Project 2: Strategies to improve outcomes for triple negative breast cancer patients involving subtype-specific targeted therapies and genomic discovery
The term “triple negative breast cancer” (TNBC) is used to classify 10% - 20% of all breast cancers that lack estrogen receptor (ER) and progesterone receptor expression as well as amplification of the human epidermal growth factor receptor 2 (HER2). TNBC is biologically more aggressive than ER+ disease, with higher rates of relapse in the early stage and decreased overall survival in the metastatic setting. Disease heterogeneity and the absence of well-defined molecular targets have made treatment of TNBC challenging. There is a major need to better understand the molecular basis of this type of breast cancer as well as to develop effective therapeutic strategies against it.
During the current project period, we have been performing a phase II clinical trial (BRE0904) to investigate the combination of paclitaxel, cisplatin ± everolimus (RAD001; mTOR inhibitor) in TNBC patients. The goals are to determine the efficacy of the treatment arms and to predict, based on tumor genomic analysis, patients who will respond to a given treatment. With 115 (of 130 planned) patients accrued to date, 65 have completed the trial and we observe a 40% pathological complete response (path CR) rate. Correlative genomic analyses on the initial tumors revealed that TNBC is a highly heterogeneous disease and subtyping was necessary to better identify effective molecular-based therapies. We leveraged TNBC specimens collected from previous and ongoing Vanderbilt-Ingram Cancer Center (VICC) Breast SPORE–funded clinical trials (VICC: BRE9936, BRE0368, BRE0904) as well as 21 publicly available datasets from eight countries to perform a comprehensive analysis of gene expression (GE) data from 3247 breast cancer tumors. This compilation of worldwide data resulted in a data set of 587 TNBCs and classification of the disease into six subtypes with differing biologies [two basal-like, two mesenchymal-like, an immunomodulatory, and a luminal subtype expressing androgen receptor (LAR)]. We have since validated these using The Cancer Genome Atlas (TCGA) data. In addition, we identified 25 TNBC cell line models representative of these subtypes. Predicted ‘driver’ signaling pathways were pharmacologically targeted in these preclinical models as proof of concept that analysis of distinct GE signatures can inform therapy selection. Cell lines representative of the basal-like subtypes have higher expression of proliferation and DNA Damage Response (DDR) genes and preferentially respond to cisplatin. Mesenchymal subtype cell lines are enriched in growth factor pathway signaling and are sensitive to PI3K/mTOR inhibitors. Cell lines representative of a luminal subtype expressing androgen receptor (LAR) are uniquely sensitive to bicalutamide (AR antagonist), and we discovered that LAR tumors and cells lines have high frequency of PIK3CA mutations.
Preclinical and clinical findings showing high frequency alterations in p53 and PI3K/mTOR signaling pathways lead to the proposed studies that translate preclinical findings to targeted therapies for TNBC patients. We will leverage cell-based models and our ever-growing, clinically annotated and integrative TNBC data set to determine mechanisms of drug response and resistance and to define biomarkers for patient selection. We propose three specific aims to test the following interrelated hypotheses:
In TNBC patients with the LAR subtype (~10%), AR and PI3K signaling synergistically drive tumor growth, and treatment of these patients with an AR antagonist (bicalutamide) in combination with a PI3K pathway inhibitor will be an effective therapy. In the remaining 90% of TNBC patients, the high frequency of p53 mutations and PI3K signaling pathway alterations in their tumors will result in therapeutic vulnerability to the genotoxic agent cisplatin given in combination with a PI3K inhibitor. Comprehensive analysis of well-characterized tumors from patients on hypothesis-driven clinical trials will allow discovery of mechanisms of sensitivity and resistance as well as biomarkers that can be used in the development of new treatment regimens and selection of patients for future trials.
Specific Aim 1: 1: a) To evaluate the efficacy, as measured by clinical benefit rate (CBR) of bicalutamide + the pan-PI3K inhibitor (GDC-0941) in patients with AR+ metastatic TNBC. b) To evaluate the efficacy, as measured by overall response rate (ORR) of cisplatin + GDC-0941 versus cisplatin alone in patients with ARmetastatic TNBC. We will determine if a set of genomic markers can predict sensitivity or resistance to the regimens tested in this phase II clinical trial (BRE1287).
Specific Aim 2: To determine mechanisms of inherent and acquired resistance to cisplatin and PI3K inhibitors in the TNBC setting. Integrating data from synthetic lethal and forward genetic screens with genomic analyses of clinical specimens from BRE0904 and BRE1287, we will identify pathways leading to resistance.
Specific Aim 3: To develop validated clinical biomarkers for TNBC subtyping and use in selection of patients for future clinical trials or new standard treatments. We will leverage comprehensive genomic information from our ever-growing data set of TNBC tumors and additional sets of TNBC tumors for validation of biomarkers.