A vital dye exclusion method following a 4 day tumor sensitivity assay was used to predict clinical response to chemotherapeutics in 16 patients with acute nonlymphoblastic leukemia (n = 12), acute lymphoblastic leukemia (n = 2), diffuse poorly differentiated lymphoma (n = 1) and oat cell carcinoma (n = 1). After 4 days of incubation, controls and drug-supplied cells were stained with fast green and hematoxylin-eosin. A tumor cell survival of 30% was used to distinguish in vitro resistance and in vitro sensitivity. Eight patients responded and five did not respond to the chemotherapy protocol. This agreed with the in vitro results. Two patients who were sensitive in vitro responded only partially in vivo but two additional patients sensitive in vitro did not respond. These results indicate that this short-term method may give a simple estimation of drug resistance.