MUC1-specific cytotoxic T lymphocytes eradicate tumors when adoptively transferred in vivo.

We have reported previously that MUC1 transgenic mice with spontaneous tumors of the pancreas (designated MET) naturally develop MHC class I-restricted, MUC1-specific CTLs as tumors progress (P. Mukherjee et al., J. Immunol., 165: 3451-3460, 2000). From these MET mice, we have isolated, expanded, and cloned naturally occurring MUC1-specific CTLs in vitro. In this report, we show that the CTL line is predominantly CD8+ T cells and expresses T-cell receptor Vbeta chains 5.1/5.2, 11, 13, and 2 and Valpha chains 2, 8.3, 3.2, and 11.1/11.2. These CTLs recognize several epitopes on the MUC1 tandem repeat with highest affinity to APGSTAPPA. The CTL clone, on the other hand, is 100% CD8+ cells and expresses a single Vbeta chain of 5.1/5.2 and Valpha2. It recognizes only the H-2Db class I-restricted epitope of MUC1, APGSTAPPA. When adoptively transferred, the CTLs were effective in eradicating MUC1-expressing injected tumor cells including mammary gland cells (C57mg) and B16 melanomas. These results suggest that MUC1-specific CTLs are capable of possibly preventing, or at least substantially delaying, MUC1-expressing tumor formation. To our knowledge, this is the first evidence that demonstrates that the naturally occurring MUC1-specific CTLs isolated from one tumor model has antitumor effects on other MUC1-expressing tumors in vivo. Therefore, our data confirm that MUC1 is an important tumor rejection antigen and can serve as a target for immunotherapy.

MUC1-specific CTLs are non-functional within a pancreatic tumor microenvironment.

Pancreatic cancer is a highly aggressive, treatment refractory disease and is the fourth leading cause of death in the United States. In humans, 90% of pancreatic adenocarcinomas over-express altered forms of a tumor-associated antigen, MUC1 (an epithelial mucin glycoprotein), which is a target for immunotherapy. Using a clinically relevant mouse model of pancreas cancer that demonstrates peripheral and central tolerance to human MUC1 and develops spontaneous tumors of the pancreas, we have previously reported the presence of functionally active, low affinity, MUC1-specific precursor cytotoxic T cells (pCTLs). Hypothesis for this study is that MUC1-based immunization may enhance the low level MUC1-specific immunity that may lead to an effective anti-tumor response. Data demonstrate that MUC1 peptide-based immunization elicits mature MUC1-specific CTLs in the peripheral lymphoid organs. The mature CTLs secrete IFN-gamma and are cytolytic against MUC1-expressing tumor cells in vitro. However, active CTLs that infiltrate the pancreas tumor microenvironment become cytolytically anergic and are tolerized to MUC1 antigen, allowing the tumor to grow. We demonstrate that the CTL tolerance could be reversed at least in vitro with the use of anti-CD40 co-stimulation. The pancreas tumor cells secrete immunosuppressive cytokines, including IL-10 and TGF-beta that are partly responsible for the down-regulation of CTL activity. In addition, they down-regulate their MHC class I molecules to avoid immune recognition. CD4+ CD25+ T regulatory cells, which secrete IL-10, were also found in the tumor environment. Together these data indicate the use of several immune evasion mechanisms by tumor cells to evade CTL killing. Thus altering the tumor microenvironment to make it more conducive to CTL killing may be key in developing a successful anti-cancer immunotherapy.

Mice with spontaneous pancreatic cancer naturally develop MUC-1-specific CTLs that eradicate tumors when adoptively transferred.

Pancreatic cancer is a highly aggressive, treatment refractory cancer and is the fourth leading cause of death in the United States. In humans, 90% of pancreatic adenocarcinomas overexpress altered forms of a tumor-specific Ag, mucin 1 (MUC1; an epithelial mucin glycoprotein), which is a potential target for immunotherapy. We have established a clinically relevant animal model for pancreatic cancer by developing a double transgenic mouse model (called MET) that expresses human MUC1 as self molecule and develops spontaneous tumors of the pancreas. These mice exhibit acinar cell dysplasia at birth, which progresses to microadenomas and acinar cell carcinomas. The tumors express large amounts of underglycosylated MUC1 similar to humans. Tumor-bearing MET mice develop low affinity MUC1-specific CTLs that have no effect on the spontaneously occurring pancreatic tumors in vivo. However, adoptive transfer of these CTLs was able to completely eradicate MUC1-expressing injectable tumors in MUC1 transgenic mice, and these mice developed long-term immunity. These CTLs were MHC class I restricted and recognized peptide epitopes in the immunodominant tandem repeat region of MUC1. The MET mice appropriately mimic the human condition and are an excellent model with which to elucidate the native immune responses that develop during tumor progression and to develop effective antitumor vaccine strategies.