Abstract 7331: Establishing a single-cell eQTL dataset of lung tissues from Asian never-smokers to identify cell-type specific lung cancer susceptibility genes

Abstract Lung cancer is one of the most prevalent and deadly forms of cancer worldwide. Whilst smoking is the main determinant, genetic factors also play a crucial role as genome-wide association studies (GWAS) have identified 50 loci associated with lung cancer risk. Yet, for most of these loci, it is still unknown how they contribute to lung cancer risk. Expression quantitative trait loci (eQTL) studies have been powerful in linking GWAS variants to potential target genes, providing genetic mechanisms underlying common diseases such as cancers. However, the current eQTL resources lack ancestral diversity and are primarily based on bulk tissues. Emerging single-cell eQTL (sc-eQTL) approaches can detect context-specific gene regulation but are mainly of blood samples or cultured cells and still representing European populations. This limits our abilities to test GWAS variants in cancer-relevant cell types as well as in diverse populations. To generate a resource to characterize lung cancer GWAS loci, we are building a lung sc-eQTL dataset of Asian population while addressing common challenges of tissue sc-eQTL. Namely, processing fresh tissue in a population scale is logistically challenging and costly, and epithelial cells (including cell types of lung cancer origin) are vulnerable to the dissociation and freezing/thawing process. To address these issues, we incorporated sample multiplexing and cell type balancing. We collected fresh tumor-distant normal lung tissues from 131 never-smoking Korean women and dissociated them before cryopreservation. We then performed single-cell RNA sequencing (scRNA-seq) using 10x Chromium Single Cell 3’ v3. 1 chemistry with multiplexing of ~6 samples/batch. To enrich for epithelial cells, we utilized flow cytometry with surface markers of four major lung cell types (epithelial: EpCAM+/CD45-, immune: EpCAM-/CD45+, endothelial and stromal: EpCAM-/CD45-) before 10X library preparation. Concurrently, we performed DNA genotyping and imputation using matched blood samples. Following scRNA-seq (~36, 000 reads/cell) we performed a genotype-based sample demultiplexing using Demuxlet. By integrating Demuxlet and Scrublet, we identified ~89% of the detected cells as singlets. After applying QC to filter empty droplets and low-quality cells, we obtained 428, 619 cells or 3, 272 cells/patient. Cell annotation guided by Azimuth label transfer using the Human Lung Cell Atlas, identified 28 cell types including 7 of epithelial origins. We will further perform eQTL analyses for individual cell types using pseudo-bulk and LIMIX (linear mixed model) or SAIGE-qtl (poisson regression) methods followed by aggregation across the cell types. By incorporating lung cancer GWAS data, we will identify cell-type specific susceptibility genes. Our dataset will provide a unique resource for lung cancer research. Citation Format: Thong Luong, Erping Long, Jinhu Yin, Bolun Li, Ju Hye Shin, Elelta Sisay, Alexander Kane, Alyxandra Golden, Yoon Soo Chang, Nicholas Banovich, Nathaniel Rothman, Jinyoung Byun, Qing Lan, Christopher Amos, Jianxin Shi, Jin Gu Lee, Eun Young Kim, Jiyeon Choi. Establishing a single-cell eQTL dataset of lung tissues from Asian never-smokers to identify cell-type specific lung cancer susceptibility genes 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 (6Suppl): Abstract nr 7331.