It is generally assumed that prednisolone (PRD) and dexamethasone (DXM) have equal glucocorticoid activity if PRD is given at sev enfold higher doses. Resuits of clinical studies of childhood acute lymphoblastic leukemia (ALL) suggested that DXM is more potent relative to PRD than assumed. The purpose of this study was to determine the relative antileukemic activ ity of PRD phosphate and DXM phosphate in 133 untreated childhood ALL samples in vitro, using the 3-(4,5-dimethylthiazol-2~y!)-2,5~diphenyltetrazolium-bromide (MTT) assay, There was a marked variation in antileukemic activity of both agents among the patient samples. The median LC50 (drug concentration lethal to 50% of the ALL cells) for PRD phosphate was 3.50 |jlM, for DXM phosphate 0,20 jxM. The individu ally calculated ratios oftheLCSO values for PRD and DXM phosphate showed a large range from 0.7 to > 5 0 0 ; with a median of 16.2. This 16-fold difference could not be explained by differences betw een these glucocorticoids in stability, hy drolysis into unesterified drug, adhesion to the wall o f the microculture plates, or protein bind ing. ALL cells were cross-resistant to PRD and DXM phosphate (correlation coefficient = 0.85, P < 0.000001).W e conclude that the in vitro antileukemic activity o f DXM phosphate is median 16-fold higher than that of PRD phosphate, which contrasts to the generally assumed factor of 7, Based on the higher potency of DXM, and its more favorable pharmacokinetics as reported in the literature, DXM may be pre ferred to PRD as the glucocorticoid in the treatment of ALL.