therapeutic outcome of cancer immunotherapy. Keywords: Nivolumab, non-small cell lung cancer, Anti-angiogenesis P2.07-020 Distinct Immune Status in Patients with Adenocarcinoma and Squamous Cell Carcinoma: Implication for Immunotherapy of NSCLC N. Hradilova, 1 O. Palata, 2 L. Sadilova, 2 D. My síková, 3 H. Mrazkova, 3 R. Lischke, 3 R.  Spí sek, 2 I. Adkins 2 1 Department of Immunology, Second Faculty of Medicine, Charles University, Prague/CZ, 2 Sotio A.S., Prague/ CZ, 3 Thoracic and Lung Transplantation Division, First Faculty of Medicine, Charles University in Prague and University Hospital Motol, Prague/CZ Background: Lung cancer is the leading cause of cancer mortality worldwide therefore understanding the role of immune system in anti- tumor immunity is of a great interest. Here we compared immune cell infiltration and responses in tumors and non-tumoral lung tissue from 43 adenocarcinomas (AC) and 39 squamous cell carcinomas (SCC) of non- small cell lung cancer patients. Method: In this study we compared immune cell populations, T cell responses and secreted cytokines in primary tumors and non-tumoral lung tissue as well as in blood of non- small cell lung cancer (NSCLC) patients undergoing neoadjuvant sur- gery. Moreover, we compared immune suppressive populations such as CD4 + CD25 + Foxp3 + T regulatory cells and myeloid-derived suppressor cells (MDSC). Result: Whereas T, B and NK cells infiltration was com- parable in AC and SCC, the number of dendritic cells was lower in SCC tumors. CD8 + T cell and NK cell proliferation, IFN-g-production from T cells and secretion of proinflammatory cytokines after stimulation in vitro was lower in SCC compared to AC tumors. A higher number of Tregs was detected in tumors and blood of AC patients, whereas a higher number of MDSC was found in SCC patients. The suppressive function of Tregs was comparable between AC and SCC patients, but MDSC in SCC patients displayed a higher suppressive function as shown by inhibition of CD3z expression and IL-2 and IFN-g production in T cells, Lox-1 plasma concentrations compared to AC patients and age- matched controls. Conclusion: Our results suggest that immune system of SCC patients might be subjected to higher immunosuppression than AC patients. Our observations also give rationale to target specifically MDSC in SCC patients and Tregs in AC patients for designing combi- natorial immunotherapeutic approaches. Keywords: Non-small cell lung cancer, tumor infiltrate, immune suppression, immunotherapy P2.07-021 A Checkpoint Molecule B7-H3 as a Novel Immune Therapy Target for Non-Small Cell Lung Cancer (NSCLC) K. Yonesaka, 1 K. Kudoh, 1 S. Takamura, 2 H. Sakai, 1 R. Kato, 1 K. Haratani, 1 T. Takahama, 1 K. Tanaka, 1 H. Hayashi, 1 H. Kaneda, 1 M. Takeda, 1 O. Maenishi, 3 M. Yamato, 4 M. Miyazawa, 2 K. Nishio, 5 K. Nakagawa 1 1 Department of Medical Oncology, Kindai University Faculty of Medicine, Osaka-Sayama/JP, 2 Department of Immunology, Kindai University Faculty of Medicine, Osaka-Sayama/JP, 3 Department of Pathology, Kindai University Faculty of Medicine, Osaka-Sayama/JP, 4 Daiichi Sankyo Co., Ltd., Tokyo/JP, 5 Department of Genome Biology, Kindai University Faculty of Medicine, Osaka- Sayama/JP Background: Anti-PD-1 immune therapy improved survival in NSCLC, whereas some patients were not responding to this treatment, indicating the requirement of alternative strategies for these patients. B7-H3, an immune checkpoint molecule, has known to be expressed in some cancer cells including NSCLC. In this study, we examined the therapeutic potential of targeting B7-H3 using a mouse model, also elucidated the expression levels of B7-H3 on NSCLC tumor cells. Method: Pan02 murine cancer cells were inoculated in syngeneic mice, and anti-tumor efficacy of anti-B7-H3, anti-PD-L1 antibodies were evaluated. T-cell expression of IFN gamma was evaluated in the spleen and tumor infiltrating lymphocytes using flow-cytometer. B7-H3 expression on tumor cells in patients with NSCLC (n¼69) was evaluated by immunohistochemistry. Result: In the mouse model study, the treatment with anti-B7-H3 antibody significantly pre- vented the tumor-growth as compared to isotype antibody. The numbers of CD4+ and CD8+ T-cells infiltrated in the tumor significantly increased following treatment with anti-B7-H3 antibody. Importantly, depletion of CD8+ T-cells cancelled the anti-tumor effect of anti-B7-H3 antibody treatment, indicating that the blockade of B7-H3 potentiates anti-tumor CD8+ T-cell responses. In fact, CD8+ T-cell expressions of IFN gamma in response to tumor cells were improved when mice were treated with anti-B7-H3 antibody. Furthermore, combination with anti-B7-H3 and anti-PD-L1 antibody treatment showed synergic effect in inhibiting tu- mor-growth. The expressions of B7-H3 were evident on NSCLC tumors, which consists 62% of NSCLC patients. Conclusion: Anti-B7-H3 antibody exhibited CD8+ T-cell-mediated anti-tumor effects in the mouse model study. B7-H3 was aberrantly expressed in NSCLC tumor cells. Anti-B7-H3 antibody or its combination with anti-PD-1 antibody is suggested to be effective for patients with NSCLC. Keywords: PD-1, PD-L1, B7-H3 P2.07-022 Inflammatory Cytokine Induction after Anti-PD-1 Ab Administration Relates to the Efficacy and Safety in Patients with Non-Small Cell Lung Cancer Y. Ozawa, 1 Y. Amano, 1 T. Koyauchi, 1 T. Kakutani, 1 Y. Sato, 1 H. Hasegawa, 1 T. Matsui, 1 M. Tanahashi, 2 H. Niwa, 2 K. Yokomura, 1 T. Suda 3 1 Department of Respiratory Medicine, Respiratory Disease Center, Seirei Mikatahara General Hospital, Hamamatsu/JP, 2 Division of Thoracic Surgery, Respiratory Disease Center, Seirei Mikatahara General Hospital, Hamamatsu/JP, 3 2Nd Division, Department of Internal Medicine, Hamamatsu University School of Medicine, Hamamatsu/JP Background: PD-1/PD-L1 interaction affects various immune cells, including macrophage and dendritic cells, which play crucial roles in Table. Influence of prior anti-angiogenesis treatment Prior anti-angiogenesis N DCR% median PFS (weeks) median OS (weeks) Yes 16 30.0 8.29 13.1 No 118 60.7 11.3 27.5 P-value 0.0912 0.0060 0.1697 Statistical method Fisher’ s exact Log-rank Log-rank November 2017 Abstracts S2423