Background. Dysregulation of the estrogen receptor gene (ESR1) is an established mechanism of inducing endocrine therapy resistance. We previously discovered a chromosomal translocation event generating an estrogen receptor gene fused in-frame to C-terminal sequences of YAP1 (ESR1-YAP1) that contributed to endocrine therapy resistance in estrogen receptor positive (ER+) breast cancer models. This study compares functional, transcriptional, and pharmacological properties of additional ESR1 gene fusion events of both early stage (ESR1-NOP2) late stage (ESR1-YAP1 and ESR1-PCDH11x) breast cancers to gain a better understanding of therapeutic resistance and metastasis. Understanding the role of ESR1 fusions in inducing metastasis is critical, since the primary cause of death in breast cancer patients is through metastasis to distant sites. Methods. RNA-seq screens identified ESR1 fusions from early and late stage, endocrine therapy resistant breast tumor samples. Functional experiments were conducted using ER+ breast cancer cell lines, xenograft, and PDX models to test the ability of ESR1 fusions to induce therapeutic resistance and metastasis. ChIP-seq and RNA-seq were performed to examine transcriptional properties and differential gene expression induced by the fusions which directed subsequent pharmacological experiments with a CDK4/6 inhibitor. Results. ESR1-YAP1 and ESR1-PCDH11x promoted estrogen-independent and fulvestrant-resistant growth in vitro and induced greater tumor growth and increased metastatic capacity to the lungs of xenografted mice. In contrast, the ESR1-NOP2 fusion was sensitive to low estrogen conditions in vitro, and did not promote tumor growth. RNA-seq profiling revealed E2F targets pathway as the most highly enriched pathway induced by the ESR1 fusions. IHC revealed higher levels of pRb in ESR1-YAP1 and ESR1-PCDH11x xenograft tumors and subsequent CDK4/6 inhibition completely blocked tumor growth in an ESR1-YAP1 PDX model. Integrating RNA-seq with ChIP-seq data, we discovered a set of EMT and metastasis genes bound by all ESR1 fusions and WT-ER, but whose expression was strongly and uniquely up-regulated only by the ESR1-YAP1 and ESR1-PCDH11x fusions. These studies also revealed gained sites bound only by the ESR1-YAP1 and ESR1-PCDH11x fusions, not bound by WT-ER nor ESR1-NOP2. Genes mapping to these sites have a role in metastatic biology and were highly up-regulated by the YAP1 and PCDH11x fusions, potentially mediated by long range transcriptional activation. Conclusion. ESR1-YAP1 and ESR1-PCDH11x are driver fusions that occur in drug-resistant, advanced stage breast cancer and are a new class of recurrent somatic mutation that can cause acquired endocrine therapy resistance, yet can be treated with CDK4/6 inhibition. These driver fusions also confer increased metastatic ability through their ability to drive expression of genes that contribute to EMT and metastasis. In contrast, ESR1-NOP2 did not produce functional protein and appears to be a passenger event. These studies may provide pre-clinical rationale for targeting ESR1 translocated breast tumors, since the presence of an ESR1 driver fusion places a patient in a therapeutic category where none of the currently available endocrine therapies are likely to be effective. Citation Format: Lei JT, Shao J, Zhang J, Iglesia M, Chan DW, Cao J, Anurag M, Singh P, Haricharan S, Kavuri SM, Matsunuma R, Schmidt C, Kosaka Y, Crowder R, Hoog J, Phommaly C, Goncalves R, Ramalho S, Rodrigues-Peres RM, Lai W-C, Hampton O, Rogers A, Tobias E, Parikh P, Davies S, Ma C, Suman V, Hunt K, Watson M, Hoadley KA, Thompson A, Perou CM, Creighton CJ, Maher C, Ellis MJ. ESR1 gene fusions drive endocrine therapy resistance and metastasis in 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 PD8-03.

Endocrine therapy resistance occurs in 50% of estrogen receptor positive (ER+) luminal breast cancers but the underlying mechanisms are poorly understood. To gain insight, we have taken advantage of whole-genome and RNA sequencing data of five late-stage hormone-resistant luminal breast tumors all of which have been successfully established as patient-derived mouse xenograft (PDX) models. Here we describe genetic alterations in ESR1-the gene coding for ERa-in three of these five tumors. They include an ESR1 point mutation (Y537S), a gene translocation causing an in-frame fusion between N-terminal ER and C-terminal Yes-associated protein 1 (YAP1), and ESR1 gene amplification. Functional characterization of the ESR1(Y537S) mutant and ESR1-YAP1 fusion in ER+ cell lines indicated that they both possess constitutive transcriptional activity and drive hormone-independent cell proliferation, mirroring the endocrine resistance of the originating tumors and the estradiol-independent growth of the PDX tumors. ESR1 (Y537S), a known gain-of-function experimental mutation in the ligand-binding domain of ER is not seen in primary breast cancer (TCGA data), suggesting it is a mutation associated with acquired resistance. Regarding the ESR1-YAP1 fusion gene, the truncated N-terminal ER fragment lacks the hormone-dependent transactivation domain (AF2) and the ligand binding domain but retains the hormone-independent transactivation domain (AF1) and therefore drives resistance to all endocrine approaches. As for ESR1 gene amplification, the tumor of origin, though resistant to aromatase inhibition, paradoxically responded to estradiol treatment, and this was recapitulated in the PDX model. Interestingly, acquired ESR1 gene amplification also occurred in long-term estrogen-deprived breast cancer MCF-7 cells which similarly regress upon estradiol exposure. Thus, ESR1 amplification may be a biomarker for paradoxical therapy with estradiol. Together, our focused study of advanced endocrine resistant luminal breast tumors revealed three distinct mutational mechanisms affecting the ESR1 gene that drive endocrine therapy resistance. Prevalence studies using RNAseq are underway to determine the frequency of somatic changes in the ESR1 gene in advanced breast cancer samples and additional ER+ PDX models. These additional data will be presented at the meeting. Citation Information: Cancer Res 2013;73(24 Suppl): Abstract nr S3-05.

#6060 Background. Mutations in the alpha catalytic subunit of phosphoinositol-3-kinase (PIK3CA) occur in approximately 30% of ER positive breast cancers but the impact of PIK3CA mutation status on the response to endocrine therapy has not been adequately studied.
 Methods. cDNA and genomic DNA derived from baseline formalin-fixed core biopsy specimens from two neoadjuvant endocrine therapy trials (P024 and the letrozole alone arm of the RAD 2222 trial) were sequenced to detect exon 9 (helical domain - HC) and exon 20 (kinase domain - KD) mutations in PIK3CA. Interactions between mutation status and clinical, pathological and biomarker response to neoadjuvant letrozole and tamoxifen were determined.
 Results. No impact of PIK3CA KD mutation on the efficacy of letrozole or tamoxifen on either clinical, pathological or Ki67 biomarker response to neoadjuvant endocrine therapy was detected in either study. In the 2222 trial, PI3KCA mutation status (KD and HD) was associated with higher levels of pAKT (P=0.01) confirming these mutations activate the PI3 kinase pathway. Despite the lack of an interaction with short term endocrine therapy efficacy endpoints, PIK3CA KD mutation, but not HD mutation, was a favorable prognostic factor for relapse free survival in the P024 trial (RFS log rank P=0.02) and the protective effect was maintained in a Cox proportional hazards model for relapse that included post treatment (surgical), Ki67 and ER and pathological node status and tumor size (HR for no KD mutation, 14, CI:1.9-105 P=0.01). The KD mutation-associated protective effect was particularly dramatic in a tumor subset that exhibited a biomarker profile indicative of endocrine therapy resistance and poor outcome (RFS Log Rank P=0.007 within worst outcome group).
 Conclusion: In agreement with other studies, PIK3CA KD mutations but not HD mutations are associated with a favorable outcome in ER+ breast cancer but the protective effect of KD mutation is apparently independent of the degree of sensitivity to endocrine treatment. An alternative explanation for the protective effect of the presence of a KD domain mutation is that this mutation class reduces the metastatic potential of breast cancer. Pharmacological strategies that plan to specifically target tumors with PIK3CA KD mutations should take this possibility into account. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 6060.

#3063 Background: Endocrine therapy with estrogen deprivation (ED) induces cell cycle arrest in sensitive estrogen receptor positive (ER+) breast cancers, but typically not apoptosis, thus relapse is common despite ED due to residual surviving tumor cells. The phosphoinositide 3-kinase (PI3K) pathway, which promotes cell survival, is hyperactivated in ER+ breast cancer due to mutations of genes in the PI3K signaling cascade, most commonly PIK3CA. We therefore investigated the role of p110α and p110β, the gene products of PIK3CA and PIK3CB, in promoting cell survival in ER+ breast cancer.
 Materials and Methods: RNA interference (RNAi) was used to selectively inhibit p110α and p110β expression in the ER+, PIK3CA mutant MCF-7 and T47D cell lines; the ER+, PIK3CA wild-type but PIK3CB gene amplified HCC712 cell line; and the ER-, PIK3CA and PIK3CB wild-type MDA-MB-231 cell line. Western blotting was used to confirm p110α and p110β knockdown efficiencies and to determine the effects of specific knockdowns on PI3K pathway signaling. Cell line growth was measured using a resazurin reduction assay. Apoptosis was quantified using TUNEL/Hoechst staining. Gene copy number for PIK3CB was assessed by FISH in cell lines and tumor samples.
 Results: PIK3CA RNAi inhibited growth and activated apoptosis in all tested ER+ cell lines under ED, whereas the addition of estrogen reversed PIK3CA RNAi-induced apoptosis. PIK3CB RNAi inhibited proliferation and promoted apoptosis in the absence of estrogen only in the PIK3CB-gene amplified HCC712 cell line. In combination, PIK3CA and PIK3CB gene knockdowns produced a synergistic apoptotic effect on T47D and HCC712 cell lines, but not in MCF-7 cells which contain low levels of p110β protein. In contrast PIK3CA and PIK3CB RNAi did not promote apoptosis in the ER- MDA-MB-231 cell line. Finally, since PIK3CB gene amplification has not been described previously we conducted a screen by FISH and array CGH and detected low-level PIK3CB copy number gain or amplification in ∼5% of clinical samples from a series of breast cancer patients representing diverse clinical phenotypes (ER+, ER-, ERBB2+).
 Conclusions: Inhibition of the PI3K pathway through suppression of PI3K catalytic subunit expression produces a synthetic lethal effect in the absence of estrogen specifically in ER+ breast cancer cells. These data establish a preclinical rationale for combining endocrine therapy with PI3K pharmacological inhibitors in ER+ breast cancer, but suggest that both PIK3CA and PIK3CB are therapeutic targets and gene mutation or amplification in both catalytic subunits may have to be taken into account when identiying a sensitive population for PI3 kinase-directed therapy. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 3063.

Background: We have recently shown that acquired resistance to aromatase inhibitors (AI) can, in some instances, be reversed low-dose estradiol treatment until progression followed by repeat treatment with an aromatase inhibitor (Ellis, M. JAMA 302:774, 2009). We have also demonstrated that simultaneous inhibition of phosphoinositol-3-kinase (PI3K) and ER (though estrogen deprivation) activates cell death through apoptosis, which is reversed by estradiol, revealing a link between ER activation and the efficacy of PI3 kinase inhibition (Crowder, R. Cancer Res 69:3955, 2009). Consistent with this concept we have also shown that long term estrogen deprived (LTED) cells are co-resistant to both classes of agent (SABCS 2009 Abs#3131). We therefore studied ways to overcome resistance to PI3 kinase inhibitors in LTED cell lines using second line endocrine therapy strategies that could be offered to patients with AI resistant advanced disease. Methods: ER+ MCF7 and T47D (PIK3CA mutant) cells were cultured in phenol-red free media containing charcoal-stripped serum for 12 months to create long term estrogen deprivation (LTED) variants that mimicked clinical endocrine therapy resistance. A variant of the MCF7 LTED line that regained estrogen-dependent proliferation was created by re-exposure to estradiol for 4 months (MCF7 LTED-R). The effects of the PI3K isoform inhibitor BKM120, the mTOR inhibitor RAD001 and the dual PI3K/mTOR inhibitor BGT226 in MCF7 LTED and T47D LTED were assessed in combination with estrogen deprivation and fulvestrant. MCF7 LTED-R cells were treated with PI3 kinase pathway inhibitors after acute estradiol withdrawal. Apoptosis was measured using TUNEL flow cytometry. Results: MCF7 LTED were resistant to all three PI3 kinase inhibitors, but combination treatment with fulvestrant restored apoptosis to levels seen with parental MCF7 cells after acute estrogen-deprivation. Acute estrogen deprivation in MCF7 LTED-R cells was also effective in restoring an apoptotic response to BGT226. T47D LTED cells, which, unlike MCF7 cells, have markedly down-regulated ER expression, become more sensitive to PI3 kinase monotherapy than the parental T47D cell line, particularly in response to BKM120, and fulvestrant did not increase the apoptotic response. Conclusions: The MCF7 cell results suggest that PI3 kinase inhibitors will be effective in endocrine therapy resistant advanced breast cancer if these agents are administered with appropriate second line endocrine strategies. Potentially efficacious approaches include fulvestrant combinations and second-line aromatase inhibition, but only after restoration of AI sensitivity with estradiol. The T47D experiments suggest that in the setting of ER+ tumors that lose ER upon disease progression PI3 kinase inhibitors may be very effective. Citation Information: Cancer Res 2010;70(24 Suppl):Abstract nr P2-19-01.