This group is supposed to be a mixed group of AML with differences in the underlying pathogenesis. For more individualized treatment options it would be helpful to better characterize this large AML group not only by molecular mutations but also use these markers for the definition of minimal residual disease (MRD). However, so far the cytogenetically intermediate AML has been lacking suitable markers for PCR-based MRD detection like the fusion genes in the prognostically favourable subgroups. The suitability of the FLT3-LM as a target for PCR-based MRD was discussed controversially as it seemed to be a rather unstable marker. Thus, we aimed at the evaluation of FLT3-LM as a marker for residual disease in a large cohort of AML. Paired samples of 97 patients with AML at diagnosis and at relapse were analyzed. It could be shown that in only four cases a loss of the length mutation was detected. This is in the range of other well-characterized AML relapsing with a different geno- and/or phenotype. In contrast, a change in the ratio of the mutated allele in comparison to the wild-type allele was frequently observed. In detail, the FLT3-LM showed a tendency to accumulate during disease progression and was found more frequently at relapse than at diagnosis. In addition, 45 patients were analyzed at different time points during and after therapy. Using conventional PCR it clearly could be shown that for most of the patients positive at presentation FLT3-LM is a reliable PCR marker for monitoring treatment response. Even an early detection of relapse was possible in some cases. Copyright (C) 2004 S. Karger AG, Basel.