Background: Triple negative breast cancer (TNBC), which accounts for 10-17% of all breast cancers, is challenging to treat due to its heterogeneity and the absence of identified targeted therapies. Heat shock protein 90 kDa (Hsp90) is a molecular chaperone protein essential for regulating the stability and activity of several client proteins required for oncogenesis, and selective Hsp90 inhibitors have shown anti-cancer activity in multiple cancer models including TNBC. Ganetespib, a second generation Hsp90 inhibitor, has shown promising clinical activity in TNBC patients but responders develop acquired resistance. Here, we generated an in vitro model of ganetespib-resistant TNBC cells and investigated the mechanisms of resistance to ganetespib in TNBC using a 96-well based small molecule screen.
Methods: Ganetespib treatment sensitivity in a panel of seven TNBC cell lines (HS578T, HCC1143, HCC1937, MDA-MB-468, MDA-MB-231, MDA-MB-453 and BT20) was assessed using a resazurin-based cell proliferation assay. HS578T cells were incubated in a high concentration of ganetespib (>IC50) to generate ganetespib-resistant clones. The effects on Hsp90 oncogenic client proteins (AKT and EGFR) and Hsp70 were determined by western blotting. Cross resistance to other Hsp90 inhibitors was also assessed. 96-well based small molecule screen with a library of 326 bioactive small molecule inhibitors was performed on parental HS578T and ganetespib-resistant clones in the absence and presence of ganetespib. Resazurin-based cell proliferation assay was used as an end point. Positive hits were identified as log10(percentage of proliferation in parental/resistant) ≥ 2SD.
Results: TNBC cell lines showed differential sensitivity to ganetespib (IC50 range: 4.75 to 20.16 nM), with HS578T being the most sensitive and HCC1143 the most resistant. Independent ganetespib-resistant HS578T clones (CR2 and CR3) were successfully generated by incubating parental HS578T cells in ganetespib (30 nM) for 42 days. IC50 values for the ganetespib-resistant clones (CR2: 15.57 ± 1.90 nM and CR3: 20.28 ± 2.75 nM) were 3 to 4-fold higher than the parental HS578T cells (IC50: 4.79 ± 0.32 nM). The client proteins were not further down-regulated after ganetespib treatment in the ganetespib-resistant clones compared to the parental HS578T cells. Resistant clones also showed cross resistance to 1st generation (17-AAG) and 2nd generation (NVP-AUY922) Hsp90 inhibitors. The small molecule screen was reproducible as the independent runs and duplicates strongly correlated (R2 >0.7). The ganetespib-resistant CR3 clones showed selective sensitivity to 11/326 compounds (Z score ≥ 2), only in the presence of ganetespib when compared to the parental HS578T cells, most notable being LY2784544, a JAK2 inhibitor.
Conclusion: Ganetespib-resistant HS578T cells provide a promising model to study the resistance mechanisms to ganetespib in TNBC. Our current data suggest that TNBC cells are responsive to JAK2 inhibition in the presence of ganetespib despite prior acquisition of resistant to ganetespib. Activation of a JAK2/STAT3 survival pathway may be a potential resistance mechanism to Hsp90 inhibition and combined Hsp90 and JAK2 inhibition could potentially be a useful therapeutic strategy for TNBC.
Citation Format: Nuramalina H. Mumin, Sivan Bokobza, Fiona Cahill, Neele Drobnitzky, Yanyan Jiang, Luiza Madia Lourenco, Aoife Vaughan, Anika Weber, Anthony Kong, Anderson Ryan. Mechanisms of resistance to the Hsp90 inhibitor ganetespib in triple negative breast cancer. [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2015 Nov 5-9; Boston, MA. Philadelphia (PA): AACR; Mol Cancer Ther 2015;14(12 Suppl 2):Abstract nr B88.