A systems approach reveals SMARCD3 regulon drives alternate growth signaling and vulnerability to MEK inhibitors in everolimus resistant breast cancer cells

Estrogen receptor-positive breast cancer (ER+BC) represents nearly 75% of all breast tumors, with nearly 20-40% patients eventually developing metastases. Everolimus is an mTOR inhibitor which has been approved for use in combination with exemestane in patients with metastatic ER+BC. However, many patients eventually develop resistance to everolimus leading to poor survival outcomes. Systematic characterization of the molecular phenotypes underlying everolimus resistance could reveal new therapeutic targets for metastatic ER+BC. We analyzed transcriptomic profiles from multiple pairs of everolimus resistant or sensitive ER+BC cell lines, to identify signaling mechanisms that distinguish resistant cells both at baseline and under everolimus treatment. Using a linear mixed-effects model, we identified strong activation of the growth-factor signaling meta-phenotype in resistant cells at baseline, which persisted even under everolimus treatment. This meta phenotype included alternate growth signaling mechanisms including IGF1R/insulin, ESR1, RAS and MAPK signaling. To understand the regulatory mechanism behind alternate growth-factor signaling activation, we adopted a network reconstruction architecture to identify SMARCD3 regulons as key transcriptional regulatory networks in resistant cells. In a neo-adjuvant trial for ER+BC patients, the expression levels of SMARCD3 and its regulon’s activity were significantly elevated in tumors refractory to everolimus. To develop a therapeutic strategy against everolimus resistance, we used transcriptomic and drug sensitivity profiles of 926 cancer cell lines treated with 395 compounds. We found that the activation of SMARCD3 regulons was consistently associated with sensitivity to several known MEK1/2 inhibitors, including trametinib. To validate these findings, we evaluated the efficacy of everolimus in combination with trametinib and demonstrated that the combination treatment significantly reduced the growth of resistant cells compared to everolimus. Our results demonstrate that everolimus resistant ER+BC cells utilize alternate growth-factor signaling to escape therapeutic control. We found that this phenotype is strongly linked with activation of the SMARCD3 regulons and can be therapeutically targeted using MEK1/2 inhibitors.