Abstract Background: Our knowledge in early lung carcinogenesis is rudimentary. Atypical adenomatous hyperplasia (AAH) is the only preneoplasia type that has been recognized by histopathology and It has been postulated that AAH can progress to adenocarcinoma in situ (AIS), to minimally invasive adenocarcinoma (MIA), and finally to invasive adenocarcinoma (ADC). However, the definition of these lesions remains controversial due to limited supporting molecular evidence. Carcinogenesis results from accumulation of multiple genetic and epigenetic events, with selection for events conferring phenotypic advantage. Immune surveillance, particularly anti-tumor response from T cells is an important host protection process to inhibit carcinogenesis. However, the T cell repertoire landscape and its evolution during lung carcinogenesis have not been well defined. Methods: We performed T cell receptor (TCR) sequencing on multiple spatially separated regions (2-6 regions per lesion) from 23 AAH, 26 AIS, 54 MIA and 14 ADC lesions and paired histologically normal lung tissues from 52 patients clinically presenting with indeterminate pulmonary nodules. Forty-one patients had multifocal diseases and 23 patients carried more than one type of pathology. Results: Compared to pre-/micro-invasive neoplastic lesions (AAH, AIS, MIA) or invasive ADC, normal lung tissues demonstrated significantly less T cell infiltration (p<0.0001) but a significantly higher clonality (p<0.0001). Furthermore, T cell diversity increased and as did evenness in pre-/micro-invasive neoplastic lesions (AAH, AIS, MIA) (p<0.0001) and invasive ADC (p<0.005) accompanied with higher frequency of the top T cell clones observed in the normal lung (p<0.05). Interestingly, homology with normal lung T cell repertoire decreased in invasive ADC compared to pre-/micro-invasive neoplastic lesions (p<0.05). Distinct T cell repertoire overlap was also observed between pre-/micro-invasive neoplastic lesions and invasive ADC (p<0.01). Conclusions: Our preliminary analyses demonstrate the distinct T cell repertoire between the normal lung and pre-/micro-invasive neoplastic lesions and invasive ADC, and that immunosuppression of T cells may have occurred prior to the development of pre-neoplastic states and T cell repertoire becomes progressively suppressed with disease evolution. The distinct overlap of pre-/micro-invasive neoplastic lesions and invasive ADC suggests that with disease evolution, the immune microenvironment of invasive ADC increases in complexity. Integration of the molecular landscape with T cell and immune profiling across different stage lesions is underway. Citation Format: Runzhe Chen, Junya Fujimoto, Alexandre Reuben, Lisha Ying, Xin Hu, Chi-Wan Chow, Jaime Rodriguez Canales, Wenyong Sun, Jinlin Hu, Edwin R. Parra Cuentas, Carmen Behrens, Chang-Jiun Wu, Latasha Little, Curtis Gumbs, Diana Wiesnoski, Guangchun Han, Won-Chul Lee, Paul Scheet, Humam Kadara, Mara Antonoff, Ara A. Vaporciyan, Stephen Swisher, Jianhua Zhang, John Heymach, Waun Ki Hong, Ignacio Wistuba, Andrew Futreal, Dan Su, Jianjun Zhang. T cell repertoire evolution from the normal lung to invasive lung adenocarcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 4686.