In the present study, we established a dependable system by which human pre-B- and non-T/non-B-acute lymphoblastic leukemia (ALL) cells are efficiently transplanted into nude mice; the transplanted tumors provide a useful model for investigating the efficacy of antitumor agents in the in vivo therapy of human cancer.
NALM-6 (a pre-B-ALL cell line) cells were transplanted under varying conditions as the pre-B-leukemia cells, whereas REH (a non-T/non-B-ALL cell line) cells were transplanted as the non-T/non-B-leukemia cells. Under optimal and near optimal conditions, 71 of 101 X-irradiated mice (70%) developed distinct tumors approximately 2 wk after i.d. inoculation of a mixture of NALM-6 cells and X-irradiated human fibrosarcoma cells. Under the same conditions, 9 of 11 mice (82%) developed tumors following i.d. inoculation of REH cells admixed with X-irradiated human fibrosarcoma cells. Examination of the tumor tissues demonstrated that the tumors are of leukemia origin but not of fibrosarcoma origin. To demonstrate the usefulness of the present tumors for investigating the efficacy of antitumor agents in the in vivo therapy of human cancer, immunotoxins were tested for their specific suppressive activity against growing tumors of the transplanted NALM-6 cells.
To this end, monoclonal antibodies SN5 and SN6 which define a common ALL antigen, termed CALLA, and a novel leukemia-associated cell surface glycoprotein, termed gp160, respectively, were separately conjugated with the A-chain subunit of ricin, a plant toxin; CALLA and gp160 are expressed on the cell surface of various human non-T-leukemia cells including NALM-6 cells. The conjugates of SN5 and SN6 with ricin A-chain (RA) showed specific activity against the leukemia cells but not against control cells in an in vitro assay. To investigate their in vivo efficacy in suppressing tumor growth, nude mice which had been inoculated i.d. with NALM-6 cells 25 days in advance and bore distinct palpable tumors (5 to 6 mm in diameter) were divided into five groups. One group of mice was nontreated as a control. Each of the remaining four groups of mice was given an injection of one of the following agents: (a) purified control mouse IgG (IgG1); (b) purified antibodies SN5 (IgG1) and SN6 (IgG1); (c) control IgG-RA conjugate; or (d) SN5-RA and SN6-RA. Tumors in all mice of the first four groups including the untreated group grew continuously, causing the mice to die. In contrast, tumors in all five mice of the fifth group which were given injections of specific immuntoxins SN5-RA and SN6-RA regressed completely, although the tumor in one of the mice later relapsed. The tumor-free mice appear to be completely healthy. The strong in vivo cytotoxic action of SN5-RA plus SN6-RA against NALM-6 tumors was reproducible in a repeated experiment. The results show the usefulness of the present human tumor model in evaluating anti-human cancer agents for their in vivo therapeutic efficacy. Furthermore, the results suggest the potential of these immunotoxins for clinical utility.
This work was supported by USPHS Grants CA42683 and CA19304 awarded by the National Cancer Institute, Department of Health and Human Services.