Antiproliferative and immunomodulatory actions of beta-interferon and double-stranded RNA, individually and in combination, on human bladder tumor xenografts in nude mice.

A cell line, RT4, derived from a human transitional cell carcinoma of the bladder, was grown as a xenograft in athymic mice. The growth of the xenografts was inhibited by beta-interferon (IFN-beta), polyriboinosinic acid.polyribocytidilic acid, the mismatched double-stranded RNA analogue r(I)n . r(C12,U)n, and to a lesser extent recombinant IFN-beta, when treatment was initiated at the time of tumor inoculation. In contrast, the growth rate of established tumors, approximately 6 mm in diameter at the initiation of therapy, was inhibited by both double-stranded RNAs, but not natural IFN-beta, indicating a possible tumor size dependence on the effectiveness of IFN-beta. Combinations of natural or recombinant IFN-beta with either polyriboinosinic acid.polyribocytidilic acid or r(I)n.r(C12,U)n gave an antagonistic effect regardless of tumor mass at the initiation of treatment. This antagonism could be overcome by alternating r(I)n. r(C12,U)n and natural IFN-beta treatment. Natural killer cell activity against RT4 cells in culture was augmented in the spleens of mice treated with r(I)n.r(C12,U)n, but not in those treated with natural IFN-beta. RT4 cells treated in culture with IFN-beta, however, were significantly less efficient as targets for natural killer cells from r(I)n.r(C12,U)n-treated and control spleens. These results indicate that: the effectiveness of IFN-beta may be related to the tumor mass; double-stranded RNAs appear to work, at least partially, in an indirect, immunomodulatory manner; combination therapy can yield an antagonistic rather than an additive or synergistic antitumor effect; and strategic scheduling can overcome the antagonistic effect of combination therapy.