Abstract.Immunological parameters were measured in order to elucidate a postoperative immunosuppression mechanism in transthoracic esophagectomy for patients with esophageal cancer. Moreover, lymphokine-activated killer (LAK) cells were transferred just after the surgery to overcome the postoperative immunosuppression. Fifteen consecutive patients who underwent transthoracic esophagectomy were subjected to the postoperative measurement of immunological parameters. Ten patients who underwent open cholecystectomy served as controls. Heparinized venous blood was obtained pre-and postoperatively, and serum levels of cytokines IL-6 and IL-10 and immunosuppressive acidic protein (IAP) were measured. Peripheral blood lymphocytes were harvested and analyzed by flow cytometry for phenotype detection and by a mixed lymphocyte reaction for detecting concanavalin (Con)-A-induced or -non-induced suppressor activity. Another 29 consecutive patients who underwent transthoracic esophagectomy were randomly enrolled in a postoperative immunotherapy trial either with or without lymphokineactivated killer cells. It was found that, in the esophagectomy group, IL-6 and IL-10 increased postoperatively and peaked on day 1, followed by an increase in IAP, peaked again on day 4, with a profound decrease in helper and cytotoxic T-cell subsets, followed by increases in Con-A-induced (on day 7 or later) and spontaneous (on day 10) suppressor activities. These changes were minimal in the cholecystectomy group. LAK cell transfer restored the postoperative decrease in the helper and cytotoxic T-cell population, and there was a trend of reduction for postoperative remote infection such as pneumonia and surgical site infection in the LAK therapy group. Taken together, we would like to propose the existence of a postoperative immunosuppression cascade consisting of increases in cytokines and immunosuppressive proteins, decreases in helper and cytotoxic T-cell populations, and the development of suppressor T-cell activities in surgery for esophageal cancer. Postoperative adoptive transfer of LAK cells may be a novel clinical application in surgery for esophageal cancer as a means of treating this postoperative immunosuppressive condition that may be identical to the status of compensatory anti-inflammatory response syndrome (CARS).

After the discovery of interleukin-2 (IL-2), lymphokine-activated killer (LAK) cells, tumor-infiltrating lymphocytes (TILs), and cytotoxic T lymphocytes (CTLs) sensitized with the mixed lymphocyte-tumor culture (MLTC) system have been conducted in adoptive immunotherapy (AIT) trials during past 15 years. Although the overall response rate of tumor shrinkage was marginal (9%), locoregional administration of TILs for malignant effusions was effective (77%) for a decrease or disappearance of the effusions even in terminally-ill patients, resulting in an improvement of QOL. Recent advances for molecular understanding of antigen presentation and recognition have promoted us to enhance the efficacy of AIT by using cultured dendritic cells (DCs) for generating antigen-specific CTLs in vitro. The peptide-pulsed DC-activated killer (PDAK) cells showed tumor recognition against antigen-expressing cells, and were efficiently propagated with the IL2 plus immobilized anti-CD3 antibody (IL-2/CD3) culture system. Clinical trials using PDAK cells against patients with lung metastases are now progressed, in which peptides suitable for generating CTLs were chosen in individual patients using the method designated as host-oriented peptide evaluation (HPOE) approach. Moreover, DCs were introduced with tumor-derived RNA, which was amplified with the T7 promoter system, and then were used for stimulating lymphocytes. The tumor RNA-introduced DC-activated killer (TRiDAK) cells showed tumor-specific interferon-gamma spots even in a patient in whom we failed to generate PDAK cells using DCs and peptides, suggesting that the clinical trial of AIT using TRiDAK cells is warranted for the treatment of patients with metastatic cancer. Thus, more understanding of antigen-presentation and -recognition mechanisms and immune regulation systems may promote clinical applications of AIT to establish a novel modality of cancer treatment.

In total, 16 patients with cytologically proven malignant effusion from colorectal cancer were treated by locoregional administration of the streptococcal preparation OK-432 alone or OK-432 plus the T-cell growth factor interleukin (IL)-2, and the action mechanism of the treatment was studied. A positive clinical response, showing a cytologic disappearance of cancer cells and decrease of effusion, was observed in nine of 11 (82%) patients treated with OK-432 alone and in all five patients treated with OK-432 plus IL-2. Flow cytometric analysis revealed that OK-432 plus IL-2 locally induced acute inflammation-like responses, including serial cellular infiltrations of granulocyte migration within a matter of hours, and activation of macrophages and T lymphocyte involvement within the following days, and that a predominant expansion of CD3 þ CD4 þ lymphocytes (CD: cluster of differentiation) was induced by in vitro stimulation with IL-2 of locoregional cells after the OK-432 administration (OK/IL-2AK cells). The OK/IL-2AK cells produced tumour necrosis factor-a and interferon-g, but these cells did not produce IL-4 and IL-6. The OK/IL-2AK cells expressed potent killing activity against autologous tumour cells. This activity was abrogated by treatment of the lymphocytes with anti-CD3, -CD4, -TCRab antibody, and by the treatment of target cells with anti-human leukocyte antigen (HLA)-DR antibody. The OK/IL-2AK cells expressed Fas-L gene, and flow cytometric analysis demonstrated HLA-DR expression in approximately 75% of CEA þ or cytokeratin þ effusion cells. TCRVb gene analysis of the OK/IL-2AK cells showed an oligoclonal usage of TCRb20, which was also involved in the cytotoxic mechanism of the OK/IL-2AK cells. Single-strand conformational polymorphism analysis demonstrated the clonotypes for the TCRVb20 gene, and the CDR3s of the gene were sequenced. The clonotypic PCR using the TCRVb20-CDR3 sequences could detect the CDR3-identical TCRs in effusion lymphocytes from the other patients. Taken together, it is suggested that locoregional administration of OK-432 plus IL-2 is highly effective for the management of malignant effusion from colorectal cancer. OK-432 plus IL-2 induces autologous tumour-reactive CD4 þ Th1 killer lymphocytes, which recognise tumour antigen(s) presented with HLA class II molecules on effusion tumour cells by means of preferential usage of TCRVb20. The clonotypic PCR using the TCRVb20-CDR3 sequences may be informative for treating malignant effusion from colorectal cancer using OK-432 plus IL-2.

Anti-tumor effector cells were generated by stimulating peripheral blood lymphocytes with cultured dendritic cells (DCs) and mRNA extracted from the gastric cancer cell line MKN45 or ascites tumor cells of gastric cancer patients. DCs were generated from an adherent fraction of peripheral blood mononuclear cells (PBMCs) in the presence of GM-CSF and IL-4. mRNA was extracted from tumor cells and subjected to T7-amplification. The DCs were electroporated (150 V/150 μF) with amplified mRNA and used after maturation with TNF-• to stimulate PBMCs to generate tumor RNA-introduced DC-activated killer (TRiDAK) cells. It was found that tumor RNA could efficiently be introduced into cultured DCs by electroporation (55% efficiency, 78% viability), and tumor RNA-introduced DCs could reproducibly stimulate lymphocytes to be tumor-reactive TRiDAK cells. The TRiDAK cells expressed an IFN-Á response specific for tumor cells, but not for normal cells. Mock DCs or normal cell RNA-introduced DCs did not induce any killer cells. RNAspecific recognition of the effector cells generated was demonstrated using an amplified EGFP-mRNA system. The tumor killing activity of TRiDAK cells was inhibited not only with the anti-HLA class I antibody but also with the anti-HLA class II antibody as well as the anti-TCR antibody. TRiDAK cells reactive with autologous tumor cells could be generated in a CEA-positive gastric cancer patient with malignant ascites, in whom effector cell generation using DCs and CEA peptides had failed. These results suggest that TRiDAK cell generation is safe, feasible, and active in gastric cancer patients with malignant ascites, and is superior to other effector cell generation systems using DCs and epitope peptides. The adoptive immunotherapy of cancer using TRiDAK cells may be warranted in a clinical setting. This is the first study investigating anti-tumor effector cell generation using cultured DCs and tumor mRNA from gastric cancer cells.

No abstract