Satoshi Morita, Natsuko Onishi, Hiroshi Ishiguro, Michio Yoshimura, Takahiko Koyama, Hiroshi Yoshibayashi, Maya Honda, Hirofumi Suwa, Atsushi Yonezawa, Masahiro Takada, Kosuke Kawaguchi, Takeshi Kotake, Tatsuki R. Kataoka, Wakako Tsuji, Yuki Himoto, Rikiya Yamashita, Laxmi Parida, Masako Kataoka, Hironori Kato, Hiroyasu Yamashiro, Reitaro Tokumasu, Masakazu Toi, and Ryuji Uozumi
BACKGROUND: Advances in imuuno-oncology (IO), i.e., a programmed cell death protein 1 (PD-1) inhibitors are revolutionizing cancer treatment by offering hope for a cure. However, the benefit of IO agents in metastatic breast cancer (MBC) is still limited. For immunotherapy to be successful, it is essential to understand the tumor immune contexture. Tumor immune microenvironment (TIME) alterations, such as increased effector-cell composition, inflamed immune cells, and the activated effector immune cells create conditions that enhance immunotherapy in MBC. The abscopal effect induced by Radiation therapy (RT) is known as a modulator of TIME. Then, abscopal effect is considered to be a systemic anti-tumor immune response. In this study, we integrated time series multi-omics data to identify alteration of the TIME signatures with RT combined with N\nivolumab, a PD-1 inhibitor, using a machine learning method. PATIENTS AND METHODS: This study was conducted as a translational research of the KBCRN-B-002 trial, which is a multicenter phase Ib/II study for evaluating the safety and efficacy of nivolumab in combination with RT in patients with HER2-negative MBC (Takada M et al. will make presentation the main results of the study. (UMIN: UMIN000026046; ClinicalTrials.gov: NCT03430479)). Twenty-nine patients were included in the translational analysis set. The multi-omics data included data from RNAseq, DNAseq, mass cytometry (CyTOF), multiple cytokines, human leukocyte antigen, radiomics, drug concentrations, tuberculin reaction using peripheral blood mononuclear cells (PBMCs), serum, plasma, imaging, and Formalin-Fixed Paraffin-Embedded (FFPE) samples. We collected time series data, which included data for the baseline, two weeks after starting therapy, four weeks after starting therapy, and the timing of progressive disease. For the integrated analysis, a unique machine learning method was developed. RESULTS: Our integrated analysis identified the alteration of the tumor immune microenvironment signatures using nivolumab combined with RT for MBC. These results included the signature of the immune cell composition, effector immune cells, and immune suppressor cells, via single-cell analysis using CyTOF. The types of analysis are given in the table. CONCLUSIONS: We identified candidates for TIME signatures for actionable hallmarks of RT combined with IO agent by our machine learning method. Type of AnalysisSampleTime PointBaseline2 Weeks4 WeeksProgressive DiseaseRNAseqFFPE✓RNAseqPBMCs✓✓✓✓DNAseqFFPE✓CyTOFPBMCs✓✓✓✓CytokinesSerum✓✓✓✓Nivolumab ConcentrationSerum✓✓✓✓TILsFFPE✓HLAPBMCs✓Tuberculin ReactionSkin✓ Citation Format: Kosuke Kawaguchi, Masahiro Takada, Takeshi Kotake, Michio Yoshimura, Ryuji Uozumi, Masako Kataoka, Takahiko Koyama, Reitaro Tokumasu, Maya Honda, Rikiya Yamashita, Atsushi Yonezawa, Yuki Himoto, Natsuko Onishi, Hironori Kato, Hiroshi Yoshibayashi, Hirofumi Suwa, Wakako Tsuji, Hiroyasu Yamashiro, Tatsuki Kataoka, Hiroshi Ishiguro, Laxmi Parida, Satoshi Morita, Masakazu Toi. Alteration of the tumor immune microenvironment signatures by nivolumab combined with radiation therapy for patients with metastatic breast cancer (Translational Research of the KBCRN-B-002 trial) [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P4-10-33.