Background : Human melanomas have shown only limited responsiveness to clinical therapy with interferon (IFN). Purpose : Our aim was to determine the most effective class of IFN for inhibiting growth of melanoma cells and to establish whether variation exists in response of various cell lines to different IFNs. Methods : We compared the direct antiproliferative effects of the type I IFN α-2b, IFN α-4a, and IFN-β and the type II IFN-γ on eight melanoma cell lines grown in vitro. We did this comparison by determining the concentration of each IFN that resulted in 50% growth inhibition, using the MTT [3-(4,5-dimethyi-2-thiazolyl)-2, 5-diphenyl-2H-tetrazolium bromide] dye uptake method. We also tested IFN α-2a and IFN-β for their ability to inhibit the growth of xenografts of the LiBr melanoma cell line in vivo in nude mice. Receptor binding was determined using [ 35 S]methionine-labeled IFN α-4a, in competition with unlabeled IFN α-2b, IFN α-4a, and IFN-β. Results : The melanoma cell lines differed markedly in their sensitivity to the IFNs tested: Five were sensitive to low concentrations (>30 pM ) of IFN-P, only one was sensitive to similar concentrations of IFN α-2b, and none were sensitive to IFN a-4a at concentrations up to 920 pM. For all cell lines, the antiproliferative potency of the type I IFNs was IFN-β > IFN α-2b > IFN α-4a. IFN-γ was less active than IFN-P on all except one of the cell lines. Similarly, IFN-β was more potent than IFN α-2a in inhibiting the growth of the LiBr xenograft in nude mice. Labeled IFN α-4a bound with high specificity in all four melanoma lines tested, and competitive binding experiments showed that the order of binding affinity (IFN-β ≶ IFN α-2b ≶ IFN α-4a) correlated with the order of antiproliferative potency. Conclusion : The finding that melanoma cell lines differ intrinsically in their sensitivity to IFNs may explain differences in clinical response. Our results suggest that IFN-P may be the most effective IFN in the treatment of melanoma, although confirmation will require clinical trials involving large numbers of patients. [J Natl Cancer Inst 84:1185–1190, 1992]