Abstract 6083: Evolutionary trajectories during carcinogenesis and metastasis in gallbladder adenocarcinoma

Abstract Background: There is no study so far that analyzed the patterns of clonal evolution from the initiation of carcinogenesis to distant metastasis in patients with gallbladder adenocarcinoma (GBAC). This study aimed to elucidate the evolutionary trajectories of GBAC using multi-regional and longitudinal tumor samples. Methods: Whole-exome sequencing was conducted on tumors and matched normal samples with the coverage of 300x and 200x, respectively. Using PyClone, CITUP, MapScape, and TimeScape, phylogenetic trees were visualized in each patient considering each tumor’s location and timing of acquisition. Mutational signatures were analyzed using Mutalisk. Results: Between 2013 and 2018, a total of 11 patients (male, 5) including 2 rapid autopsy cases were enrolled. The median age was 70 (range, 59-75) years. The dataset consisted of 11 normal samples, 4 biliary intraepithelial neoplasia (BilIN), 11 primary tumors, and 30 metastatic tumors. The most frequently altered gene was ERBB2 (54.5%), followed by TP53 (45.5%) and FBXW7 (27.3%). Of 6 patients with analyzable tumor ploidy, 2 patients (33.3%) had whole genome doubling (WGD) in both primary and metastatic tumors, and 1 patient (16.7%) had WGD not in the primary tumor but in the liver metastasis. In the BilIN analysis (n=4), most mutations in frequently altered genes in GBAC were detectable from the BilIN stage, but some of them were subclonal. In these 4 patients, the fittest subclone in BilIN underwent linear and branching evolution by acquiring additional subclonal mutations and thus expanded in the primary tumor, suggesting a selective sweep phenomenon. In combined analysis with metastatic tumors (n=11), branching and linear evolution was identified in 9 (81.8%) and 2 (18.2%) patients, respectively. Of the 9 patients with branching evolution, eight (88.9%) had a total of 11 subclones expanded at least 7-fold in regional or distant metastasis. These subclones harbored putative metastasis-driving mutations in tumor suppressor genes such as SMAD4, ROBO1, and DICER1. Metastases were polyclonal in all patients. However, metastatic lesions in the same or adjacent organs showed similar clonal compositions and there was evidence of metastasis-to-metastasis spread. In mutational signature analysis, we identified 6 mutational signatures: signatures 1 (age), 3 (DNA double-strand break-repair), 7 (ultraviolet), 13 (APOBEC), 22 (aristolochic acid), and 24 (aflatoxin) (cosine similarity values ≥ 0.9). Among them, signatures 1 and 13 were dominant at the carcinogenesis stage while signatures 22 and 24 were dominant at the metastasis stage. Conclusions: We have constructed evolutionary trajectories of individual patients, highlighting the role of each clone and the dynamics among clones during carcinogenesis and metastasis. This novel approach may help us move forward to precision medicine that enables early detection of carcinogenesis and metastasis. Citation Format: Minsu Kang, Hee Young Na, Soomin Ahn, Ji-Won Kim, Sejoon Lee, Soyeon Ahn, Ju Hyun Lee, Jeonghwan Youk, Haesook T. Kim, Kui-Jin Kim, Koung Jin Suh, Jun Suh Lee, Se Hyun Kim, Jin Won Kim, Yu Jung Kim, Keun-Wook Lee, Yoo-Seok Yoon, Jee Hyun Kim, Jin-Haeng Chung, Ho-Seong Han, Jong Seok Lee. Evolutionary trajectories during carcinogenesis and metastasis in gallbladder adenocarcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 6083.