The automated fluorometric microculture cytotoxicity assay (FMCA) was used for chemotherapeutic drug sensitivity testing of fresh and cryopreserved tumor cells from patients with acute lymphoblastic leukemia (ALL) at diagnosis and relapse. The technique success rate was 87% for fresh and 81% for cryopreserved samples. Up to 16 different cytotoxic drugs were routinely tested, but neither asparaginase nor methotrexate produced dose-response related cell kill. FMCA data showed good correlation to the well established Disc assay and the drug sensitivity reported by the FMCA was in good agreement with known clinical activity. Samples from children and initial ALL tended to be more drug sensitive than those from adults and ALL at relapse, respectively. For 36 samples clinical outcome was correlated to the quartile position in comparison to all other samples for the most in vitro active drug actually given to the patient. For patients with samples in the first, second, third, and fourth quartiles, the probabilities of complete remission were 89, 57, 38, and 0%, respectively. Using the median value as cut-off line, the sensitivity and specificity of the assay were 87 and 62%, respectively. It is concluded that the FMCA with a minimum of effort and with high success rate report clinically relevant drug sensitivity profiles for ALL.