Chemo-adoptive immunotherapy of nude mice implanted with human colorectal carcinoma and melanoma cell lines.

The antitumor effects of chemotherapy, recombinant human interleukin-2 (IL-2), recombinant human interferon alpha A/D (IFN alpha), allogeneic human lymphokine-activated killer (LAK) cells, and antitumor monoclonal antibody (mAb), administered alone and in various combinations, were tested in athymic nude mice carrying human tumor xenografts. Treatment began 6-18 days after i.v. or i.p. inoculation of colorectal carcinoma or melanoma cell lines, when macroscopic growths were evident. Chemotherapy consisted of two or three courses of 5-fluorouracil (5-FU) or dacarbazine. IL-2 and/or IFN alpha were administered three to five times weekly for 1-3 weeks, usually starting 2-5 days after chemotherapy. Human LAK cells were infused once or twice weekly for 2 or 3 weeks concurrently with IL-2. In some experiments, murine anticolorectal carcinoma mAb (SF25) was administered. In both tumor systems, chemotherapy alone or immunotherapy alone (IL-2, IL-2 + LAK cells, IFN alpha, IL-2 + IFN alpha +/- LAK cells) had little or no therapeutic effects. Additive effects were obtained by combining chemotherapy with IL-2 and LAK cells or with IL-2 and IFN alpha. In the majority of the experiments, the most effective combination was chemotherapy + IL-2 + IFN alpha + LAK cells. Treatment with mAb was beneficial in the colorectal carcinoma system when combined with 5-FU + IL-2 or 5-FU + IL-2 + IFN alpha. Homing experiments with radiolabeled human and mouse LAK cells injected i.v. showed increased early accumulation in the liver and lungs, whereas freshly explanted mouse splenocytes localized mostly in the spleen and liver. The tissue distribution pattern of human LAK cells was similar in normal and tumor-bearing mice (with lung metastases). These findings suggest that combination of chemotherapy with cytokines and LAK cells can be partially effective for advanced solid human tumors even in the absence of the host's T-cell immune response. Preliminary experiments showed that tumor-specific, anti-melanoma T-cell clones were effective in local (s.c.) tumor growth inhibition (Winn assay) following coinjection with the autologous tumor cells.