Ryozo Eifuku

Successful Induction of Tumor‐specific Cytotoxic T Lymphocytes from Patients with Non‐small Cell Lung Cancer Using CD80‐transfected Autologous Tumor Cells

Cytotoxic T lymphocytes (CTL) against human lung cancer cells are difficult to induce by a conventional method using tumor cell stimulation probably due to an insufficiency of tumor antigens (TA) or costimulatory molecules such as CD80. We, therefore, investigated the potential of CD80‐transfected tumor cells as stimulators of the in vitro induction of autologous tumor‐specific CTL from regional lymph node lymphocytes in patients with lung cancer. Five non‐small cell lung cancer cell lines (two adenocarcinomas, 1 squamous cell carcinoma, 1 large cell carcinoma and 1 adenosquamous cell carcinoma) were established from surgical specimens and were successfully transduced with a plasmid constructed with expression vector pBj and human CD80 cDNA, using a lipofection method. CD80‐transfected tumor cells (CD80‐AT) significantly augmented the proliferation of autologous lymphocytes from all cases as compared with non‐transfected tumor cells (AT). AT‐stimulated lymphocytes from 4 out of 5 cases did not show any cytotoxicity against AT; however, lymphocytes stimulated with CD80‐AT exhibited substantial cytotoxicity against parental AT in all 5 cases tested. AT‐stimulated lymphocytes derived from only one out of 5 cases showed major histocompatibility complex (MHC)‐class I‐restricted cytokine production in response to AT, while the MHC‐class I‐restricted responses were found in CD80‐AT‐stimulated lymphocytes from 4 out of 5 cases. These results indicate that CD80 on tumor cells could be a beneficial costimulatory molecule to elicit CTL against lung cancer, and also show that TA recognized by CTL was frequently expressed on lung cancer cells.

Tumor-Infiltrating B-Cell-Derived IgG Recognizes Tumor Components in Human Lung Cancer

Abstract Tumor-infiltrating lymphocytes consist predominantly of T cells, whereas B cells, plasma cells, and natural killer cells are observed with different degrees of frequency. We investigated the nature of tumor-infiltrating B lymphocytes (TIB) in human lung cancer. First, to examine the ability of immunogloblin production by TIB, cancer tissues were subcutaneously transplanted in severe combined immuno-deficient mice, and the murine serum was examined for the concentration of human immunogloblin. Human IgG (hulgG) was detected in the serum of all 12 mice engrafted with lung cancer tissues. hulgM was almost undetectable. The levels of hulgG reached a peak approximately 6 weeks after engraftment and gradually decreased but were detectable until 20 weeks postengrafment. Serum from a large cell carcinoma-engrafted mouse reacted with a protein of 60 kDa derived from lung cancer cell lines (PC-9, Sq-1) and autologous tumor cells but did not react with cell lysates of normal lung tissue. Serum from an adenocarcinoma-engrafted mouse reacted with two proteins, 33 and 55 kDa, derived from lung cancer cell lines (PC-9, Sq-1, A549) and autologous tumor cells hut did not react with the lysate of normal lung tissues. These results suggest that B cells infiltrating lung cancer tissues produce IgG that recognizes common tumor-specific antigen.

Autologous Tumor‐specific Cytotoxic T Lymphocytes in a Patient with Lung Adenocarcinoma: Implications of the Shared Antigens Expressed in HLA‐A24 Lung Cancer Cells

Human lung adenocarcinoma‐specific cytotoxic T lymphocytes (CTL) were generated by multiple stimulations with autologous tumor cells (named A110L) from regional lymph node lymphocytes and tumor‐infiltrating lymphocytes expanded by solid‐phase anti‐CD3 monoclonal antibody (mAb) and recombinant interleukin‐2. The CTL lysed A110L but failed to kill either autologous B lymphocytes immortalized by the Epstein‐Barr virus or K562. The killing activity of the CTL against autologous A110L was inhibited by anti‐MHC class I mAb (W6/32), but not by anti‐MHC class II mAb. The CTL produced interferon‐γ and GM‐CSF in response to A110L and the production was completely blocked by the addition of anti‐MHC class I mAb. The HLA type of the CTL was HLA‐A2/A24, B52/B54, Cw1/‐. Allele‐specific deletion of HLA‐A2 molecules was observed in A110L by staining with anti‐HLA‐A2 mAb. A partial blocking effect on the cytokine production from the CTL was also obtained with anti‐CD8, and anti‐HLA‐A24 mAbs, but not with anti‐MHC class II, anti‐CD4 and anti‐HLA‐A2 mAbs. To analyze further the mechanism of antigen recognition by the CTL, the cross reactivity of the CTL against several HLA‐A locus‐matched (HLA‐A24+) and mismatched allogeneic tumor cells (HLA‐A24‐) was investigated. The A110L‐specific CTL showed a weak but significant cytotoxicity against some HLA‐A24 positive lung cancer cell lines, such as Sq‐1 (HLA‐A11/A24, squamous cell carcinoma) and PC‐9 (HLA‐A2/A24, adenocarcinoma), but failed to kill HLA‐A locus‐mismatched allogeneic tumors. This cross reactivity of the CTL against Sq‐1 and PC‐9 was blocked by anti‐MHC class I mAb. These results thus demonstrate that shared common tumor antigens might exist among lung cancer cells in the context of HLA‐A24.

Unfavorable prognosis of patients with non-small cell lung carcinoma associated with HLA-A2

BACKGROUND HLA class I molecules present antigenic peptides to cytotoxic T lymphocytes and, thus, play an important role in immune surveillance. Since 1970s there have been many reports of an increased frequency of one or more HLA haplotype in association with autoimmune disease, and malignancy. We studied types of HLA class I antigens in 204 resected non-small cell lung carcinoma (NSCLC) patients and also examined its correlation with clinicopathologic features and prognosis. METHOD Serological typing for HLA class I antigens was performed using a microcytotoxicity test. The disease-free survival curves were calculated by the Kaplan-Meier method and then compared using the Logrank test. Multivariate analysis was carried out by Coxs proportional hazard method. RESULTS The difference in disease-free survival time between the HLA-A2 present group and A2 absent group was significant (P = 0.040). The 3-year disease-free survival rate of all patients was 44% in HLA-A2 present group and 66% in A2 absent group. When a comparison was made within the group with stage I, expression of HLA-A2 was the only independent factor that affected survival time by multivariate analysis (P = 0.0457). CONCLUSIONS Expression of HLA-A2 was considered as one of the unfavorable prognostic factors in NSCLC patients. Our results suggested expression of HLA-A2 in NSCLC patients was one of the mechanisms of escape from immune surveillance.

Analysis of MAGE-3 derived synthetic peptide as a human lung cancer antigen recognized by cytotoxic T lymphocytes.

AbstractBackground. The human MAGE-3 gene was originally discovered in melanoma cells that encode tumor antigens, and has been reported to be expressed in various types of tumors, including lung cancer, but not in normal tissues other than testis or placenta. Our aim in this study was to clarify whether HLA-A2 restricted MAGE-3 peptide (FLWGPRALV) could be a lung cancer antigen recognized by cytotoxic T lymphocytes (CTL). Methods.MAGE-3-derived peptide-specific CTL were induced from the peripheral blood mononuclear cells (PBMC) of HLA-A0201-positive healthy donors and the regional lymph node lymphocytes (RLNL) of HLA-A2-positive patients with lung cancer by multiple stimulations with peptide-pulsed HLA-A0201-positive antigen-presenting cells. Results. Lymphocytes stimulated with MAGE-3 peptide exhibited specific lysis of Epstein-Barr virus-transformed B cells (EBV-B) pulsed with MAGE-3 peptide, but not with control peptide derived from influenza matrix protein, erbB-2, or wild type p53. Specific activity for MAGE-3-presenting targets was found after the second stimulation, and increased depending on the number of stimulations. The peptide-specific activity was inhibited by the addition of monoclonal antibodies against MHC class I and HLA-A2. Such CTL also recognized tumor cell lines expressing both HLA-A2 and MAGE-3 in an MHC class I-restricted manner, but did not recognize tumor cell lines that did not express HLA-A2 or MAGE-3. Conclusion. These results suggested the MAGE-3 peptide could be a potential target of specific immunotherapy for HLA-A2 patients with lung cancer.

Effects of Interleukin‐12 on the Induction of Cytotoxic T Lymphocytes from the Regional Lymph Node Lymphocytes of Patients with Lung Adenocarcinoma

Lung cancer‐specific cytotoxic T lymphocytes (CTL) were induced by repeated stimulations of regional lymph node lymphocytes (RLNL) in lung cancer patients with either autologous or HLA‐A‐locus‐matched tumor cells. To investigate the effect of interleukin‐12 (IL‐12), IL‐12 was added during the stimulation of RLNL from HLA A24 / adenocarcinoma patients with either autologous tumor cells or HLA A24‐positive adenocarcinoma cells (PC‐9) in combination with, or instead of interleukin‐2 (IL‐2), and then the cytotoxic activity, cytokine production and populations of the lymphocyte subsets were examined. The addition of IL‐12, or the substitution of IL‐2 by IL‐12 was found to enhance the cytotoxic activity and the cytokine production (IFN‐γ, GM‐CSF) of the CTL as compared with IL‐2 alone. The cytotoxic activity and cytokine production were both partially inhibited by anti‐MHC‐class I monoclonal antibody. The CTL thus induced by IL‐12 had a higher proportion of CD3+/CD56+ cells than the CTL induced with IL‐2 alone. The positively selected CD8+/CD56– lymphocytes showed PC‐9‐specific cytotoxic activity, because the population did not show any cytotoxicity to K562 or A549 (HLA‐A26/A30). However, the CD3+/CD56+ lymphocytes were cytotoxic to both PC‐9 and K562. In conclusion, IL‐12 is considered to be a useful cytokine for both the induction of lung‐cancer specific CTL and the augmentation of non‐MHC‐restricted cytotoxicity against tumor cells, and may be applicable for adoptive immunotherapy using CTL.

Induction of tumor-specific cytotoxic T lymphocytes from regional lymph node lymphocytes of human breast cancer

BackgroundIn this study we activated breast cancer-specific cytotoxic T lymphocytes (CTL) from regional lymph node lymphocytes (RLNL) of HLA-A2-positive patients with breast cancer.MelthodsFreshly isolated RLNL were stimulated with solid phase anti-CD3 monoclonal antibody followed by expansion with recombinant interleukin-2. Subsequently, the RLNL were stimulated with an irradiated HLA 0201 breast cancer cell line, MCF-7, at a responder/stimulator ratio of 10/1 once a week for 2 weeks.ResultsThe cultured RLNL exhibited specific lysis against MCF-7 in all 5 HLA-A2-positive patients tested, but not in 2 HLA-A2-negative patients. Cytotoxicity against MCF-7 was substantially inhibited by addition of anti-HLA-A2 mAb. In 3 of 5 HLA-A2-positive patients, anti-MCF-7 CTL also exhibited a substantial level of reactivity against PC-9, an HLA-A0206-positive lung adenocarcinoma cell line. Conversely, anti-PC-9-specific CTL were inducible by multiple stimulations of RLNL with PC-9 cells in 2 of 3 patients.ConclusionsThese results suggest that several common tumor antigens might exist among HLA-A2-positive breast cancers, some of which may be shared with lung adenocarcinomas.

Establishment of an immortalized T-cell line from lung cancer tissue: phenotypic and functional analyses

AbstractBackground. In order to investigate host defense against solid tumors, valuable information could be provided by ex-vivo analyses of functional immune cells in tumor tissues. However, available sources of fresh tumor-infiltrating T cells (TIL) are usually very limited, and it is often difficult to establish TIL lines. In this study, we analyzed the phenotypic and functional characteristics of TIL, using an immortalized cell line prepared by cotransfection of human c-myc and c-Ha-ras. Methods. A human T-cell line was established by cotransfecting c-Ha-ras and c-myc oncogenes to T lymphocytes freshly isolated from human lung large-cell carcinoma tissue. The phenotypes were assessed by flow cytometry. T-cell receptor (TCR) Vα- and β-usage was analyzed by polymerase chain reaction (PCR) and Southern blot. Cytotoxic activity against autologous and allogeneic tumor cell lines was examined by standard 51Cr-release cytotoxicity assays. Cytokine production by the established T-cell line, 904-T1, in response to stimulation by autologous tumor cells was assayed by using an enzyme-linked immunosorbent assay. Results. 904-T1 (CD3+, CD8+, CD56+, CD16-, CD161-, TCR Vα 9, 13, and Vβ 1, 5) displayed a broad range of MHC-nonrestricted tumoricidal activity against various human tumor cell lines, but did not lyse autologous B cells transformed by Epstein-Barr virus. The cytotoxicity of 904-T1 was not mediated by a T-cell antigen receptor or by Fas-ligand, but by perforin-based cytolytic pathways, and was enhanced by interleukin (IL)-12. 904-T1 cells produced large amounts of interferon (IFN)-γ, but not tumor necrosis factor (TNF)-α or IL-4 in response to autologous tumor cells, and produced high levels of IFN-γ and TNF-α, and a substantial level of IL-4 following stimulation with anti-CD3 monoclonal antibody. Conclusions. Our results suggest that 904-T1 cells were natural killer T (NKT)-like cells with regard to their nonspecific killing, cytokine repertoire, and sensitivity to IL-12, although the repertoire of the TCR variable region was not compatible with that of NKT cells.

Adoptive immunotherapy using specific cytotoxic T lymphocytes against human lung-cancer-engrafted severe combined immunodeficiency mice

OBJECTIVE We studied the ability of human lung-cancer-specific cytotoxic T lymphocytes to suppress the growth of human lung adenocarcinoma (PC-9) engrafted in severe combined immunodeficiency mice. METHODS PC-9-specific cytotoxic T lymphocytes were generated by multiple stimulation with irradiated PC-9 cells of regional lymph node lymphocytes from lung cancer patients expressing the same human leukocyte antigen-A locus haplotype as PC-9 following expansion due to the administration of immobilized anti cluster of differentiation 3 mAb and interleukin-2. Cytotoxic T lymphocytes showed specific cytotoxicity against PC-9 cells in vitro. Severe combined immunodeficiency mice with a subcutaneous graft of PC-9 were treated with a PC-9-specific cytotoxic T lymphocyte by i.v. injection and/or with interleukin-2 by s.c. injection. RESULTS Cytotoxic T lymphocyte treatment suppressed PC-9 graft growth significantly an effect, significantly enhanced when combined with interleukin-2 injection. To evaluate the in vivo specificity of anti-PC-9 cytotoxic T lymphocytes, each mouse was subcutaneously inoculated in the right flank with PC-9, and in the left flank with A549 or Sq-1. Cytotoxic T lymphocytes plus interleukin-2 treatment was found to suppress PC-9 growth selectively, but not A549 or Sq-1 growth. CONCLUSIONS These results provide sufficient rationale for conducting further clinical trials on immunotherapy using cytotoxic T lymphocyte for lung cancer patients.

Interleukin-12 enhances the antitumor activity of cytotoxic T lymphocytes against lung adenocarcinoma engrafted in severe combined immunodeficient mice

AbstractBackground. Through a number of biologic activities, interleukin 12 (IL-12) has proven to be a potential antitumor cytokine in mice bearing a variety of malignancies. However, in clinical trials in humans, the eradication of solid tumors remains difficult. Methods. A lung cancer cell line (PC-9)-specific cytotoxic T lymphocytes (CTL) were generated by multiple stimulations, with irradiated PC-9 cells, of regional lymph node lymphocytes obtained from patients with lung cancer whose cells expressed the same HLA-A locus haplotype as PC-9 (HLA-A24). Severe combined immunodeficient (SCID) mice bearing a subcutaneous graft of PC-9 were then intravenously injected with anti-PC-9-specific CTLs. Under these conditions, the in-vivo effect of recombinant human (rh) IL-2 and rh IL-12 was evaluated, based on tumor growth. Results. Mice that received either rh IL-2 or rh IL-12 exhibited no inhibitory effect on tumor growth. However, mice that received adoptive immunotherapy (AIT) alone exhibited a significant inhibition of tumor growth in the PC-9 graft in comparison to untreated mice. When mice were treated with AIT combined with rh IL-2 + rh IL-12 administration, tumor growth was significantly suppressed. A significant difference was observed in the growth of the PC-9 graft between AIT + IL-2 + IL-12 treatment and AIT + IL-2 treatment. Four of eight mice in the AIT + IL-2 + IL-12-treated group showed complete tumor regression. Conclusion. IL-12 showed a synergistic effect with adoptive immunotherapy, using CTL in a tumor-engrafted SCID model. These results are therefore considered to provide a sufficient rationale for IL-2 + IL-12-based immunotherapy using CTL transfer for patients with lung cancer.