Abstract P3-03-01: A novel high-throughput secreted factor screen and bioinformatics pipeline identifies microenvironment-derived FGF2 as a mechanism of resistance to anti-estrogens, PI3K, and mTOR inhibitors in ER+ breast cancer

• Published in: Cancer research (Chicago, Ill.) Vol. 77; no. 4_Supplement; pp. P3 - P3-03-01
• Main Authors: Shee, K, Hinds, JW, Hampsch, RA, Golub, TR, Straussman, R, Miller, TW
• Format: Journal Article
• Language: English
• Published: 15.02.2017
• ISSN: 0008-5472; 1538-7445
• DOI: 10.1158/1538-7445.SABCS16-P3-03-01

Abstract Resistance to anti-estrogen therapy is a serious obstacle to the treatment of patients with ER+ breast cancer. Unlike the majority of preclinical studies that focus on cancer cell-intrinsic mechanisms of resistance, we hypothesized that the tumor microenvironment significantly contributes to drug resistance, and that individual factors present in the microenvironment differentially modulate response to therapy. To systematically test this hypothesis, high-throughput screens were performed in which 297 unique secreted proteins (cytokines, growth factors, and extracellular matrix) were tested on ER+ breast cancer cell lines (MCF-7 and T47D) treated +/- anti-estrogen fulvestrant or the PI3K inhibitor GDC-0941 (pictilisib). Cytokines that rescued cells (hits) were validated in two additional ER+ breast cancer cell lines (ZR75-1 and HCC-1500), and expanded to include the anti-estrogen tamoxifen and the mTORC1 inhibitor everolimus. Hits were also found to rescue from combinations of fulvestrant/GDC-0941 and fulvestrant/everolimus in models with acquired resistance to fulvestrant. Multiple factors, such as ErbB family ligands and fibroblast growth factors (FGFs), were among the top validated hits. To parse out which hits are most likely to be relevant in the setting of the ER+ breast tumor microenvironment, a bioinformatics filter was developed to incorporate gene and protein expression data for secreted proteins in non-cancer human tissues relevant to ER+ breast cancer. These tissues include primary breast tissue components (breast mammary, adipose, primary fibroblasts) and common metastatic sites (bone marrow, lung, liver). After filtering, the top hit was fibroblast growth factor 2 (FGF2), which significantly drives resistance to anti-estrogens, PI3K, and mTOR inhibitors, and is highly expressed in non-cancer tissues relevant to ER+ breast cancer. FGF2-mediated rescue was completely abrogated by the ATP-competitive pan-FGFR inhibitor PD173074, confirming FGFR kinase specificity of the rescue phenotype. Immunoblot data confirmed that FGF2 induced phosphorylation of FGFR and the FGFR effector FRS2, and a consistent pattern of downstream MEK-ERK-Rsk90 activation for rescue to all drugs. Flow cytometry data showed that FGF2-mediated rescue leads to decreases in drug-induced apoptosis and G1 cell cycle arrest, which correspond at the protein level to increased degradation of the pro-apoptotic protein Bim and upregulation of the cell cycle driver cyclin D1, respectively. Mice bearing MCF-7 xenografts were treated with vehicle, FGF2 (20 ug/kg/d, s.c.), fulvestrant (5 mg/wk, s.c.), or the combination. Vehicle-treated and FGF2-treated mice showed similar rates of tumor growth. Single-agent fulvestrant significantly suppressed tumor growth, while the addition of FGF2 rescued from the growth-inhibitory effect of fulvestrant. Studies in other tumor models are ongoing. These data collectively suggest that stroma-derived FGF2-mediated drug resistance is a novel therapeutic opportunity in ER+ breast cancer. Citation Format: Shee K, Hinds JW, Hampsch RA, Golub TR, Straussman R, Miller TW. A novel high-throughput secreted factor screen and bioinformatics pipeline identifies microenvironment-derived FGF2 as a mechanism of resistance to anti-estrogens, PI3K, and mTOR inhibitors in ER+ breast cancer [abstract]. In: Proceedings of the 2016 San Antonio Breast Cancer Symposium; 2016 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2017;77(4 Suppl):Abstract nr P3-03-01.