Cellular drug resistance profiles in childhood acute myeloid leukemia: differences between FAB types and comparison with acute lymphoblastic leukemia.

Determining in vitro drug resistance may reveal clinically relevant information in childhood leukemia. Using the methyl-thiazol-tetrazolium assay, the resistance of untreated leukemic cells to 21 drugs was compared in 128 children with acute myeloid leukemia (AML) and 536 children with acute lymphoblastic leukemia (ALL). The differences between 3 French-American-British (FAB) types (M1/M2, M4, and M5) were also compared. AML was significantly more resistant than ALL to the following drugs, as noted by the median resistance: glucocorticoids (greater than 85-fold), vincristine (4.4-fold), L-asparaginase (6.9-fold), anthracyclines (1.8- to 3.4-fold), mitoxantrone (2.6-fold), etoposide (4.9-fold), platinum analogues (2.4- to 3.4-fold), ifosfamide (3.5-fold), and thiotepa (3.9-fold). For cytarabine and thiopurines, the median LC50 values (the drug concentration that kills 5% of the cells) were equal. Also, busulfan, amsacrine, teniposide, and vindesine showed no significant differences, but the numbers were smaller, and the median LC50 values were 1.3- to 5.2-fold higher in AML. None of the drugs demonstrated greater cytotoxicity in AML. FAB M5 was significantly more sensitive than FAB M4 to most drugs frequently used in AML, as indicated by the following ratios of median sensitivities: the anthracyclines (2.6- to 3.2-fold), mitoxantrone (12.5-fold), etoposide (8.7-fold), and cytarabine (2.9-fold). For etoposide and cytarabine (5.4- and 3.4-fold, respectively) FAB M5 was also significantly more sensitive than FAB M1/M2. FAB M5 was equally sensitive to L-asparaginase and vincristine as ALL. Only 15% of the AML samples were "intermediately" sensitive to glucocorticoids, mainly in FAB M1/M2. The poorer prognosis of childhood AML is related to resistance to a large number of drugs. Within AML, FAB M5 had a distinct resistance pattern. These resistance profiles may be helpful in the rational design of further treatment protocols. (Blood. 2000;96:2879-2886)