The Hyal-l locus, which we have previously described and mapped to mouse chromosome 9, influences the serum levels and molecular weight forms of hyaluronidase. We have also shown that the growth of two transplantable tumors, the 3LL carcinoma and the B16F10 melanoma, is influenced by the alleles at Hyal-1, in that the tumors develop more slowly in congenic B6.C-Hyal-1a (also called HW23) mice than in the parental Hyal-lb C57BL/6 mice. Here we present evidence that tumor development is stimulated and mortality is accelerated in B6.C-Hyal-la mice grafted with 3LL carcinoma cells when treated with a/(3 interferon (IFN-a/13) or with IFN-f, whereas in IFN-treated C57BL/6 mice 3LL tumor growth is inhibited. Likewise, in B6.C-Hyal-la mice grafted with B16F10 melanoma cells, IFN-a/P treatment results in stimulation of tumor growth, whereas in IFN-treated C57BL/6 mice tumor growth is inhibited and mortality delayed. Thus, IFN-a/,B treatment of B6.C-Hyal-la mice results in stimulation of tumor development and sometimes in accelerated mortality. This is the opposite of the usually described effect of IFN treatment in mice, which is inhibition of tumor development and delayed mortality, as was indeed observed in the C57BL/6 mice in the present experiments. These results provide the first indication that host genes can up-or down-regulate the antitumor activity of IFN and that, on some genetic backgrounds, IFN treatment enhances rather than inhibits tumor development. This may help to explain the apparent discordance between mouse model studies, which hitherto have consistently reported inhibition of tumor formation by IFN, and the clinical trials, in which only a limited percentage ofindividuals show tumor regression while others have no beneficial effect or even have progression of disease in spite of the IFN treatment.Because of their manifold interesting biologic activities, an increasing number of cytokines are used currently in the clinic or are being considered for such use. Most cytokines act on a great variety of cells and can up-or down-regulate the synthesis of many different proteins, and, probably as a result of this complex activity, pronounced individual variations are observed in humans treated with these substances (1). Based on their antitumor activities, the first cytokines to have found large-scale use in the clinic were the interferons (IFNs), and human IFN-a and -/ are currently administered to a great variety of patients with various tumor conditions for which IFN treatment has shown some effectiveness-for example, metastatic melanoma, different forms of carcinoma, myeloma, and ovarian cancers (2). Tumor regression and clinical improvement are observed only in a limited percentage of treated individuals, whereas in others, in spite of identical treatment dosage and regimen and tumors of apparently comparable histological type, there is tumor progression (3). It would be of obvious interest to understand the differences in the individual responses to IFN treatment in particular and to cytokine t...