Techniques for detection of minimal residual disease

Analysis of leukemia-specific and leukemia-associated markers following standard or high-dose treatments is crucial in order to evaluate the efficacy of therapeutic strategies. During the last decade, several techniques have been proposed and used for detecting minimal residual disease (MRD). Each approach is characterized by advantages and limitations, mainly related to its sensitivity and specificity. The general limitations of such tests originates from the size of the sample which can be analysed and the heterogeneous distribution of leukemia after treatment. Clinically useful methods for detecting residual leukemia require not only sensitivity but also speed and reproducibility. The rate of false negative tests is low with polymerase chain reaction as well as flow cytometric analysis. Usually, patients without persistent cells carrying leukemia-associated markers have a lower risk of relapse. However, the detection of a persistent marker at one time point in complete remission cannot be considered a reliable indicator of MRD, whereas increase of positive signals or reappearance of leukemic markers usually precedes relapse. It is likely that one single approach will not allow the monitoring of the majority of patients and that a combination of techniques will be needed. Definitive results will be obtained only through prospective studies in patients receiving standardised therapy. Studies in which therapeutic strategies are designed according to the results provided by techniques for detecting MRD will be necessary to assess the relevance of their contribution to the treatment of leukemia.