Fibrosarcoma-SA-I -tumor-bearing mice were treated S.C. in the vicinity of tumors (peri-tumorally) or intravenously, with recombinant human TNF-a lacking I to 3 amino acids from N-terminal part (TNF-aNv3). Tumor growth delay, observed after both routes of TNF-aNv3 application, was statistically significant, though a better anti-tumor effect was achieved after peri-tumoral application. TNF-aNv3 serum levels were determined in these animals and compared with TNF-aNv3 serum levels in healthy animals, which were treated with TNF-olNv3 either S.C. or i.v. The peak serum levels of TNF-aNv3 applied peri-tumorally/s.c. were significantly higher in tumor-bearing than in healthy mice, whereas smaller differences in peak serum levels were found after i.v. application, which might correlate with anti-tumor activity. Whatever the route of application, TNF-aNv3 elimination from the serum of tumor-bearing mice was slower than that in healthy animals. Also, comparison of TNF-aNv3 pharmacokinetic parameters for tumor-free and sarcoma-or melanoma-bearing mice has demonstrated that the pharmacokinetics of TNF-aNv3 are modified in tumorbearing animals.
o I993 Wiley-Liss, bzc.
Tumor necrosis factor-a, a protein secreted by mononuclear phagocytes in response to various stimuli (Matthews, 1981; Agganval et al., 1985), was recognized as an efficient anticancer drug, because it causes hemorrhagic necrosis of some tumors in vivo and lysis of cancer cells in vitro (Carswell et al., 1975;Helson et al., 1975). Later, it was established that TNF-a initiates multiple effects on cells (Beutler and Cerami, 1988; Old, 1989; Stalcet al., 1992) and that excess of TNF-CY can have profoundly deleterious effects in animals and humans.
Clinical trials with TNF-a have been complicated in cancer patients by toxic side effects after its application (Blick et al., 1987; Kramer and Sherwin, 1989). To define the best route of TNF-a administration and dosage schedule in order to diminish the side effects, it is important to know the pharmacokinetic mechanisms of TNF-a.
Despite the fact that the action of TNF-a is not without species specificity, as was originally believed, animal models were often used to study the pharmacokinetics of human TNF-a (