Study on acquisition timing of KRAS mutations in the pancreatic ductal epithelium

[Introduction] Pancreatic cancer is a highly aggressive malignancy with a poor prognosis, underscoring the need for early diagnosis KRAS mutations are among the earliest genetic events in pancreatic carcinogenesis, but the timing of KRAS mutation acquisition in the pancreatic ductal epithelium remains unclear. [Methods] From surgical specimens, we collected non-neoplastic pancreatic tissues distant from the macroscopic tumor, from which pancreatic duct-derived organoids were established. Paired tumor/normal DNA was extracted and subjected to whole-exome sequencing (WES). Phylogenetic trees were reconstructed based on shared somatic mutations. In patients with pancreatic cancer or intraductal papillary mucinous neoplasm (IPMN), WES was also performed on DNA from formalin-fixed, paraffin-embedded (FFPE) tumor tissue to assess relationships between cancer and non-cancerous KRAS -mutant clones. In a subset of bulk organoids, single cell-derived organoids were further established to determine mutation accumulation rates. [Results] We analyzed 430 organoid samples from patients with (n=25) and without (n=23) pancreatic cancer. A median of 33 mutations per sample was identified. Eighteen KRAS mutations were detected in 16 cancer patients compared with four in four non-cancer controls, indicating a higher prevalence and wider distribution of KRAS -mutant clones in cancer patients. Phylogenetic analysis showed that 13 of 22 KRAS mutations mapped to the major trunk, which typically spread into multiple (n=6 on average) branches, whereas KRAS -wild-type clones rarely extended to multiple branches, suggesting more localized expansion. We further analyzed 56 single-cell-derived organoids from 19 cases, of which 7 harbored KRAS mutations. In KRAS -wild-type cells, the mutational burden positively correlated with age with a rate of 0.29 mutations per exome per year. KRAS-mutant cells carried significantly more mutations than wild-type cells, suggesting an increased mutation rate after KRAS activation. Based on this mutation rate, we estimated that these KRAS mutations were acquired between 21 and 44 years of age. Among 16 cases with KRAS -mutant PDAC or IPMN, we analyzed 8 FFPE tumors. In one patient with metachronous intraductal papillary mucinous carcinoma, the primary tumor, recurrent tumor, and KRAS -mutant organoid clone all arose from a common ancestral clone. Phylogenetic analysis indicated that a clone shared by the primary and recurrent tumors first branched from this ancestor and subsequently gave rise to both lesions. In other cases, KRAS -mutant clones in non-neoplastic ducts were phylogenetically independent of the cancer clone, indicating parallel evolution of multiple KRAS-mutant lineages. [Conclusions] We elucidated the evolutionary history of clones in non-cancerous pancreatic ductal epithelium. These findings improve our understanding of the early events in pancreatic carcinogenesis.