Activation of FGFR2 Signaling Suppresses BRCA1 and Drives Triple-Negative Mammary Tumorigenesis That is Sensitive to Immunotherapy.

Fibroblast growth factor receptor 2 (FGFR2) is a membrane-spanning tyrosine kinase that mediates FGF signaling. Various FGFR2 alterations are detected in breast cancer, yet it remains unclear if activation of FGFR2 signaling initiates tumor formation. In an attempt to answer this question, a mouse model berrying an activation mutation of FGFR2 (FGFR2-S252W) in the mammary gland is generated. It is found that FGF/FGFR2 signaling drives the development of triple-negative breast cancer accompanied by epithelial-mesenchymal transition that is regulated by FGFR2-STAT3 signaling. It is demonstrated that FGFR2 suppresses BRCA1 via the ERK-YY1 axis and promotes tumor progression. BRCA1 knockout in the mammary gland of the FGFR2-S252W mice significantly accelerated tumorigenesis. It is also shown that FGFR2 positively regulates PD-L1 and that a combination of FGFR2 inhibition and immune checkpoint blockade kills cancer cells. These data suggest that the mouse models mimic human breast cancers and can be used to identify actionable therapeutic targets.

Preclinical Combination Studies of an FGFR2 Targeted Thorium-227 Conjugate and the ATR Inhibitor BAY 1895344.

PURPOSE: Fibroblast growth factor receptor 2 (FGFR2) has been previously reported to be overexpressed in several types of cancer, whereas the expression in normal tissue is considered to be moderate to low. Thus, FGFR2 is regarded as an attractive tumor antigen for targeted alpha therapy. This study reports the evaluation of an FGFR2-targeted thorium-227 conjugate (FGFR2-TTC, BAY 2304058) comprising an anti-FGFR2 antibody, a chelator moiety covalently conjugated to the antibody, and the alpha particle-emitting radionuclide thorium-227. FGFR2-TTC was assessed as a monotherapy and in combination with the DNA damage response inhibitor ATRi BAY 1895344. METHODS AND MATERIALS: The in vitro cytotoxicity and mechanism of action were evaluated by determining cell viability, the DNA damage response marker γH2A.X, and cell cycle analyses. The in vivo efficacy was determined using human tumor xenograft models in nude mice. RESULTS: In vitro mechanistic assays demonstrated upregulation of γH2A.X and induction of cell cycle arrest in several FGFR2-expressing cancer cell lines after treatment with FGFR2-TTC. In vivo, FGFR2-TTC significantly inhibited tumor growth at a dose of 500 kBq/kg in the xenograft models NCI-H716, SNU-16, and MFM-223. By combining FGFR2-TTC with the ATR inhibitor BAY 1895344, an increased potency was observed in vitro, as were elevated levels of γH2A.X and inhibition of FGFR2-TTC-mediated cell cycle arrest. In the MFM-223 tumor xenograft model, combination of the ATRi BAY 1895344 with FGFR2-TTC resulted in significant tumor growth inhibition at doses at which the single agents had no effect. CONCLUSIONS: The data provide a mechanism-based rationale for combining the FGFR2-TTC with the ATRi BAY 1895344 as a new therapeutic approach for treatment of FGFR2-positive tumors from different cancer indications.

Preclinical Evaluation of the Pan-FGFR Inhibitor LY2874455 in FRS2-Amplified Liposarcoma.

Background: FGFR inhibition has been proposed as treatment for dedifferentiated liposarcoma (DDLPS) with amplified FRS2, but we previously only demonstrated transient cytostatic effects when treating FRS2-amplified DDLPS cells with NVP-BGJ398. Methods: Effects of the more potent FGFR inhibitor LY2874455 were investigated in three DDLPS cell lines by measuring effects on cell growth and apoptosis in vitro and also testing efficacy in vivo. Genome, transcriptome and protein analyses were performed to characterize the signaling components in the FGFR pathway. Results: LY2874455 induced a stronger, longer-lasting growth inhibitory effect and moderate level of apoptosis for two cell lines. The third cell line, did not respond to FGFR inhibition, suggesting that FRS2 amplification alone is not sufficient to predict response. Importantly, efficacy of LY2874455 was confirmed in vivo, using an independent FRS2-amplified DDLPS xenograft model. Expression of FRS2 was similar in the responding and non-responding cell lines and we could not find any major difference in downstream FGFR signaling. The only FGF expressed by unstimulated non-responding cells was the intracellular ligand FGF11, whereas the responding cell lines expressed extracellular ligand FGF2. Conclusion: Our study supports LY2874455 as a better therapy than NVP-BGJ398 for FRS2-amplified liposarcoma, and a clinical trial is warranted.

3-Dimensional micropillar drug screening identifies FGFR2-IIIC overexpression as a potential target in metastatic giant cell tumor.

We established two patient derived tumor cells (PDCs) from right and left pulmonary metastatic lesions respectively of a patient with giant cell tumor. At that time, patient-derived tumor cells from right and left surgical specimens were collected and cultured. High-throughput screening (HTS) for 24 drugs was conducted with a micropillar/microwell chip platform using giant cell tumor PDCs. Using 6 doses per drug in 6 replicates for giant cell tumor PDCs, the dose response curves and corresponding IC50 values were calculated from the scanned images using the S+ Chip Analyzer. A sensitive response was more significantly achieved for AZD4547 (FGFR2 inhibitor) in giant cell tumor PDCs originated from the right pulmonary nodule under the micropillar/microwell chip platform using 3D culture. This sensitivity was consistent with the target expression patterns of giant cell tumor PDCs (FGFR2-IIIC mRNA expression in giant cell tumor PDCs originated from the right pulmonary nodule was increased significantly as compared to those originated from left). However, in a conventional 2D cultured MTT assay, there was no difference for IC50 values of AZD4547 between giant cell tumor PDCs originated from right and left pulmonary nodules. An HTS platform based on 3D culture on micropillar/microwell chips and PDC models could be applied as a useful preclinical tool to evaluate the intrapatient tumor/response heterogeneity. This platform based on 3D culture might reflect far better the relation between the tumor-biology and the matched targeted agent as compared to a conventional 2D cultured MTT assay.

KGFR as a possible therapeutic target in middle ear cholesteatoma.

CONCLUSION: We demonstrated that repression of keratinocyte growth factor (KGF) receptor (KGFR) could be a potentially useful strategy in the conservative treatment of middle ear cholesteatoma. OBJECTIVES: Recently, the use of a selective inhibitor of the KGFR, SU5402, in an in vitro experiment resulted in the inhibition of the differentiation and proliferation of epithelial cells through KGF secretion by fibroblasts isolated from the cholesteatoma. In this study, we investigated the effects of the KGFR inhibitor during middle ear cholesteatoma formation in vivo. METHODS: Based on the role of KGF in the development of cholesteatoma, Flag-hKGF cDNA driven by CMV14 promoter was transfected through electroporation into the external auditory canal of rats five times on every fourth day. Ears transfected with empty vector were used as controls. KGFR selective inhibitor (SU5402) or MEK inhibitor (PD0325901) was administered in the right ear of five rats after vector transfection. In the control, 2% DMSO in PBS was administered in the other ears after vector transfection. RESULTS: The use of a selective KGFR inhibitor, SU5402, completely prevented middle ear cholesteatoma formation in the rats.

Enhanced FGFR signalling predisposes pancreatic cancer to the effect of a potent FGFR inhibitor in preclinical models.

BACKGROUND: Fibroblast growth factor receptor (FGFR) signalling has been implicated in pancreas carcinogenesis. We investigated the effect of FGFR inhibition in pancreatic cancer in complementary cancer models derived from cell lines and patient-derived primary tumour explants. METHODS: The effects of FGFR signalling inhibition in pancreatic cancer were evaluated using anti-FRS2 shRNA and dovitinib. Pancreatic cancers with varying sensitivity to dovitinib were evaluated to determine potential predictive biomarkers of efficacy. Primary pancreatic explants with opposite extreme of biomarker expression were selected from 13 tumours for in vivo dovitinib treatment. RESULTS: Treatment with anti-FRS2 shRNA induced significant in vitro cell kill in pancreatic cancer cells. Dovitinib treatment achieved similar effects and was mediated by Akt/Mcl-1 signalling in sensitive cells. Dovitinib efficacy correlated with FRS2 phosphorylation status, FGFR2 mRNA level and FGFR2 IIIb expression but not phosphorylation status of VEGFR2 and PDGFRß. Using FGFR2 mRNA level, a proof-of-concept study using primary pancreatic cancer explants correctly identified the tumours' sensitivity to dovitinib. CONCLUSION: Inhibiting FGFR signalling using shRNA and dovitinib achieved significant anti-cancer cancer effects in pancreatic cancer. The effect was more pronounced in FGFR2 IIIb overexpressing pancreatic cancer that may be dependent on aberrant stimulation by stromal-derived FGF ligands.