Abstract SP068: The unique epigenetic state of invasive lobular breast cancers

Although the majority of invasive lobular breast cancers (ILC) are strongly hormone receptor positive and are rarely high-grade, they present a number of unique clinical problems compared with invasive ductal cancer (IDC). These include challenges for early detection, decreased response to chemotherapy and evidence of relative resistance specifically to adjuvant tamoxifen treatment. In addition to uniform loss of e-cadherin, ILCs possess genetic alterations that are suggestive of a unique estrogen receptor axis, including an increase in the frequency of FOXA1 mutations and decrease in GATA3 mutations. In this study we performed a comprehensive analysis of the estrogen receptor axis in pre-clinical models and clinical samples of ILC and IDC. We found that ILC has a unique chromatin state that stems from gained FOXA1 chromatin binding and leads the remodeling of ER chromatin binding. We demonstrated that FOXA1 recruitment can classify primary hormone receptor positive breast cancers to ILC and IDC histological subtypes. Moreover, we show that gained FOXA1 recruitment leads to the transcription of a gene signature associated with resistance to adjuvant tamoxifen in estrogen receptor positive breast cancers and is a mechanism for the divergent pattern of response to specific endocrine treatments in ILC. Collectively, these studies unveil key insights to the fundamental biology unique to ILC, set the stage for precision endocrine treatment in estrogen receptor positive breast cancer, and offer potential treatment targets to improve outcomes for patients with ILC.

Citation Format: R Jeselsohn. The unique epigenetic state of invasive lobular breast cancers [abstract]. In: Proceedings of the 2020 San Antonio Breast Cancer Virtual Symposium; 2020 Dec 8-11; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2021;81(4 Suppl):Abstract nr SP068.

Abstract MS2-2: Targeting the cell cycle - Beyond CDK4/6 inhibition

Over the past several years three CDK4/6 inhibitors were approved for the treatment of metastatic ER+ breast cancer. This class of drug has had a major impact on patient outcomes and more recent analyses showed improvement in overall survival in addition to a significant delay in the need for chemotherapy and increase in progression free survival. Despite the remarkable clinical benefit, there are patients who have primary resistance to CDK4/6 inhibitors and the majority of patients will ultimately develop resistance to this class of drugs. Multiple pre-clinical studies and some clinical data have provided insights to the mechanisms of resistance to CDK4/6 inhibitors guiding the development of clinical trials after progression on a CDK4/6 inhibitor. In this presentation, I will review the mechanisms of resistance to CDK4/6 inhibitors and the ongoing clinical trials investigating novel therapeutic strategies to enhance responses to CDK4/6 inhibitors and for treatment after the acquisition of resistance to CDK4/6 inhibitors.

Citation Format: R Jeselsohn. Targeting the cell cycle - Beyond CDK4/6 inhibition [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr MS2-2.

A phase II feasibility study of palbociclib in combination with adjuvant endocrine therapy for hormone receptor-positive invasive breast carcinoma

BACKGROUND

The CDK4/6 inhibitor palbociclib prolongs progression-free survival in hormone receptor-positive/HER2-negative (HR+/HER2-) metastatic breast cancer when combined with endocrine therapy. This phase II trial was designed to determine the feasibility of adjuvant palbociclib and endocrine therapy for early breast cancer.

PATIENTS AND METHODS

Eligible patients with HR+/HER2- stage II-III breast cancer received 2 years of palbociclib at 125 mg daily, 3 weeks on/1 week off, with endocrine therapy. The primary end point was discontinuation from palbociclib due to toxicity, non-adherence, or events related to tolerability. A discontinuation rate of 48% or higher would indicate the treatment duration of 2 years was not feasible, and was evaluated under a binomial test using a one-sided α = 0.025.

RESULTS

Overall, 162 patients initiated palbociclib; over half had stage III disease (52%) and most received prior chemotherapy (80%). A total of 102 patients (63%) completed 2 years of palbociclib; 50 patients discontinued early for protocol-related reasons (31%, 95% CI 24% to 39%, P = 0.001), and 10 discontinued due to protocol-unrelated reasons. The cumulative incidence of protocol-related discontinuation was 21% (95% CI 14% to 27%) at 12 months from start of treatment. Rates of palbociclib-related toxicity were congruent with the metastatic experience, and there were no cases of febrile neutropenia. Ninety-one patients (56%) required at least one dose reduction.

CONCLUSION

Adjuvant palbociclib is feasible in early breast cancer, with a high proportion of patients able to complete 2 years of therapy. The safety profile in the adjuvant setting mirrors that observed in metastatic disease, with approximately half of the patients requiring dose-modification. As extended duration adjuvant palbociclib appears feasible and tolerable for most patients, randomized phase III trials are evaluating clinical benefit in this population.

CLINICALTRIALS.GOV REGISTRATION

NCT02040857.

Abstract ES10-1: Recurrent ESR1 mutations activating mutations in breast cancer

Endocrine treatments are the cornerstone treatments in estrogen receptor positive (ER+) breast cancer. Despite their efficacy, resistance to endocrine treatments remains an important clinical challenge. A number of years ago, recurrent estrogen receptor ligand binding domain (LBD) mutations were identified as the most common genomic mechanism of acquired resistance to endocrine treatment in metastatic breast cancer. These mutations were found in about 20% of ER+ metastatic tissue specimens and in more than 30% of patients with metastatic ER+ breast cancer when testing circulating tumor DNA. Pre-clinical studies showed that these mutations lead to constitutional ligand independent transcriptional activity and tumor growth. Recently, we showed that the ER mutations have neomorphic properties that promote metastases. Clinical data, derived mainly from retrospective analyses of clinical trials, indicate that the ER mutations confer resistance to aromatase inhibitors and fulvestrant, and decreased overall survival. A recent study also showed that the Y537S LBD ER mutation is acquired during fulvestrant and palbociclib treatment. Collectively, the ER mutations are important drivers of resistance standard endocrine treatments and confer worse outcomes in ER+ metastatic breast cancer. Clinical trials stratifying patients by ER mutant status or dedicated to patients with the ER mutations are needed to identify therapeutic strategies to target these mutations.

Citation Format: Jeselsohn R. Recurrent ESR1 mutations activating mutations in breast cancer [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr ES10-1.

Abstract P4-04-03: Hyperactive FOXA1 activates super-enhancer-engaged HIF2α/EPAS1 to promote endocrine-resistant metastatic ER-positive breast cancer

Background: We have recently reported that acquired endocrine resistance (Endo-R) in multiple ER+ breast cancer (BC) Endo-R cell models is driven by high levels of FOXA1 (High-FOXA1), via gene amplification and/or overexpression (OE), leading to coordinated reprogramming of the FOXA1 genomic binding (cistrome) and transcriptome. Forced FOXA1 OE in parental (P) cells induced similar transcriptional reprogramming leading to Endo-R and metastasis. Recent clinical data showing enrichment of FOXA1 amplification in ER+ metastases further support the clinical importance of our findings. However, the molecular components and the mechanism of High-FOXA1-induced transcriptional reprogramming in Endo-R and metastasis are unknown.

Methods: High-FOXA1-containing MCF7 tamoxifen-resistant (TamR) and P/FOXA1-OE cells were used in this study. An integrative multi-OMICS approach was employed to analyze transcriptome (RNA-seq), FOXA1 cistrome, and histone H3K27 acetylation (ac) (ChIP-seq). Intersection of High-FOXA1-induced transcriptome and distinct FOXA1 cistrome-predicted genes defined a High-FOXA1 core gene signature (CGS). Gene Set Enrichment Analysis (GSEA) and Gene Ontology (GO) were used for functional annotation. Cell growth and migration/invasion were measured by a bright-field automated cell counter and Transwell insert system. Altered gene expression was measured by RT-qPCR. High-FOXA1 signaling inhibition included gene knockdown (siRNA) or pharmacologic blockade (the EPAS1 inhibitor PT2385). The predictive role of EPAS1 and the associated gene signature were analyzed using publicly available BC datasets.

Results: FOXA1 OE reprogrammed the FOXA1 cistrome in P cells to resemble that of the TamR cells. The FOXA1 cistrome was significantly correlated with the deposition of H3K27ac in TamR vs. P cells (P<2.2e-16). Similarly, the differentially expressed genes in TamR vs. P cells were enriched for FOXA1 binding at enhancers demarcated by H3K27ac (P=8e-125). The FOXA1-CGS was linked to multiple metastasis-related GO terms including “hypoxia response”, enriched for the cancer secretome gene set (P=4.1e-16), and highly represented in the Endo-R transcriptome across our multiple cell models (MCF7, 600MPE, and CAMA1) (P<0.01). Integrative analysis of H3K27ac-defined super-enhancers (SEs) and altered cistrome/transcriptome upon High-FOXA1 nominated EPAS1, a hypoxia-inducible transcription factor (TF), as a top candidate of SE-activated TFs amplifying High-FOXA1 signaling. EPAS1 blockade markedly repressed the secretome genes (e.g., IL8 and S100P) and cell migration and invasion in TamR cells. Primary ER+ tumors (TCGA) with high EPAS1 are enriched for a cancer secretome gene set (P=3e-4). High EPAS1 predicts poor distant metastasis-free survival in ER+ BC treated with endocrine therapy (P=.034).

Conclusions: High-FOXA1 induces transcriptional reprogramming by coordinating histone enhancer marks to activate EPAS1 via an SE mechanism, which in turn mediates transcriptional reprogramming, partly via inducing a pro-metastatic secretome, to promote Endo-R and metastasis. Targeting the High-FOXA1/EPAS1 axis to block transcriptional reprogramming may offer a new therapeutic strategy to prevent and treat Endo-R metastatic ER+ BC.

Citation Format: Fu X, Pereira R, De Angelis C, Veeraraghavan J, Shea MJ, Nanda S, Feng Q, Jeselsohn R, O'Malley BW, Brown M, Osborne CK, Schiff R. Hyperactive FOXA1 activates super-enhancer-engaged HIF2α/EPAS1 to promote endocrine-resistant metastatic ER-positive breast cancer [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P4-04-03.

Abstract P4-03-05: AP-1 as a potential mediator of resistance to the cyclin-dependent kinase (CDK) 4/6-inhibitor palbociclib in ER-positive endocrine-resistant breast cancer

Background: The CDK4/6-inhibitor palbociclib (Palbo) in combination with endocrine therapy (ET) substantially improves progression-free survival compared to ET alone. However, almost all initial responders eventually develop resistance and relapse. Delineating the early adaptive signaling and the mechanisms underlying resistance to CDK4/6 inhibition is therefore crucial to identify new biomarkers and therapeutic targets to enhance the efficacy of Palbo and improve patient outcome.

Materials and Methods: MCF7 parental (P) cells and derivative lines made resistant (R) to tamoxifen (TamR), estrogen deprivation (EDR), or fulvestrant (FulR) were used. The MCF7P line and its endocrine-R (EndoR) derivatives were exposed to increasing concentrations of Palbo to generate acquired Palbo-R (PDR) models. The proteomic/signaling profiles of P and EndoR cells upon short-term Palbo treatment and as PDR develops were determined using reverse-phase protein arrays (RPPA). Transcriptional activity of the activator protein-1 (AP-1) transcription factor (TF) was measured by luciferase reporter assay. Global AP-1 blockade was achieved using a pINDUCER system to express doxycycline (Dox)-inducible dominant-negative (DN) c-Jun that lacks the transcriptional activation domain. Cell growth and colony formation were assessed using methylene blue staining and clonogenic assays, respectively. Levels of phosphorylated (p)-RB and CDK2 were assessed by Western Blot.

Results: In P and all EndoR cell models, Palbo inhibited cell growth and colony formation in a dose-dependent manner, though the inhibitory effect was greater in the EndoR cells compared to P cells [IC50 value of P cells >3 times that of EndoR lines (p<0.001); clonogenic % inhibition at 100nM = 54 in P and >85 in EndoR lines (p<0.001)]. Across the P and all EndoR models, short-term Palbo treatment resulted in increased levels of several membrane and intracellular signaling molecules, TFs, and enzymes. Among these, the AP-1 TF components c-Jun and p-c-Jun showed the highest increase across all models, with the utmost change observed in the TamR model (Fold-change = 4.4, 4.0 for total and p-c-Jun, respectively). Since we also observed that acquired resistance to Palbo in the TamR model was associated with higher AP-1 transcriptional activity and increased total and p-c-Fos levels, we assessed the efficacy of combining Palbo with AP-1 blockade in the TamR model. In two independent TamR clones ectopically expressing inducible DN-c-Jun, AP-1 blockade (+Dox) in combination with Palbo was highly effective in inhibiting cell growth and reducing p-RB and CDK2 levels compared to single agent treatments. In addition, in both the TamR/DN-c-Jun-expressing clones, the combination of Palbo, AP-1 blockade, and fulvestrant resulted in cell death and a significantly greater cell growth inhibition compared to any dual or mono treatments.

Conclusion: Our results suggest activation of AP-1 as a potential mechanism of resistance to Palbo in ER+ EndoR models. Transcriptomic profiling of the Palbo-sensitive and R cells, currently underway, will provide an in-depth understanding of the role of AP-1 as well as other key targets and associated molecular mechanisms in Palbo resistance.

Citation Format: De Angelis C, Nardone A, Cataldo ML, Veeraraghavan J, Fu X, Giuliano M, Malorni L, Jeselsohn R, Osborne KC, Schiff R. AP-1 as a potential mediator of resistance to the cyclin-dependent kinase (CDK) 4/6-inhibitor palbociclib in ER-positive endocrine-resistant breast cancer [abstract]. In: Proceedings of the 2017 San Antonio Breast Cancer Symposium; 2017 Dec 5-9; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2018;78(4 Suppl):Abstract nr P4-03-05.

Abstract PD2-04: FOXA1 induces a pro-metastatic secretome through ER-dependent and independent transcriptional reprogramming in endocrine-resistant breast cancer

Background: Metastasis in ER-positive (+) breast cancer (BC) occurring years to decades after initial diagnosis presents a daunting challenge for clinical care and preclinical research due to limited known key players and experimental models. FOXA1 is a pioneer factor for ER-chromatin binding and function, and is highly expressed in ER+ BC metastases, yet the underlying mechanism is unclear. Tumor-secreted proteins play a crucial role in the reciprocal interplay between cancer cells and host microenvironmental factors at both primary and secondary sites. We hypothesized that high FOXA1 provokes an ER-dependent transcriptional program that includes a unique pro-tumorigenic secretome essential for promoting ER+ BC metastasis. Methods: A lentiviral doxycycline (Dox)-inducible FOXA1 overexpression vector and a dual luciferase/GFP (LG) tracking vector were integrated to construct a stable MCF7-LG/FOXA1 cell model. Ovariectomized nude mice bearing MCF7-LG/FOXA1 xenografts in the presence of exogenous estrogen (E2) were randomized to ± Dox, each with continued E2, E2 deprivation (ED), or tamoxifen (Tam). Survival surgery removing the therapy-naïve (E2 arm) and relapsed (ED/Tam arms) tumors was performed when tumors reached ∼1000 mm3. All mice then received ED/Tam 'adjuvant' therapy, with longitudinal luminescence imaging to monitor local/distant recurrences. Mice were or will be euthanized at the ethical end-point. Integrative bioinformatics was performed using RNA-seq and FOXA1/ER ChIP-seq data from our preclinical models to identify secretome targets for functional intervention. Times to tumor regression (TTR) and progression (TTP) were defined by when the tumor reached half or twice the volume at randomization. Results: Median (m) TTR was achieved in ED (31/34 days, -/+Dox, P = 0.184) but not in Tam groups — Tam delayed tumor growth but failed to prevent progression in all mice with mTTP of 94/93 days (-/+Dox, P = 0.517). Despite no difference in mTTP at Tam-/+Dox, a quarter of +Dox tumors (3/12) had volume doubled by day 11. No metastases were observed by imaging in any of the mice before surgery ('neoadjuvant' setting). Local relapse and lymph-node/lung metastases were detected after surgery ('adjuvant' setting). At day 90 in the adjuvant Tam group with previously relapsed tumors, +Dox mice succumbed to metastasis more often than -Dox mice (7/8 vs. 3/10, P = 0.023). Compared to the adjuvant Tam+Dox mice with previous therapy-naïve tumors, the Tam+Dox with previously relapsed tumors showed higher distant metastasis rate (7/8 vs. 5/14, P = 0.026). Analysis of the ED setting is pending due to late recurrence. Data integration and functional study revealed a set of cytokines, growth factors, and extracellular matrix components (including IL-8, CTGF, and LOX), regulated by FOXA1 often in conjunction with ER, that are highly involved in FOXA1-induced metastasis. Global secretome profiling by mass spectrometry and target validation are ongoing. Conclusions: FOXA1 overexpression increases metastatic potential in ER+ BC. We established a pertinent metastatic xenograft mouse model to characterize a pro-metastatic secretome with diagnostic and therapeutic potential for treating metastatic ER+ BC.

Citation Format: Fu X, Pereira R, Zhao D, Jung SY, Jeselsohn R, Creighton CJ, Shea M, Nardone A, Angelis CD, Tsimelzon A, Wang T, Gutierrez C, Huang S, Edwards DP, Rimawi MF, Hilsenbeck SG, Brown M, Chen K, Osborne CK, Schiff R. FOXA1 induces a pro-metastatic secretome through ER-dependent and independent transcriptional reprogramming in endocrine-resistant 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 PD2-04.

Abstract S5-05: Impact of pre-operative exercise on breast cancer gene expression

Background: Exercise is linked to a lower risk of developing and dying from breast cancer, but the biological mechanisms through which exercise could impact breast cancer are unclear. In animal models, exercise impacts tumor formation and progression, but there are few data regarding direct effects of exercise on tumor tissue in humans. The Pre-Operative Health and Body (PreHAB) Study was a randomized window of opportunity trial designed to explore the impact of exercise on molecular pathways in women with breast cancer.

Methods: Inactive women with Stage I-III breast cancer were enrolled through Dana-Farber Cancer Institute and Yale University prior to surgery. Participants were randomized 1:1 to an aerobic and strength training exercise intervention or mind body control intervention and participated in the interventions between enrollment and the time of surgery. Tumor tissue was collected at enrollment and surgery; samples were reviewed by a breast pathologist and were macrodissected to include sections of tumor with at least 10% cellularity. Capture RNA-sequencing of the transcriptome coding regions was performed using the Illumina Truseq RNA access platform.

Results: 49 women were randomized (27 exercise and 22 control). At baseline, mean age was 52.6, BMI was 30.2kg/m2 and exercise was 49 min/wk. Mean time between enrollment and surgery was 4.2 weeks. Participants in the exercise arm significantly increased exercise vs. controls (increase of 203 vs. 23 min/wk, p<0.0001). Transcriptomic analysis was performed on the tumors from the pre and post intervention biopsies from 32 patients (16 exercise and 16 control). Quality Control analysis of the RNA-sequencing data showed an average read depth of 25 million reads per sample, mapping ∼79% to exonic regions. Principal Component Analysis revealed no read bias or batch effects and unsupervised clustering showed that pre- and post-operative samples clustered together by patient. Differential gene expression analysis by DEseq2 revealed a limited number of individual genes with significant changes after the intervention. KEGG pathway analysis, however, of 214 KEGG pathways using the bioconductor package GAGE (Generally Applicable Gene-Set Enrichment for Pathway Analysis) demonstrated upregulation of 13 unique pathways between the baseline biopsy and surgical excision in exercise participants and none in mind body participants (q<0.1). The top ranked upregulated pathway was cytokine-cytokine receptor interactions (q=6.93E-05, set size=238 genes). Il6, CCL3 and other cytokines are among the genes upregulated in this pathway. Analysis also demonstrated downregulation of 13 unique pathways (q<0.1) including cell cycle, RNA transport and DNA replication pathways, in exercise participants over the intervention period.

Conclusions: A pre-operative exercise intervention led to alterations in gene expression in tumor tissue in women with breast cancer. Validation in additional data sets and an analysis of which cellular compartments within the tumor are responsible for the changes is needed. These findings demonstrate that exercise may have a direct effect on breast tumor tissue in humans, providing new insights into the biologic mechanisms through which exercise could lower the risk of developing and dying from breast cancer.

Citation Format: Ligibel JA, Irwin M, Dillon D, Barry W, Giobbie-Hurder A, Frank E, Winer EP, McTiernan A, Cornwell M, Pun M, Brown M, Jeselsohn R. Impact of pre-operative exercise on breast cancer gene expression [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 S5-05.

Abstract P3-05-09: Exploring bazedoxifene and palbociclib as potential therapeutic strategies for overcoming ESR1-mediated endocrine resistance

INTRODUCTION

Despite effective endocrine treatments, endocrine resistance remains a major clinical challenge. We and other groups have recently detected ligand-binding domain ESR1 mutations in metastatic estrogen receptor positive ( ER+) breast cancers. Our preclinical studies showed that these mutations confer constitutive activity and relative resistance to tamoxifen and fulvestrant. In this study we sought to investigate therapeutic strategies to overcome resistance rendered by the ESR1 mutations. Since our previous studies showed relative resistance to tamoxifen or fulvestrant, we hypothesized that bazedoxifene, a high affinity third generation SERM/SERD could overcome resistance driven by the ER mutant resistance. Additionally, we hypothesized that inhibiting cyclin D1, a key ER transcriptional target gene and cell cycle regulator, is a second potential therapeutic strategy to circumvent resistance rendered by mutant ER. Therefore in this study we tested the effects of bazedoxifene, palbciclib and their combination on cell proliferation in the presence and absence of the ESR1 mutations.

METHODS

For this study we established doxycycline inducible MCF7 cell lines expressing the ER-LBD mutations (Y537S, Y537N and D538G) and WT-ER as control. Cell proliferation response to bazedoxifene, tamoxifen, fulvestrant, palbociclib and the bazedoxifene-palbociclib combination was evaluated.

RESULTS

Cells harboring mutant ER were relatively resistant to tamoxifen and fulvestrant, as expected, and remained sensitive to single agent bazedoxifene and palbociclib. The combination of bazedoxifene and palbociclib was found to be superior to the single agents and exhibits synergistic activity.

CONCLUSION

The combination of bazedoxifene and palbociclib inhibits mutant ER cell growth and other cell models of endocrine resistance and is a potential therapeutic combination.

Citation Format: Nguyen M, Buchwalter G, Luo F, Garraway L, Brown M, Jeselsohn R. Exploring bazedoxifene and palbociclib as potential therapeutic strategies for overcoming ESR1-mediated endocrine resistance. [abstract]. In: Proceedings of the Thirty-Eighth Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2015 Dec 8-12; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2016;76(4 Suppl):Abstract nr P3-05-09.

P4-01-18: AP-1 Blockade Potentiates the Anti-Tumor Effect of Endocrine Treatment and Reverts the Resistant Phenotype in Hormone Receptor-Positive Breast Cancer

Background: Resistance to endocrine therapy is a major clinical issue. The transcription factor AP-1 is a key regulator of cell growth and survival as well as a downstream signaling component of several pathways deregulated in endocrine-resistant breast cancer. We have previously shown that acquired endocrine resistance is associated with increased AP-1 activity. AP-1 has also been shown to interact with and modulate the ER network and transcriptional program, especially under hyperactive growth factor signaling, which is commonly associated with endocrine resistance. We hypothesized that interfering with AP-1 function would circumvent endocrine resistance possibly due to its role in modulating ER transcriptional activity.

Methods and results: We inhibited AP-1 function by a genetic approach. We used two different MCF7 clones stably transfected with a Doxycycline (Dox)-inducible dominant-negative (DN) c-Jun (MCF7/Tet-Off Tam67 clones 62 and 67) and two vector-alone control MCF7 clones. Xenografts of these clones were established in ovariectomized nude mice supplemented with estrogen (E2). Mice were then randomized to continued E2 supplementation (control) or to endocrine therapy with either estrogen deprivation (ED) or tamoxifen (Tam), all in the presence or absence of Dox to induce the DN c-Jun expression. AP-1 blockade in both MCF7/Tet-Off Tam67 clones significantly enhanced sensitivity to Tam by reducing time to tumor size halving (p=.014 and p=.006 for clone 62 and 67, respectively) and time to complete tumor disappearance (p=.001 and p=.0034 for clone 62 and 67, respectively). Similar results were obtained with ED treatment. In addition, AP-1 blockade significantly delayed the onset of Tam resistance by increasing time to tumor size doubling (p=.0028). Furthermore, induction of DN c-Jun resulted in a dramatic shrinkage of growing tumors after long-term Tam treatment, suggesting reversal of endocrine resistance with AP-1 blockade. None of the above effects was observed in control clones upon Dox removal. Interestingly, no significant effect of AP-1 blockade was observed on E2-stimulated tumor growth. IHC analysis showed that AP-1 blockade induced tumor response by reducing proliferation (i.e., decreased % of Ki67- and phospho-Histone 3-positive cells) and by inducing apoptosis (i.e., increased % of cleaved caspase 3/7-positive cells). Bioinformatic analyses were conducted to intersect our MCF7 xenograft/Tam-resistant gene signature and the datasets of genes associated with ER DNA-binding sites obtained by whole-genome ER cistromic analysis under estrogen or epidermal growth factor (EGF) stimulation of MCF7 cells. A significant enrichment of the genes associated with the EGF-unique ER DNA-binding sites was observed within our Tam-resistant signature (p<2E-16). Remarkably, 90% of these DNA binding sites harbored an AP-1 motif.

Conclusions: We show that AP-1 blockade increases tumor sensitivity and circumvents resistance to endocrine therapy, thus warranting the development of AP-1-targeted therapy to improve endocrine treatment outcomes. Overall, we suggest that AP-1 is critical in induction of a switch in the ER transcriptional program and may be a new hallmark of endocrine resistance.

Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr P4-01-18.