High dimensional analysis of untreated early lung cancer lesions reveals novel myeloid immune responses to tumor
Yonit Lavin1, Soma Kobayashi1, Andrew Leader1, El-Ad David Amir1,2, Camille Bigenwald1, Robert Sweeney1, Romain Remark1, Christian Becker1, Naama Elephant3, Jacob Levine2, Klaus Meinhof1, Andrew Chow1, Seunghee Kim-Shulze2, Andrea Wolf1, Emanuella Taioli1, Raja Flores1, Sacha Gnjatic1, Adeeb Rahman1, Dana Pe’er2, Ido Amit3, Miriam Merad1
1Icahn School of Medicine at Mount Sinai, Oncological Sciences, New York, NY 10029, USA,
2 Departments of Biological Sciences, Systems Biology and Computer Science, Columbia University, New York, NY, USA
3 Weizmann Institute of Science, Department of Immunology, Rehovot, Israel
Recently, monoclonal immunotherapy agents, particularly anti-PD-1 has demonstrated unprecedented clinical responses in patients with late stage and metastatic non-small cell lung carcinoma. These therapies target T cells, as they are the major effector cells in the tumor. These cells work in concert with innate cells which provide a first line of protection and are critical for antigen presentation. However, while many studies have focused on tumor infiltrating lymphocytes, little is known about the diversity of the myeloid repertoire at the tumor site. Work in our laboratory and others, have suggested that in various mouse tumor models, targeting specific cells of the myeloid compartment may increase response to immunotherapy and tumor regression. Therefore, understanding the composition of the myeloid compartment may be critical for developing new immunotherapy targets. Moreover, little is known about the nature of the immune response to early stage lung tumors.
To address this unmet need, we developed a multiscale immune profiling strategy to map the immune micro-environment of patient early lung adenocarcinoma lesions. We used high-dimensional analysis including mass-cytometry and multiplex-immunohistochemistry to map the immune microenvironment of 28 untreated lung adenocarcinoma. We demonstrate that early stage lung tumors have a robust immune response Utilizing a unique barcoding strategy that allows the simultaneous analyses of the tumor lesion, adjacent non-involved lung tissue and blood cells, we identified a response of both T cells and innate cells that was specific to the tumor. We propose that the tumor immune mapping helps stratify patients with early lung cancer lesions based on distinct immune patterns that will help tailor immunotherapy strategies and improve tumor responses.