20P - Preclinical characterization of FGFR1-4 variants of unknown significance

Background

Fibroblast Growth Factor Receptors (FGFR1, FGFR2, FGFR3, FGFR4) are oncogenes mutated in 5% - 10% of cancer patients. Oncogenic activation of FGFRs can occur through rearrangements or point mutations. Functional information on point mutations outside of hotspots or the tyrosine kinase domain is scarce, and such mutations are commonly classified as variants of unknown significance (VUS). Preclinical testing of unknown mutations can provide functional information about their oncogenic potential and possible treatment options.

Methods

We searched the Catalogue Of Somatic Mutations In Cancer (COSMIC) for recurring variants located in the extracellular, transmembrane, or juxta-membrane protein domains of FGFRs. Candidates for in vitro characterization were chosen based on mutational frequency, position, lack of published data, and presumed functional importance. NIH/3T3 fibroblasts expressing FGFR VUS were generated using site-directed mutagenesis and retroviral transduction. The activating potential of FGFR variants was assessed with high-throughput screen and functional assays testing growth in the absence of serum, anchorage-independency (anoikis), morphological transformation, and inhibition by FGFR targeting drugs.

Results

We selected 76 FGFR mutations for functional preclinical characterization. High-throughput screening identified 22 candidates with potential activating effects. Hits were subsequently validated and functionally characterized in vitro confirming 12 novel variants with activating function. Results from in vitro assays were compared to functional predictions made by computational models. We report drug sensitivity profiles for nine approved tyrosine kinase inhibitors or inhibitors in development which reveal potential therapeutic vulnerabilities for all newly identified activating variants.

Conclusions

We characterized novel FGFR variants with activating functional effect in vitro and explored their sensitivity to targeted therapies. Data generated by preclinical investigations can contribute to overcoming the knowledge gap for FGFR VUS and indicate potential treatment options.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

German Cancer Aid.

Disclosure

S. Loges: Financial Interests, Institutional, Full or part-time Employment: University Medical Centre Mannheim, German Cancer Research Centre Heidelberg; Financial Interests, Personal, Advisory Role: Lilly; Non-Financial Interests, Personal, Advisory Board: Sanofi, BerGenBio, Novartis, Boehringer Ingelheim, BMS, Roche, AstraZeneca, MSD, Merck, Sanofi Aventis, Janssen, Takeda, Pfizer, Amgen, Bayer, Medac, Daiichi Sankyo; Financial Interests, Personal, Financially compensated role: Lilly, Sanofi, BerGenBio, Novartis, Boehringer Ingelheim, BMS, Roche, AstraZeneca, MSD, Merck, Sanofi Aventis, Janssen, Takeda, Pfizer, Amgen, Bayer, Medac, Daiichi Sankyo; Financial Interests, Institutional, Research Grant: Roche, BerGenBio, Lilly, ADC Therapeutics, Daiichi Sankyo. All other authors have declared no conflicts of interest.