Growth-inhibitory activity of human recombinant beta-interferon (GKT-beta) against 20 human cultured cell lines derived from leukemias and lymphomas was measured quantitatively by regrowth assay. Daudi cells were the most sensitive to GKT-beta. Two T-cell lines (RPMI-8402, HUT78), three B-cell lines (Raji, P3HR-1, A3/Kawakami), one non-T, non-B acute lymphoblastic leukemia (ALL) cell line (KOPN-1) and one monocytoid cell line (U937) were moderately sensitive to GKT-beta. Although the levels of sensitivity of these cell lines to GKT-beta were different, the cells could be killed by GKT-beta. Morphological changes of the sensitive cells treated with GKT-beta were decrease in mitosis, pyknosis and segmentation of cells. Twelve other cultured cell lines, comprising four T-cell lines, four B-cell lines, one non-T, non-B ALL cell line and three myelomonocytoid cell lines, were not sensitive to GKT-beta. The results indicated that the growth-inhibitory activity of GKT-beta was not always cell lineage-specific or differentiative stage-specific. GKT-beta was instable in vitro and its antiviral activity was reduced to about 10% during the first 24 hr of incubation in culture medium with or without cells. This instability was reflected in a similar reduction of its growth-inhibitory activity. It was demonstrated that GKT-beta had a time-dependent, but not a concentration-denpendent antiproliferative action. This suggests that, in the clinical use of the interferon, direct antiproliferative activity of GKT-beta may be expected only through the use of therapeutic schedules which are suitable for its time-dependent action, such as through daily long-term treatment, but not through a single large-dose therapy.