Abstract 1075: Buccal mucosal remodeling predicts esophageal cancer risk

Abstract

Abstract Lifestyle habits, including alcohol consumption, tobacco smoking, and germline variants of ALDH2 and ADH1B, increase the risk of developing esophageal squamous cell carcinoma (ESCC). However, current prediction methods that rely on self-reported lifestyle information lack accuracy. We previously demonstrated that driver-mutated clones remodel the esophageal epithelium with age and that this remodeling is accelerated in heavy drinkers and smokers. Nonetheless, predicting cancer from clonal expansion in normal tissues remains challenging, due to limited sample availability beyond the blood. The squamous epithelium continuously covers the esophagus, throat, and oral cavity. Therefore, we evaluated buccal mucosal remodeling by mutant clones and its potential as a predictive biomarker of ESCC. A total of 387 buccal mucosa samples were collected from 110 subjects, including 62 with ESCC and 48 without, using swabs of three different sizes, followed by error-corrected sequencing targeting 26 driver genes that showed positive selection in samples from ESCC and physiologically normal esophageal tissues, as well as ALDH2 and ADH1B. Somatic mutations were detected in all samples (median, 27 (1-134)). In total, 13,235 mutations, including 12,085 nonsynonymous and 1,150 synonymous mutations, with VAFs ranging from 0.0016-0.12 (median, 0.0025), were detected. Ten of the 26 genes, TP53, CHEK2, NOTCH1, CDKN2A, NOTCH2, FAT1, AJUBA, PPM1D, ARID1A, and ZNF750 showed positive selection (dN/dS >1, q<0.1). Correlation analysis demonstrated that driver mutations in TP53, FAT1, and PPM1D were positively correlated with not only germline risk variants of ALDH2 but also alcohol consumption. The impact of alcohol consumption on the number and sum of VAFs of mutations in each of the three genes was analyzed and stratified by the presence or absence of the germline risk variant of ALDH2, and positive correlations were observed only among subjects with risk variants. Compared to subjects without ESCC, patients with early-stage or advanced ESCC displayed significantly higher numbers and sums of VAFs of driver mutations (p<0.05). These results indicated that buccal mucosal remodeling driven by positively selected mutations is associated with lifestyle and germline risk factors for ESCC, thus motivating the establishment of a predictive model for ESCC using genetic swab data. The prediction model for ESCC using sequencing data outperformed a model using self-reported lifestyle and alcohol flushing reaction and a model using lifestyle and germline risk variants (AUC of 0.94, 0.72, and 0.88, respectively). In conclusion, this is the first study to use clonal expansion in normal tissues to predict the presence of solid organ cancer, and our study successfully established a prediction model for ESCC based on buccal mucosal remodeling and germline variants. Citation Format: Akira Yokoyama, Tomonori Hirano, Koichi Watanabe, Masashi Tamaoki, Kenshiro Hirohashi, Yoshihiro Ishida, Yoshikage Inoue, Yasuhide Takeuch, Yo Kishimoto, Soo Ki Kim, Chikatoshi Katada, Yasuhito Nanya, Hiroshi Seno, Seishi Ogawa, Manabu Muto, Nobuyuki Kakiuchi. Buccal mucosal remodeling predicts esophageal cancer risk [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 1075.