Seattle Cancer Consortium Breast SPORE
Project Co-Leaders: Bruce Clurman, MD, and V.K. Gadi, MD
Co-Investigator: Peggy Porter, MD
The finding that mutations in genes that regulate the cell cycle are among the most common genetic changes in breast cancer cells led to the expectation that cell cycle-directed therapy would radically change therapy and outcomes. However, while this promise is now being realized with the use of CDK4/6 inhibitors in breast cancer therapy, CDK2, which mediates many oncogenic signaling pathways in breast cancer, has been more recalcitrant to this approach. Our work is focused to new ways to target CDK2 activity to enhance breast cancer therapy, through both p27 modulation, and through abnormal inhibitory phosphorylation. If successful, these approaches may be rapidly translated to new breast cancer treatment strategies. Our original Project 1, led by Dr. James Roberts, aimed to determine if cytoplasmic p27 acts as an oncogene and to evaluate the relationship between p27 cytoplasmic mislocalization and recurrence/survival from breast cancer or response to breast cancer therapy. Dr. Bruce Clurman was selected to lead the project in 2014 when Roberts left Fred Hutch. Clurman’s leadership of the project and the development of additional data and insights into CDK control mechanisms, led to new translational opportunities and to the current proposal to therapeutically exploit CDK2 activity in breast cancers.
Project Co-Leaders: Stanley Riddell, MD, Lupe Salazar, MD, and Jennifer Specht, MD
Co-Investigators: Shelly Heimfeld, PhD, and Ted Gooley, PhD
Project 2 builds on more than a decade of research in cancer adoptive cell therapy (ACT) at our center and on promising new evidence that immunotherapy might improve outcomes in breast cancer. The translational goal of this SPORE project is to evaluate the adoptive transfer of tumor-specific T cells derived or engineered from defined subsets of T cells to treat breast cancer. The immune system is designed to distinguish diseased from normal cells with exquisite specificity and sensitivity, and there is increasing evidence that tumor development and progression is restrained by adaptive host T cell responses to tumor-associated antigens. However, harnessing endogenous T cells to provide therapeutic benefit in breast cancer requires identifying antigens that are expressed by tumor cells and can be safely targeted, and developing methods to achieve potent and durable T cell immunity in patients. Many candidate tumor associated antigens have been discovered in breast cancer and we have focused on using genetic engineering of T cells to target the tyrosine kinase-like orphan receptor ROR1, which is highly expressed at the cell-surface of breast cancers, particularly triple negative breast cancer (TNBC) and can serve as a target for ACT. In other epithelial malignancies, ROR1 mRNA expression has been associated with a poor prognosis and a metastatic phenotype, suggesting that this molecule may contribute to resistance to conventional therapies. In studies developed as part of the current SPORE project and with SPORE developmental project funding, the Riddell lab designed and optimized ROR1-specific chimeric antigen receptors (CARs) that when expressed in primary human T cells have potent antitumor activity in vitro and in pre-clinical tumor models studies against ROR1+ breast cancer cells. Safety studies have been performed in non-human primates, a clinical grade lentiviral vector produced, and an investigational new drug application for a clinical trial of ROR1 CAR-T cells has been prepared for submission to the FDA. In collaboration with Dr. Specht in the clinical core, a first-in-human clinical trial of ACT with ROR1 CAR-T cells will be conducted in triple negative breast cancer, a breast tumor subtype that shows a high rate of ROR1 expression.
Project Co-Leaders: David Hockenbery, MD, and Jennifer Specht, MD
Co-Investigators: Lara Gamble, PhD, Li Hsu, PhD, PhD, Fionnuala Morrish, PhD, Savannah Partridge, PhD, Andrew Shields, MD, Ted Gooley PhD
The use of systemic chemotherapy for breast cancer has contributed to the recent decline in breast cancer mortality; however, an unacceptable number of patients fail systemic therapy and die of disseminated disease. Identifying factors important in resistance and directing patients towards more effective treatment is the translational goal of Project 3. Using quantitative PET imaging to measure glucose metabolism, and more recently dynamic contrast-enhanced (DCE) MRI to measure blood flow, we have identified an in vivo metabolic signature for locally advanced breast cancer (LABC) resistant to neoadjuvant chemotherapy as (1) a pre-therapy mismatch between metabolism and perfusion, (2) persistent or even increased tumor perfusion despite treatment, and (3) an altered pattern of glucose metabolism relative to glucose delivery after treatment. This pattern predicts incomplete response, early relapse and death independent of established prognostic factors, including pathologic primary tumor and nodal pathologic response. We have also found this pattern is more profoundly associated with triple-negative (TN, ER/PR/HER2 negative) tumors versus those that express ER/PR and/or over-express HER2. The Project’s ongoing clinical trial, “Quantitative Dynamic PET and MRI and Breast Cancer Therapy," includes pre- and post-therapy fluorodeoxyglucose (FDG) PET and MR imaging with concurrent biopsies on patients with locally advanced breast cancer. Data collected include quantitative and serial FDG PET and DCE-MRI at baseline, mid-therapy and post-therapy; gene expression profiles and tumor intrinsic subtypes from tissue specimens assessed by microarray; tumor histology and grade; mitotic count; protein expression of ER, PR, Her2, Ki-67, EGFR, CK 5/6, CAIX, CD68 and CC3 and HIF-1α; detailed clinical data; and measurement of cell metabolites by ToF-SIMS, a novel proteomic assay performed on prepared frozen tissue specimens. Analyses are underway that will help identify breast cancer patients likely to fail systemic chemotherapy and direct therapy towards those biologic targets most likely to overcome resistance.
Project Co-Leaders: Christopher Li, MD, PhD, and Peggy L. Porter, MD
Co-Investigators: Li Hsu, PhD, and Kathleen Malone, PhD
The three most common molecular subtypes of breast cancer, luminal A (40%), luminal B (30%), and basal-like (15%) vary considerably in their prognosis. There are currently no tools for predicting prognosis among patients with luminal B and basal-like breast cancer. These patients have comparatively high risks of recurrence and poor survival rates. A recent study with lengthy follow-up observed that the 10-year survival rate for women with luminal A tumors was 70%, but only 54% and 53% for women with luminal B and basal-like disease, respectively. Oncotype DX and MammaPrint testing have demonstrated the potential for molecular tumor characteristics to be clinically useful means of predicting risk of recurrence and guiding treatment decisions. However, at present Oncotype DX is only useful for women with hormone receptor-positive/HER2-negative breast cancers, which are primarily comprised of lower risk, less aggressive luminal A tumors; and MammaPrint is not specific to any particular molecular subtype of breast cancer. Thus, there is a clear clinical need for tools specifically tailored to identifying which patients with the more aggressive subtypes of breast cancer, luminal B and basal-like, have a high risk of recurrence and may benefit from more aggressive treatment and follow-up. There is also considerable value in identifying the substantial proportion of these patients who have a low risk of recurrence and could potentially avoid certain treatments.
Our research team is uniquely positioned to address this need for new data and this project was chosen in 2014 to replace the original Project 4. Using a population-based strategy that builds on the substantial number of women with breast cancer enrolled in completed and ongoing studies and preliminary gene expression profiling data that demonstrates marked differences in the sets of molecular markers that are related to risk of recurrence in estrogen receptor (ER) positive vs. negative disease, the team is working to discover and validate molecular predictors of recurrence and breast cancer specific mortality among patients with luminal B and basal-like breast cancers. Drs. Li and Malone have enrolled more than 2000 patients with ER+ breast cancer (of whom ~800 are expected to be luminal B) and 1000 patients with triple-negative breast cancer (of whom ~700 are expected to be basal-like) having obtained detailed risk factor data and treatment information as well as tumor tissue specimens. The project was not selected to go forward in the next grant cycle. However, the size, depth of available information, and collection of biospecimens from this set of patients makes it uniquely suited for discovery of this type and the investigators are seeking non-SPORE funding to complete this important research.
Project Co-Leaders: Amanda Paulovich, PhD, Kathleen Malone , PhD
The aims of the original Project 4 were to 1) characterize the DNA damage response (DDR) of primary human mammary epithelial cells, 2) develop a multi-analyte panel of reporters that en masse measures activity of multiple components of the signal transduction pathway in breast cancers, 3) Test the multi-analyte panel for its predictive and prognostic capability (relative to known clinico-pathologic features) in a well-characterized population cohort. A protein analyte panel including 53BP1, ATM pS1981, H2AXpS139, Smc1(pan), Smc1pS966, 53BP1pS25, BRCA1 (clone MS110), CHK1pS317, CHK2 pT68, DNA-PKpS2056, Nbs1pS343/Nibrin, P53pS15, P53 pan (pAb 1801), Smc1S957 , Rad17pS645, TD p27, Rad17(pan) was optimized and tested in the Pathology core on a triage TMA of more than 150 tumors with known clinical outcome representing all tumor subtypes. The markers most strongly associated with breast cancer mortality (BCM) in Cox proportional-hazards models after adjustment for age, stage, and subtype were: CHK2(pT68), SMC1(pS957), SMC1(pan). For Luminal B tumors, SMC1(pan) (HR=3.9, 95% Confidence Interval (CI)=1.1-13.6) and for Triple Negative tumors, CHK2(pT68)(HR=11.0 CI=1.4-85.7). Although these associations showed promise, it was not clear that they would produce a stronger prognostic biomarker panel than those currently available. The majority of protein changes associated with IR-induced DNA damage were in the phosphoproteome, not the unmodified proteome. This suggested that assay methods other than the planned IHC would be better suited for assay development. Dr. Paulovich’s work with the international NCI Clinical Proteomic Tumor Analysis Consortium (CPTAC) and her development of a targeted Multiple Reaction Monitoring Mass Spectrometry (MRM-MS) assay (selected as “Method of the Year” by Nature Methods58) that enables precise and highly specific quantification of proteins that can be highly multiplexed, standardized, reproduced, and shared across laboratories and instrument platforms, is the methodology that will be used to revisit this panel development in the future. Dr. Paulovich also has DRP funding to test the feasibility of using her immuno-MRM (I-MRM-MS) technology59, which combines peptide immunoaffinity enrichment with MRM for quantifying low-abundance analytes, to address the unmet clinical need for accurate hormone receptor determination in breast cancer bone metastasis biopsies.