Abstract Introduction: Lung cancer remains the leading cause of cancer death worldwide. Although prevention holds the promise to reduce lung cancer mortality, the results from randomized trials have been disappointing, likely due to our rudimentary knowledge of early lung cancer carcinogenesis. Cancer initiation and progression involve complex processes dependent on the dynamic interplay of individual cells within the tumor microenvironment (TME). Our understanding of the multicellular composition, interaction, spatial distribution, and functional dynamics during the early evolution from precancer to invasive LUAD is limited. Methods: We performed spatial single cell immune microenvironment analysis of 10 normal, 40 AAH, 22 AIS, 18 MIA, and 34 IAC by Imaging Mass Cytometry (IMC) and analyzed high-dimensional images of 1,618 regions-of-interest (ROIs) using a 34-plex marker. Using machine learning approaches, we classified 4,828,879 cells into 14 major cell types, which were subjected to further analysis for multicellular composition, interaction, spatial distribution, and functional dynamics in LUAD precursors of different stages. Results: From AAH to AIS, MIA, and IAC, we observed a gradual escalation in the density (number of cells/mm2) of all immune subsets. Proportionally, adaptive immune cells exhibited a progressive increase, whereas innate immune cells demonstrated a concurrent decrease with the neoplastic progression from normal lung to AAH, AIS, MIA, and ADC. Notably, macrophages emerged as the predominant immune cells within the immune microenvironment of these LUAD precursors. Upon closer examination of subtypes, protumor M2 macrophages showed a gradual increase from AAH to AIS, MIA, and IAC, whereas anti-tumor M1 macrophages increased from normal to AIH, AIS and subsequently decreased in AIS, MIA, and IAC stages. These alterations resulted in a progressive increase in M2/M1 ratios from AAH to IAC stages. Furthermore, M2 macrophages exhibited higher heterotypic interactions with epithelial cells than M1 across LUAD precursors of different stages, especially in the later stages. Co-expression analysis revealed 5 modules that distinguishes each histological stage. Tim-3 module was significantly enriched in AAH lesions, while NK cell-related features were enriched at AIS stage, B cell-related features enriched at MIA stage and Ki-67-related adaptive immune features enriched in IAC stage. Conclusion: The spatial single cell immune microenvironment analysis revealed a transition from innate to adaptive immunity during initiation and early progression of LUAD progression and innate immunity may play critical roles during early LUAD development. Therefore, reprograming innate immunity such as repolarization of M2/M1 may provide avenues to intercept LUAD precursors to prevent invasive LUAD. Citation Format: Bo Zhu, Pingjun Chen, Muhammad Aminu, Junya Fujimoto, Lingzhi Hong, Andre L. Moreira, Jian-Rong Li, Yanhua Tian, Luisa M. Solis Soto, Parra Cuentas Edwin Roger, Ou Shi, Hong Chen, Frank R. Rojas, Eduardo Vilar, Anirban Maitra, Ken Chen, Nicolas Navin, Cara L. Haymacker, Vamsidhar Velcheti, Daniel H. Sterman, P Andrew Futreal, Don L. Gibbons, Ignacio I. Wistuba, John V. Heymach, Chao Cheng, Harvey I. Pass, Jia Wu, Jianjun Zhang. Spatial single-cell immune microenvironment analysis reveals the transition from innate to adaptive immune response during early lung adenocarcinoma carcinogenesis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 6098.