Abstract While immune checkpoint inhibitors (ICI) have revolutionized treatment for various cancers including non-small cell lung cancer (NSCLC), only a subset of patients receives durable clinical benefit. Discovery of blood-based biomarkers that reflect dynamic change of tumor microenvironment, and predict response to ICI will markedly improve current treatment regimens. Emerging evidence has shown that the frequency of peripheral blood (PB) CD8+ T cells expressing CX3CR1, a marker of effector T-cell differentiation increases in patients receiving ICI. However, no studies have rigorously evaluated the utility of CX3CR1 as a PB biomarker to predict response to ICI, and identify responders and non-responders. Here, we evaluated the utility of circulating T-cell biomarkers to predict response to ICI therapy in preclinical models and NSCLC patients (n=35). We found successful treatment with ICI significantly increases the frequency of PB CX3CR1+ CD8+ T-cell subsets that enrich neoantigen- as well as shared tumor-associated antigen-specific T cells in mice bearing MC38 and CT26 colon adenocarcinoma. Significantly increased expression of IFN-γ and Ki67 suggests that these subsets are highly proliferative effector CD8+ T cells. Interestingly, upregulation of Ki67 was transient; however, expression of CX3CR1 remained high on tumor-specific tetramer+ CD8+ T cells. Moreover, high-throughput sequencing of the T-cell receptor (TCR) identified significantly higher clonality and overlapping TCR repertoires in between peripheral CX3CR1+ CD8+ T cells and CD8+ tumor-infiltrating lymphocytes (TILs), suggesting the potential utility as a dynamic biomarker early on-treatment. Most importantly, changes in the frequency of PB CX3CR1+ CD8+ T cells associate with response to ICI therapy in individual mice. Phenotypic analysis of PB CD8+ T cells from 35 NSCLC patients undergoing PD-1/PD-L1 blockade revealed that the maximum percent change of these T-cell subsets from baseline identified responders and non-responders as early as 3 weeks from the initiation of ICI therapy with higher sensitivity, specificity, positive and negative predictive value compared to PD-L1 expression in the tumor, tumor mutational burden, and pre-existing TILs. Importantly, TCR sequencing confirmed clonally-expanding TCR repertoires within the T-cell subsets that were also found in TILs. Collectively, the frequency of circulating CX3CR1+ CD8+ T cells correlates with response to ICI therapy, and identifies responders vs non-responders early on-treatment. These findings provide a solid foundation for future development of clinical trials with large cohorts of patients undergoing ICI therapy not only for NSCLC, but also for a wide variety of malignancies. Citation Format: Takayoshi Yamauchi, Toshifumi Hoki, Takaaki Oba, Kristopher Attwood, Sebastiano Battaglia, Igor Puzanov, Hongbin Chen, Grace Dy, Brahm Segal, Marc Ernstoff, Fumito Ito. Frequency of circulating CX3CR1+ CD8+ T cells to predict response to immune checkpoint inhibitor therapy [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 1044.