[P-16] Characterisation of genomic alteration of FGFR3 by integrating analysis of the Cancer Genome Atlas

Thao Quyen Nguyen, Quoc Thang Pham, Thanh Tu Nguyen, Trong Hieu Le, Tuan Dung Vu, Thi Thanh Tam Bui, Thanh Tu Duong, Dang Anh Thu Phan and Quoc Dat Ngo

Department of Pathology, University of Medicine and Pharmacy at Ho Chi Minh City, Vietnam

Background and Objectives: Mutations and fusions of the fibroblast growth factor receptor (FGFR) gene family occur in various cancer types. Recently erdafitinib has been approved by the FDA for the treatment of FGFR3-altered urothelial cancer. We performed an integrated analysis of the TCGA-pancancer atlas including 32 cancer types (n = 10,953) to reveal the novel alteration of FGFR3.

Materials and Methods: To clarify FGFR3 alterations we performed an analysis on the server https://www.cbioportal.org/. The FGFR3 alterations and clinicopathologic data were also downloaded for further investigation. We used the CellMiner-web-server https://discover.nci.nih.gov/cellminercdb/ with the NC-60 and GDSC cancer cell line database to validate the cytotoxicity of FGFR3 inhibitors.

Results: FGFR3 alterations were found in 351 (3%) of queried patients of which amplification, mutations and rearrangements were 120 (34.2%), 172 (49%) and 38 (10.8%), respectively. Seventy-one (41.3%) of FGFR3 mutations were putative-driver missense mutations in which the most frequent were FGFR3 p.S249C (57.7%) and Y373C/H (15.5%). FGFR3-TACC3 was the most prevalent FGFR3 rearrangement (84.2%). Tumour types with the most frequency of FGFR3 alterations were bladder (32.5%) and endometrial (10.6%) cancer. The cytotoxicity of FGFR3 inhibitors (AZD-4547, PD173074) was found negatively correlated with FGFR3 expression across the NC-60 database. Focus on the endometrial carcinoma cell lines of GDSC tested with PD173074 showed negative correlation between FGFR3 expression and response to PD173074.

Conclusion: FGFR3 alterations are present at low frequencies and consistency across many cancers. Bladder and endometrial cancers were the highest frequency detections of the FGFR3 alterations. In-silico analysis revealed FRFR3 may be a pivotal targeted therapy in endometrial cancer.