Search

Our recent studies using IL-12 protein or fibroblasts genetically engineered to secrete IL-12 have demonstrated profound antitumor effects of IL-12 in murine models. The antitumor effects of local, high level IL-12 expression were examined using a retroviral vector, which can express both IL-12 subunits (p35 and p40) and the neomycin phosphotransferase (Neo)-marker gene from a polycistronic message utilizing internal ribosome entry site sequences. All animals intradermally (i.d.) receiving MCA207 murine sarcoma cell line nontransfected or Neo-transfected had progressively growing tumor, whereas all animals injected with MCA207 transfected with IL-12 were tumor free and were subsequently determined to be immune to a rechallenge of nontransfected MCA207 i.d. Similar results were obtained in experiments using the poorly immunogenic MCA102 murine sarcoma cell line. The inoculation of live MCA207-IL-12 tumor cells also caused the regression of contralateral nontransfected MCA207 inoculated either at the same time (80% protection) or up to 3 days before (33% protection) to the therapeutic tumor inoculation. In vivo depletion studies suggest that NK cells and IFN-gamma play important roles in the development of the early phase of the antitumor response, but that T cells (both CD4+ and CD8+) play the major role in the subsequent events, leading to long-term immunity. The potent antitumor effects observed for paracrine gene-delivered administration of IL-12 have thus been confirmed for multiple tumor cell types and in multiple murine strains. We believe that these results support the feasibility of IL-12 gene therapy for the treatment of human cancer.

Using a newly described pH 3.3 acid elution technique, peptides were extracted by denaturation of class I molecules on the surface of human melanomas. HPLC fractionation of this material revealed six T cell epitopes (termed P1-P6) recognized by HLA-A2-restricted, melanoma-specific tumor infiltrating lymphocyte (TIL) lines. Three of these fractions (P1, P2, and P4) appeared to represent shared/immunodominant melanoma Ag recognized in the context of HLA-A2 because they were expressed by 4/4 HLA-A2+ melanoma cell lines and were each recognized by all four oligoclonal HLA-A2-restricted TIL lines examined. Interestingly, P1 and P2 (but not P3-P6) could also be recognized by these same TIL when presented by the HLA-Aw69 class I molecule, which is closely related to HLA-A2. P3, P5, and P6 displayed more restricted expression and were differentially recognized by the four oligoclonal TIL lines. These results suggest that synthetic peptide derived from P1, P2, and P4 sequences (when deduced) may form the basis of effective prophylactic or therapeutic melanoma vaccines by stimulating CD8+ CTL in HLA-A2+ individuals. This approach of identifying T cell epitopes presented by class I molecules should prove generally applicable to the study of other tumors recognized by class I-restricted CTL.