Summary We describe the application of a simple, rapid, semi-automated assay to the sensitivity testing of cytotoxic drugs in 23 patients with acute myeloid leukaemia (AML). The survival of blast cells from the bone marrow was measured by the MTT assay after 48h continuous exposure to drugs both singly and in combination. There was a linear relationship between the number of leukaemic ceils and the optical density of the Cytotoxic drug therapy remains the prime method of treatment in acute myeloid leukaemia. One of the recognised limitations of this therapy, however, is the inability to predict tumour sensitivity in individual patients. Most attention in the field of haematological malignancies has focused on the clonogenic assay (Marie et al., 1987), dye exclusion assays (Weisenthal et al., 1986;Bird et al., 1988) and radioactive precursor incorporation (Schwarzmeier et al., 1984;Raza et al., 1987). The advantages and disadvantages of these methods are well documented (Hill, 1983; Wiesenthal & Lippman, 1985). A simple rapid chemosensitivity test suitable for automation is what is required for routine use. The clonogenic assay is long-term, effectively measuring the chemosensitivity of dividing cells only. Evidence is emerging to suggest that non-clonogenic assays which measure cell kill in the total blast cell population may be equally valuable. The most promising of these, the dye exclusion assay, shows a good correlation with the end-point of the clonogenic assay (Weisenthal et al., 1983). However, it is not an automated technique and is therefore timeconsuming and subject to observer error.In 1983 Mosmann described a semi-automated colorometric assay based on the premise that the mitochondria of living cells reduce the tetrazolium salt MTT to formazan. A modified form of this is currently being successfully applied by the National Cancer Institute USA to the chemosensitivity testing of new drugs on cell lines (Alley et al., 1988). The technique has been adapted for chemosensitivity testing of chronic (Twentyman et al., 1989) and acute (Pieters et al., 1988) lymphatic leukaemia cells. Results compared favourably to those using the differential staining cytotoxicity (DiSC) assay, a dye exclusion technique (Pieters et al., 1989).It is important to validate this assay for each cell type, and consequently we describe its application to the chemosensitivity testing of blast cells from the bone marrow of patients with acute myeloid leukaemia. These patients have a poor prognosis even in those who achieve remission. A simple in vitro method aiding initial selection of drugs both singly and in combination and permitting retesting throughout remission induction and on relapse would be a therapeutic advance. Thirteen patients were assayed both on presentation and throughout the course of the disease, two after remission was already established, three after relapse and the remaining five patients tested initially had no follow-up as they did not survive beyond the first course of treatment.
Methods
Preparation of ce...