The Cancer Immunotherapy Project aims to investigate evidenced-based cancer immunotherapy, repeating basic and translational research. This project is focused on developing not only more effective immunotherapies but also immunological methods for recurrence suppression or cancer prevention.
Glypican-3 (GPC3)We previously reported that GPC3, uniquely overexpressed in human hepatocellular carcinoma (HCC) and melanoma, is an ideal tumor antigen for immunotherapy in mouse models. We identified HLA-A24 (A*2402) and H-2Kd-restricted GPC3298-306 (EYILSLEEL), and HLA-A2 (A*0201) -restricted GPC3144-152 (FVGEFFTDV), both of which can induce GPC3-reactive cytotoxic T cells (CTLs). After approval of the Institutional Review Board of the National Cancer Center, we started the phase I clinical study of GPC3 peptide vaccine for patients with advanced HCC.
Preclinical study in a mouse model
BALB/c mice were intradermally vaccinated at the base of the tail with Kd-restricted GPC3298-306 peptide mixed with incomplete Freund's adjuvant (IFA), CpG, αGalCer or aluminum. However, only the peptide emulsified in IFA induced peptide-specific CD8+ T cells. Furthermore proteomic analyses showed that IFA protected the peptide against degradation in the human serum. Peptide-reactive CTLs were induced by the peptide vaccine in a dose-dependent manner. In addition, compared with the proportion of CTLs elicited by one to four vaccinations at 7-day intervals, the induction of GPC3298-306-specific CTLs required at least two vaccinations with a single dose > 10 μg. Repeated vaccinations with a lower dose of GPC3298-306 did not induce peptide-specific CTLs. Similarly, induction of an Ag-specific immune response by HLA-A2 GPC3<SMALL>144-152 depended on the dose administered.
The results of this study suggest that IFA is an indispensable adjuvant for peptide-based immunotherapy, and that the immunological effect of peptide vaccine depends on the dose of the peptide injected. We devised a schedule of the Phase I clinical study based on these results (133).
HSP105 is overexpressed in various cancers, but is expressed at low levels in many normal tissues, except for the testis. The expression of HSP105 in 62 human melanoma samples (46 primary and 16 metastatic lesions) and 42 melanocytic naevi samples was assessed by immunohistochemistry. Western blotting was performed on melanoma cell lines, melanoma tissues with matched normal skin, and melanocytic naevi. Seventy-four per cent of the primary melanoma lesions and 88% of the metastatic lesions overexpressed HSP105, as shown by immunohistochemistry. The majority of melanocytic lesions (95%) were negative (P < 0.05). Western blotting detected high HSP105 expression in melanoma cell lines and tissues. HSP105 expression was related to the invasiveness of the lesions. Melanocytic naevi expressed HSP105 at a level similar to that of normal skin. Our results show that high HSP105 expression is associated with malignant melanoma, particularly advanced and metastatic lesions. The results suggest that HSP105 analysis may be a helpful tool as a poor prognostic indicator and as a diagnostic aid in problematic lesions; in addition, melanoma can be included in the growing list of tumors overexpressing HSP105 and be targeted for potential HSP105-based therapeutic strategies (134).
Toward the development of a novel cancer immunotherapy, we have previously identified several tumor-associated antigens (TAAs) and the epitopes recognized by human histocompatibility leukocyte (HLA)-A2/A24-restricted cytotoxic T lymphocytes (CTLs). In this study, we attempted to identify a TAA of lung cancer (LC) and its HLA-A2-restricted CTL epitopes to provide a target antigen useful for LC immunotherapy. We identified a novel cancer testis antigen, cell division cycle associated gene 1 (CDCA1), overexpressed in nonsmall cell LC (NSCLC) using cDNA microarray analysis. The expression level of CDCA1 was increased in the majority of small cell LC, cholangiocellular cancer, urinary bladder cancer and renal cell cancers. We used HLA-A2.1 transgenic mice to identify the HLA-A2 (A*0201)-restricted CDCA1 epitopes recognized by mouse CTL, and we investigated whether these peptides could induce CDCA1-reactive CTLs from the peripheral blood mononuclear cells (PBMCs) of HLA-A2-positive donors and a NSCLC patient. Consequently, we found that the CDCA165-73 (YMMPVNSEV) peptide and CDCA1351-359 (KLATAQFKI) peptide could induce peptide-reactive CTLs in HLA-A2.1 transgenic mice. In HLA-A2(+) donors, in vitro stimulation of PBMC with these peptides could induce peptide-reactive CTLs which killed tumor cell lines endogenously expressing both HLA-A2 and CDCA1. This indicates that CDCA1 is a novel cancer-testis antigen overexpressed in LC, cholangiocellular cancer, urinary bladder cancer and renal cell cancers, and that CDCA1 may therefore be an ideal TAA useful for the diagnosis and immunotherapy of these cancers (135).
● T. Nakatsura ●
Table of Contents