Lack of cell-cycle-specific effects of recombinant tumor necrosis factor in vivo.

Several in vitro studies have demonstrated that tumor cells arrested in the G2 and M phases of the cell cycle expressed an increased sensitivity to the tumor necrosis factor (TNF). The scope of the present study was to investigate whether this cycle dependence of TNF effects also exists in vivo. The experiments were performed by using the Lewis lung carcinoma (LLC), which had been allotransplanted to nude mice. In order to induce delays of the tumor cell cycle in G2, the animals were treated with etoposide (40 mg/kg body weight i.p.) or with local radiation (15 Gy), each increasing the G2 fraction of the LLC from 10% to 35% and 50% respectively. For combination therapy with recombinant (r)TNF, the tumor was transplanted to four groups of six mice each, one of them serving as a control group the others being treated either with a G2 inductor alone, with rTNF alone, or with rTNF and a G2 inductor combined. Administration of rTNF (125 or 250 micrograms/kg body weight i.v.) was always carried out 24 h after therapy with etoposide or radiation when the maximum of G2 accumulation had developed. The growth behavior of the treated tumors revealed that the response of the LLC to rTNF in vivo was not improved by pretreatment with a G2 inductor and, thus, obviously lacked cell-cycle specificity. It is supposed that direct interactions of TNF with the tumor cells, which are a basic requirement for cell-cycle-linked phenomena, play a minor role in the therapeutic outcome of the LLC under in vivo conditions.