Background: Immune checkpoint inhibitors (ICIs) have become a standard treatment for non-small cell lung cancer (NSCLC), yet there remains a lack of effective biomarkers to explain the heterogeneity in therapeutic response and prognosis. As ICIs directly interact with circulating immune cells, transcriptomic changes in these cells may critically influence antitumor immunity and patient outcomes. However, the extent to which these transcriptomic features are shaped by germline and somatic variants, such as clonal hematopoiesis of indeterminate potential (CHIP), remains unclear in NSCLC. Methods: In this study, I performed targeted sequencing on blood samples from 100 NSCLC patients for CHIP detection and conducted single-cell RNA sequencing (scRNA-seq) on 73 of these patients. Using these matched samples, I analyzed transcriptomic differences driven by germline or somatic variants, integrating single-cell expression quantitative trait loci (sc-eQTL) mapping and co-expression network analysis. Additionally, I examined transcriptional networks in myeloid cells to understand the effect of CHIP and across immune clusters for germline variants. Finally, I conducted survival analyses to assess clinical relevance. Results: I identified distinct, CHIP-driven myeloid-specific inflammatory signatures, notably involving NF-kB activation, in NSCLC patients with CHIP mutations. Concurrently, sc-eQTL analysis identified 3,616 eGenes, many of which showed immune-specific regulatory unique to this cohort. I also derived a functional core gene score based on co-expression network involving CD8⁺ T cell differentiation, which was significantly associated with progression-free and overall survival. Notably, while germline and CHIP-related variants regulated immune states, the CD8⁺ T cell network explaining ICI response was predominantly shaped by strong co-expression regulation rather than individual genetic variants. Conclusion: This study highlights the contribution of germline and CHIP-driven somatic mutations to immune transcriptional features in NSCLC patients treated with ICIs. However, response-explaining regulatory modules, such as those governing CD8⁺ T cell differentiation, appear to be driven more by co-expression dynamics than by direct genetic control. Our findings provide valuable resources for understanding immune regulation by genetic variants in lung cancer patients and showed the need for further validation of the identified biomarkers in independent cohorts.