Background: Aberrant induction of the epithelial to mesenchymal transition cascade (EMT) has a defining role in shaping cancer malignancy. Despite the complexity of this process, the alternative splicing of the fibroblast growth factor receptor 2 (FGFR2), between the epithelial-specific FGFR2-IIIb and mesenchymal-specific FGFR2-IIIc isoforms, has been shown to play a pivotal role in the EMT cascade. Aberrant splicing of FGFR2 in carcinomas has been shown to mark invasive cancerous cells that have undergone an EMT event and have gained the ability to thrive in diverse microenvironments. In addition to improper splicing patterns, overexpression of either isoform is sufficient to induce tumor formation by constitutively activating common oncogenic pathways. For example, a recent microarray analysis identified amplification of the FGFR2-IIIb locus as a common oncogenic signature in a significant number of treatment-resistant triple negative breast cancers (TNBC). Furthermore, inhibition of FGFR2 signaling in candidate TNBCs was sufficient to induce tumor regression. Due to the intimate link that exists between FGFR2 and breast cancer (BC), isoform-specific antibodies against the receptor may be able to identify tumors that possess deregulated FGFR2 signaling and subsequently inhibit their growth by blocking FGF ligand binding.

Methods: To develop FGFR2-IIIb and —IIIc specific antibodies, we initially cloned and expressed the alternatively spliced region of each isoform and provided sufficient amount to ThermoScientific for rabbit immunizations. After the 72-day protocol was completed, the raw anti-IIIb/IIIc serum was passed through a column containing immobilized GST-IIIc/IIIb, respectively, to remove cross-reactive antibodies. Antibody specificity was assessed through western blotting and immunofluorescence (IF). Therapeutic validity of serum was determined by its ability to inhibit the growth of cancers in-vitro.

Results: Through western blotting, we were able to demonstrate that the anti-IIIb serum was able to specifically recognize the FGFR2-IIIb isoform. However, the anti-IIIc serum exhibited substantial cross-reactivity to FGFR2-IIIb and was therefore not included in downstream studies. By IF, the anti-IIIb serum specifically recognized FGFR2-IIIb expressing cells while exhibiting background levels of reactivity to those that expressed FGFR2-IIIc, FGFR1, or FGFR4. To test the therapeutic validity of the anti-IIIb serum, we conducted cell inhibition assays using MCF-7 (FGFR2-IIIb specific), BT-549 (FGFR2-IIIc specific), and PC3 (FGFR2 negative) cancer lines to test whether the growth inhibitory effects of the antibody could specifically target FGFR2-IIIb expressing cancer cell lines. After 5 days of treatment with anti-IIIb or pre-immune rabbit serum, we observed potent growth inhibition of the MCF-7 BC cells (p = .02) in anti-IIIb treated wells and minimal effects on BT-549 and PC3 cells.

Conclusion: Based upon the observed data, the anti-IIIb serum that we developed not only identified FGFR2-IIIb expressing cells through IF, but also inhibited their growth in vitro. Taken together, our data suggest that FGFR2-IIIb may be a promising therapeutic and predictive marker.

Citation Information: Cancer Res 2012;72(24 Suppl):Abstract nr P4-06-07.