Identification and epitope enhancement of a PAX-FKHR fusion protein breakpoint epitope in alveolar rhabdomyosarcoma cells created by a tumorigenic chromosomal translocation inducing CTL capable of lysing human tumors.

Fusion proteins created by chromosomal translocations in tumors can create neoantigenic determinants at the breakpoint, which are unique to the tumor cells but shared by the vast majority of tumors of that histologic type. If the fusion protein is responsible for the malignant transformation, its expression cannot be lost by the tumor to escape immune responses against this tumor antigen. Here, we identify such a fusion protein breakpoint epitope in the PAX-FKHR fusion protein created by the t(2;13) translocation present in 80% of cases of alveolar rhabdomyosarcoma, a highly aggressive pediatric soft-tissue sarcoma. We use autologous dendritic cells pulsed with the RS10 breakpoint fusion peptide to raise a human CTL line from a normal healthy HLA-B7+ blood donor specific for this peptide. These CTLs are CD8+ (CD4-CD56-) and restricted by HLA-B7. These human peptide-specific CTL lyse human HLA-B7+ rhabdomyosarcoma tumor cells. Therefore, the fusion protein is endogenously processed to produce this natural epitope presented by HLA-B7 and thus this peptide is a bone fide human tumor antigen. We also define a substitution that increases the affinity for HLA-B7 without loss of antigenicity. This epitope-enhanced peptide may serve as a candidate cancer vaccine for HLA-B7+ patients with alveolar rhabdomyosarcoma.

Unmasking immunosurveillance against a syngeneic colon cancer by elimination of CD4+ NKT regulatory cells and IL-13.

We have previously observed a novel role of natural killer T (NKT) cells in negative regulation of antitumor immune responses against an immunogenic regressor tumor expressing a transfected viral antigen. Here, we investigated whether hidden spontaneous antitumor immunosurveillance, in the absence of a vaccine, could be revealed by disruption of this negative regulatory pathway involving CD4+ NKT cells and interleukin-13 (IL-13), in a murine pulmonary metastasis model of a nontransfected, nonregressor, syngeneic tumor, the CT26 colon carcinoma. Lung metastases of CT26 were decreased in CD4+ T cell-depleted BALB/c mice, suggesting that CD4+ T cells were involved in negative regulation of antitumor responses. CD1-knock out (CD1-KO) mice, which have conventional CD4+ T cells and CD4+CD25+ regulatory T cells but lack CD1-restricted CD4+ NKT cells, were significantly resistant to lung metastasis of CT26. The metastases were not further decreased in CD4+ T cell-depleted CD1-KO mice, implying that CD4+ NKT cells might be the primary negative regulator of antitumor immune responses in BALB/c mice. CD8+ T cells were found to act as effectors in antitumor immune responses, since the inhibition of lung metastases observed in naive CD1-KO or CD4+ T cell-depleted mice was abrogated by depletion of CD8+ T cells. Lung metastases were significantly decreased by treatment of mice with an IL-13 inhibitor, but not by deficiency or inhibition of IL-4. Thus, even for a nonregressor tumor, immunosurveillance exists but is negatively regulated via CD4+ NKT cells possibly mediated by IL-13, and can be unmasked by removal of these negative regulatory components.

Dendritic cell-induced activation of adaptive and innate antitumor immunity.

While studying Ag-pulsed syngeneic dendritic cell (DC) immunization, we discovered that surprisingly, unpulsed DCs induced protection against tumor lung metastases resulting from i.v. injection of a syngeneic BALB/c colon carcinoma CT26 or a syngeneic C57BL/6 lung carcinoma LL/2. Splenocytes or immature splenic DCs did not protect. The protection was mediated by NK cells, in that it was abrogated by treatment with anti-asialo-GM1 but not anti-CD8, and was induced by CD1(-/-) DCs unable to stimulate NKT cells, but did not occur in beige mice lacking NK cells. Protection correlated with increased NK activity, and increased infiltration of NK but not CD8(+) cells in lungs of tumor-bearing mice. Protection depended on the presence of costimulatory molecules CD80, CD86, and CD40 on the DCs, but surprisingly did not require DCs that could make IL-12 or IL-15. Unexpectedly, protection sensitive to anti-asialo-GM1 and increased NK activity were still present 14 mo after DC injection. As NK cells lack memory, we found by depletion that CD4(+) not CD8(+) T cells were required for induction of the NK antitumor response. The role of DCs and CD4(+) T cells provides a novel mechanism for NK cell induction and innate immunity against cancer that may have potential in preventing clinical metastases.