Abstract 3918: Kinase domain duplications of FGFR1 and MET are potential therapeutic targets

Abstract

Abstract Background: Kinase domain duplication (KDD) is recently recognized as an oncogenic driver. Therapeutic potential has been established in EGFR-KDDs in non-small cell lung cancer, and our group recently showed resistance mechanism in lung cancers harboring EGFR-KDDs treated with EGFR tyrosine kinase inhibitors (TKIs). However, the frequency and oncogenic role of KDDs in other receptor tyrosine kinases has been elusive. Methods: To investigate the frequency of KDDs in advanced cancer patients, we developed in-house algorithm to detect KDDs in high-depth panel sequencing dataset and evaluated its performance using large-scale panel sequencing data from Seoul National University Hospital (SNUH). In short, the algorithm collects reads with soft clips from a modified GATK best practice pipeline for clinical NGS data processing. The collected reads are then grouped according to their matched position and insert size, and evidences of each group (number of splitting and spanning reads) are collected to yield a set of initial candidates. Finally, the initial candidates were manually inspected after quality control by removing false-positive candidates, such as non-target genes or consistent findings. Then, we established cell lines transfected with KDDs of FGFR1 and MET to validate oncogenic role of identified KDDs in the cohort. Using these transformed cell lines, we investigated carcinogenicity with cell viability assay, immunoblot assay and colony formation assay. FGFR TKIs and MET TKIs were used to uncover the possibility of therapeutic targets in the KDDs. Results: Using the in-house algorithm, we identified FGFR1-KDDs from brain tumors (n=4), EGFR-KDDs from lung (n=1) and brain (n=2) tumors, and MET-KDD from a lung tumor (n=1) in a total of 3,932 cancer samples, with 0.2% frequency. All three kinds of KDDs contained full-length of kinase domains of each gene. Ba/F3 cells transfected with the FGFR1-KDD constructs proliferated well without IL-3 (1239- and 1019-fold higer than wild-type FGFR1), while wild-type FGFR1 Ba/F3 cells did not. This growth potential was diminished by FGFR TKIs BGJ398 (IC50, 0.063 ± 0.065nM), PD173074 (IC50, 1.498 ± 1.021nM), and ponatinib (IC50, 5.69 ± 3.05nM). Reduced expressions of downstream pathway proteins including phospho-FGFR1, phospho-Akt, and phospho-Erk were confirmed by immunoblot assays. NIH-3T3 cells transfected with MET-KDD construct also showed growth advantage compared to wild-type MET NIH-3T3 cells. Capmatinib, a MET TKI, inhibited colony formation of MET-KDD NIH-3T3 cells with reduced phospho-MET expression. Conclusion: In this study, we validated newly developed KDD-detection computational methods using targeted DNA sequencing data. Using this algorithm, we found 0.2% of KDDs in 3,932 cancer samples, including EGFR, FGFR1, and MET. In addition to EGFR-KDDs, our study reveals that both FGFR1- and MET-KDD can be oncogenic and therapeutic targets in brain and lung cancer patients. Citation Format: Chaelin Lee, Sungyoung Lee, Hongseok Yun, Jeonghwan Youk, Tae Min Kim, Soyeon Kim, Bhumsuk Keam, Miso Kim, Dong-Wan Kim. Kinase domain duplications of FGFR1 and MET are potential therapeutic targets. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 3918.