Introduction

The ER signaling is a well-established addictive oncogenic pathway in breast cancer cells. Approximately, 70% of breast cancers are classed as ER-positive breast cancers [1,2,3]. Endocrine therapies (ETs) targeting estrogen action have dramatically decreased mortality from breast cancer. Typical ETs that are currently used worldwide include selective estrogen receptor down-regulators (e.g., fulvestrant); selective estrogen receptor modulators (e.g., tamoxifen); and aromatase inhibitors (AIs, e.g., letrozole). However, their efficacy is limited by intrinsic and acquired endocrine resistance [4]. For example, one-third of patients exposed to tamoxifen will eventually relapse with endocrine-resistant advanced tumors within 15 years [1]. After 5 years of scheduled endocrine therapy, the breast cancer recurrences emerged at a stable rate throughout the study period from 5 to 20 years [5]. In a meta-analysis of outcomes of randomized trials of AIs compared with tamoxifen in breast tumors, the 5-year recurrence rate of tamoxifen was 12.6% and that of AIs was 9.6% [6]. Recently, in 692 cases of hormone receptor-positive (HR+) post-endocrine therapy breast cancers, clinical genomic analysis results provided perspicacity in the genomic aberrations present in advanced HR+/HER2− breast cancers [7]. Endocrine-resistant breast tumors were divided into four groups according to the genomic alterations, 18% of the tumors bearing ESR1 alterations, 13% of tumors harboring lesions in the mitogen-activated protein kinase (MAPK) pathway, 9% of the tumors with mutations in the transcriptional factors, and 60% of the endocrine-resistant cases with undiscovered mechanisms [7].

Several well-described mechanisms of ET resistance are focused on the dysregulation of the ER pathway itself. The major form of estrogen receptor in breast cancer is ERα. ERα contains a central DNA-binding domain flanked by two activation domains [8]. The AF-2 domain is located in the ligand-binding domain and its activity is estrogen dependent; moreover, the AF-2 domain interacts with co-activators which promote ERα transcriptional activity [8, 9], whereas AF-1 is activated by phosphorylation of ER [10]. ERα, a transcription factor-dominating gene correlated with cell proliferation, plays key regulatory roles during mammary gland development and in breast carcinogenesis. Blockade of ER signaling function is the mainstay therapeutics for ER+/HER2− breast cancer. But the occurrence of endocrine resistance is inevitable with advanced breast cancer. Given that endocrine therapy can be well tolerated with durations of response extending into years, the clinical benefit rate declines to around 30% for second- or third-line ETs for patients who benefited from the first-line treatment [11]. Therefore, the challenge is to improve our understanding of the mechanisms of ET resistance and to develop therapeutic strategies to extend the duration of effective therapy while minimizing toxicity. In this review, we discuss both the genetic and epigenetic rationale of ET resistance in HR+/HER2− advanced breast cancer, and highlight the striking success of the treatment developed to overcome the specific resistance mechanism.

ESR1 mutations

Dysregulation of various components of the ER signaling associated with endocrine resistance includes dysregulation of ESR1 expression [12, 13], ESR1 mutations, expression of truncated isoforms of ERα [14,15,16], post-translational modification of ERα [17,18,19,20,21,22], abnormality of differential recruitment of ER co-regulators [23, 24] and downstream regulations of ER target genes via receptor tyrosine signaling and other signaling pathways [25,26,27,28]. Much literature exist detailing these mechanisms [29, 30]. The primary mechanism of endocrine  resistance in breast cancer is lack of expression of ER, however, two studies showed that less than 10% of the cases appeared to show loss of the ER expression [31, 32], therefore, the ER pathway functions in the majority of cases of ETs resistance. Moreover, several sequencing analysis studies revealed that up to 20% of endocrine-resistant breast cancers harbored ESR1 mutations, while mutations of ESR1 appeared to be rarely happen in primary tumors [33,34,35,36]. The study of genomic landscape of ETs-resistant breast cancer showed that 18% of the tumors harbored mutations of ESR1 [7]. Taken together, the mutation rate in ESR1 in ET resistance breast cancer is 20% approximately. Yates et al. found that the driver coding mutations in ESR1 gene change significantly between primary and metastatic luminal breast cancer [37]. The driver coding mutations in the ESR1 gene confer allele-specific neomorphic properties [33, 38, 39], especially the ER Y537S and D538G mutants have unique transcriptomes and cistromes which drive endocrine resistance and metastases [40]. Spoerke et al. found that multiple different ESR1 mutations coexist in distinct drug-resistant subclonal tumor cells in patients who suffered a failure of endocrine therapy [41]. Moreover, distinct ESR1 mutations differentially affect the efficacy of ER antagonists [42,43,44,45]. Therefore, biopsy and sequencing of advanced ER+ breast cancer may be helpful in some cases, because most ET resistance acquire additional driver mutations not seen in the primary. Razavi et al. identified that ESR1 mutations and lesions in the MAPK pathway, representing different categories of endocrine resistance breast cancer [7]. Although the mutations in the multiple effectors of MAPK signaling or in MYC or other transcription factors were mutually exclusive with hotspot mutations in ESR1 at the level of individual cases in the prospective sequencing cohort, these distinct mutations were observed to coexist in a patient whose multiple tumors were available for analysis [7]. Hence, multiple biologically distinct mechanisms of ET resistance are likely to coexist in subclone cancer cells in individual patients.

Cyclin D–CDK4/6–RB pathway and CDK4/6 inhibitors

Role of cyclin D–CDK4/6–RB axis in G1–S phase transition

Genetic and biochemical characterization of D-type cyclins and their partner cyclin-dependent kinases (CDK4 and CDK6) have been extensively investigated and revealed how mammalian cells regulate G1–S phase transition in a retinoblastoma protein (RB)-dependent manner [46, 47]. RB, which undergoes periodic phosphorylation during cell division cycle, is a well-known master-regulator of the G1/S-checkpoint. RB is dephosphorylated in the mitosis phase and progressively re-phosphorylated first by cyclin D/CDK4/6 complex and later by cyclin E/CDK2 complex in the G1 phase [48]. Hypophosphorylated RB inhibits the transcriptional activation of E2F target genes as cells rested in the G0 or early G1 phase [49, 50]. RB becomes hyperphosphorylated (inactivated) in the late G1 phase, resulting in the loss of its proliferation-suppressive function and promoting the G1–S phase transition [51]. The enzymatic activities of CDK4/6 in the G1 phase are governed by cyclin D expressed in response to various extracellular signals [52, 53]. Therefore, the cyclin D–CDK4/6–RB axis is downstream of multiple mitogenic cascades, making it a valuable target for drug development [54].

The cyclin D–CDK4/6–RB axis deranged in ER+ breast cancer

The RB tumor suppressor gene is functionally inactivated in approximately 20–30% of breast cancers [55], and loss of RB expression is more commonly observed in triple-negative breast cancer [56, 57]. Therefore, RB is proficient in the majority of HR+ breast cancer [58]. Protein p16INK4, which acts as a brake on the activation of CDK4/6 in RB-proficient cells [59], is found to be inactivated in half of invasive breast cancers [60].While the activated cyclin D–CDK4/6 complex plays a central role in the G1/S phase transition in response to oncogenic pathways and cyclin D1 acts as mitogen sensor to govern G1 phase progression, activating mutations in cyclin D and CDK4/6 rarely existed. Many studies reported that overexpression of cyclin D occurred in over half of all breast cancers with or without cyclin D1 gene amplification [61,62,63,64,65,66,67]. Moreover, amplification of cyclin D1 is especially high in ER+ breast cancers (58% in luminal B subtype and 29% in luminal A subtype, respectively) [68]. Amplification of CDK4 is identified in 25% of luminal B cancers and 14% of luminal A cancers [68]. Additionally, amplification of both cyclin D1 and CDK4 is high in HER2-enriched subtype (38% and 24%, respectively) [68]. While ER and HER2 signaling seems to be drivers in the biology of about 70% and 20% of breast cancers, respectively [23, 69], the two pathways share the same downstream or end points on the cyclin D–CDK4/6–Rb axis. That is, the receptor tyrosine kinases (RTKs) signaling can potentiate cyclin D–CDK4/6–Rb axis in an ER-independent fashion.

CDK4/6 inhibitors’ clinical development in patients with breast cancer

Over the past 4 years, three orally available approved CDK4/6 inhibitors (palbociclib, ribociclib and abemaciclib) have been demonstrated to result in significant clinical benefit when combined with ETs in HR+/HER2− advanced breast cancers in the clinical settings [70,71,72,73,74,75]. Palbociclib was the first CDK4/6 inhibitor approved. The clinical trial NCT01684215 (Phase I/II PALOMA-1 study) evaluated the safety and tolerability of the combination of letrozole plus palbociclib in the first-line treatment of HR+/HER2− advanced breast cancer in postmenopausal women (Table 1). The clinical trial NCT00721409 (Phase II PALOMA-1/TRIO 18 study) revealed an impressive improvement in progression-free survival (PFS) in the palbociclib plus letrozole arm (20.2 versus 10.2 months, p = 0.0004) [72]. The OS for NCT00721409 has not been reported. Consistent with findings from the clinical trial NCT00721409, in the clinical trial NCT01942135 (Phase III PALOMA-3 study), the median PFS showed significant improvement in patients treated for 9.5 months in the palbociclib plus fulvestrant group compared with 4.6 months in the placebo plus fulvestrant group. Moreover, the median OS was 34.9 months (28.8–40.0) in the palbociclib plus fulvestrant arm and 28.0 months (23.6–34.6) in the placebo plus fulvestrant arm (HR = 0.81; p = 0.09) [76]. However, all the three drugs inhibit the proliferation of RB-positive tumor cells to induce cell cycle exit and are inactive in RB-negative tumor cells. Currently, CDK4/6 inhibitors are being increasingly employed in clinical trials combined with signaling pathway inhibitors against epidermal growth factor, phosphoinositide 3 kinase (PI3K), or others that upregulated the expression of cyclin D1 or CDK4/6 (Table 2) [77,78,79,80,81]. These combination therapy strategies designed to increase therapeutic efficiency have been extensively and comprehensively reviewed [74, 82]; moreover, the outcomes of some recent investigations showed that the CDK4/6 inhibitors strengthened the cytostatic effect induced by several signaling pathway inhibitors [83]. Further, CDK4/6 inhibition can also affect the tumor microenvironment. For example, CDK4/6 inhibition triggered antitumor immunity in patient-derived breast cancer cell xenografts model and an MMTV-HER2 mouse [84]. Cdk4/6 inhibitor plus an AI or fulvestrant was listed as the preferred treatment option in HR+/HER2− metastatic breast cancer. Therefore, an applied understanding of the outcomes of CDK4/6 inhibitors and practice patterns may generate a hypothesis for subsequent treatments to deal with the coming challenges. Given palbociclib in combination with hormone therapy led to mPFS of 20.7, 12.8, and 4.0 months when administered in the first-line, second-line and third-line in the real-world palbociclib practice pattern [85, 86], one challenge in the treatment of HR+/HER2− advanced breast cancer is deciding the optimal time to introduce a CDK4/6 inhibitor. Moreover, the cytostatic effects of CDK4/6 inhibitors are limited by primary and acquired resistance. Several studies in preclinical settings have demonstrated primary and acquired resistance to CDK4/6 inhibitors mediated by amplification of CDK6 or CCNE1 or FGFR1 gene, and loss of RB1 or FAT1 gene [87,88,89,90]. In the PALOMA-3 clinical trial, evolution of driver gene mutations (such as RB1 mutations, p = 0.041; PIK3CA mutations, p = 0.00069; and ESR1 Y537S mutation, p = 0.003) was common in patients progressing later on palbociclib combined with fulvestrant treatment [91, 92]. The other challenge is to discover how complex the outcomes would be in patients treated with additional CDK4/6 inhibitor therapy after progression. Recently, the treatment with CDK4/6 inhibitors after disease progression is under active investigation in prospective clinical trials, such as the ongoing PACE trial (NCT03147287), a randomized phase II study comparing the median PFS for fulvestrant alone versus fulvestrant + palbociclib versus fulvestrant + palbociclib + avelumab, and the TRINITI-1 trial (NCT01857193), a single-arm phase II trial assessing the antitumor activity of ribociclib + exemestane + everolimus.

Table 1 CDK4/6 inhibitor clinical trials in women with HR+ advanced breast cancer
Full size table
Table 2 Summary of ongoing clinical trials involving CDK4/6 inhibitors in HR+ breast cancer
Full size table

PI3K pathway inhibitors’ clinical development

PI3K pathway is frequently hyperactivated in HR+/HER2− advanced breast cancer and has been implicated in resistance to ETs [93,94,95]. Furthermore, genomic alterations in PIK3CA are common in ER+/HER2− metastatic breast cancer [7, 91, 96,97,98,99]. Thus, the PI3K pathway has emerged as an important therapeutic window for intervention in endocrine-resistant breast cancer. Several PI3K inhibitors combined with various endocrine therapies have been tested in the clinical trial in HR+/HER2− metastatic breast cancer. Pan-Class I PI3K inhibitors (such as buparlisib and pictilisib) have shown modest efficacy in clinical trials [100]. Several clinical trials evaluated the safety and efficacy of buparlisib plus fulvestrant in patients with HR+/HER2− metastatic breast cancer who were pretreated with everolimus plus exemestane (NCT01610284, NCT01633060) [101, 102]. Although the median PFS was significantly improved in the buparlisib versus placebo group (3.9 months vs 1.8 months; HR = 0.67, p = 0.0003), the serious adverse events generated from the off-target effects of the pan-PI3K inhibitors limited the clinical practice of these drug compounds [101, 102]. PI3Kα, which has the most frequent genomic alterations among the class I PI3K isoforms in breast tumors [7, 91, 96,97,98,99], has a prominent role in the PI3K pathway. Selective inhibitors targeting the PI3Kα isoform have been implicated to provide a therapeutic window and to reduce adverse events greatly compared to the Pan-Class I PI3K inhibitors [103]. Alpelisib, an oral selective inhibitor of PI3Kα, was proved to block tumor growth in xenograft models harboring PIK3CA mutations in the preclinical studies [104]. Moreover, alpelisib showed a tolerable clinical safety profile in phase I studies in cohorts of both Western and Japanese patients with PI3KCA-mutated advanced solid cancer (NCT01219699, NCT01387321) [105, 106]. On May 24, 2019, Alpelisib received FDA approval for the treatment of postmenopausal women, and men, with HR+/HER2− , PIK3CA-mutated metastatic breast cancer following progression on or after an endocrine-based regimen. The median OS was 11 months (7.5–14.5) in the alpelisib plus fulvestrant arm and 5.7 months (3.7–7.4) in the placebo plus fulvestrant arm (HR = 0.0.65; p < 0.001) in the cohort of patients with PI3KCA-mutated cancer, and no significant clinical benefit was observed with alpelisib on median PFS in the cohort of patients without a PI3KCA mutation [107, 108].

Histone deacetylases as a therapeutic target in HR+ breast cancer

Preclinical activity of the HDAC inhibitors

In addition to genetic alterations, epigenetic alteration including histone hypoacetylation is a putative mechanism by which tumor cells can develop drug resistance [109,110,111,112]. Aberrant histone deacetylase (HDAC) activity has been demonstrated in breast cancer. In breast cancer core biopsy specimens from 200 patients, HDAC1 expression was associated with estrogen receptor and progesterone receptor expression, and HDAC1 expression predicted significantly better disease-free survival [113]. Muller et al. presented the results of HDACs expression in a large cohort of primary breast cancer cases (n = 238) [114]. HDAC1 was increased in HR+ tumors, while HDAC2 and HDAC3 were strongly expressed in hormone receptor-negative subgroups of tumors with features of a high grade and more aggressiveness [114]. Four ERα corepressors (nuclear corepressor (NCoR), silencing mediator for retinoid or thyroid hormone receptors (SMRT), COUP-TF II and SPEN) have been shown to potentiate endocrine sensitivity in breast cancers [115]. NCoR and SMRT both repress the ERα transcriptional activation depending on HDAC3 activity [116]. COUP-TF II and SPEN attenuate hormonal responses by recruiting HDAC1 to the ERα complex at the genomic sites recognized by ERα [117, 118]. The loss of any of the four corepressors leads to abnormal recruitment of HDACs to ERα-target genes and results in endocrine resistance in breast cancer [115]. These studies have prompted the clinical testing of HDAC inhibitors as anticancer therapeutics in breast cancer [119]. A vast array of both natural and synthetic chemical compounds functioning as HDAC inhibitors were initially discovered based on drug screens for differentiation inducers in leukemias [120, 121]. The HDAC inhibitors have been investigated as therapeutic agents in cancers; for example, romidepsin, vorinostat and belinostat have been approved by the US FDA for treatment of cutaneous or peripheral T cell lymphoma. Panobinostat combined with bortezomib has been approved for the treatment of drug-resistant multiple myeloma. Tucidinostat has been approved in China for relapsed or refractory peripheral T cell lymphoma.

Laboratory research to date supports the investigation of HDAC inhibitors for the treatment of HR+ breast cancer. Several HDAC inhibitors could induce G1 and G2/M cell cycle arrest and subsequent apoptosis or differentiation of both ER-positive and ER-negative breast cancer cell lines [122,123,124,125]. HDAC inhibitors are thought to be able to relieve transcriptional repression in preclinical breast cancer models. Reactivation of silenced ER was observed with vorinostat treatment in preclinical models in hormone receptor-negative tumors [126]. The significance of re-expression of silenced ERα and restoration of sensitivity to endocrine therapy such as AIs were demonstrated in triple-negative breast cancer xenografts following treatment with both HDAC and DNMT inhibitors [127,128,129]. Moreover, entinostat sensitized triple-negative breast cancer xenografts to letrozole [130]. In addition, a significant growth inhibition was also observed in HER2-positive xenograft mouse models following treatment with entinostat plus lapatinib. Mechanistic studies revealed that these effects resulted from downregulation of HER2 and phosphorylated AKT [131, 132]. These experiments provided a strong rationale for combining HADC inhibitors with hormone therapy in advanced HR+ breast cancer clinical trials.

HDAC inhibitors’ clinical development in patients with breast cancer

Several HDAC inhibitors have been evaluated or being evaluated in a number of Phase I/II/III trials in patients with breast cancer. Vorinostat, which targets classes 1 and 2 HDACs, was the first HDAC inhibitor available for investigator-initiated trails. In a Phase II trial of single-agent vorinostat in patients with advanced breast cancer [133], 14 patients received vorinostat at a dose of 200 mg oral twice daily for 14 days of each 21-day cycle. The clinical trial revealed no complete or partial responses, and the study was terminated after the first stage. Although the study did not meet its primary end point, stable disease was observed in almost 30% (4 of 14) of the patients. The therapy was well tolerated with the most common adverse events. Given vorinostat was found to enhance the anti-proliferative actions of tamoxifen on breast cancer cells [134], a Phase II clinical trial (NCT00365599) of vorinostat plus tamoxifen treatment was designed in the hormone therapy-resistant breast cancer setting. 43 women with hormone-resistant breast cancer received oral vorinostat 400 mg daily (21 days of a 28 day cycle) and tamoxifen 20 mg daily [135]. The results showed that the objective response rate was 19% and the clinical benefit rate was 40%. In addition, the combination of these two agents was well tolerated [135]. Entinostat is a synthetic benzamide derivative HDAC inhibitor, which potently inhibits class 1 and class 4 HDAC enzymes. Several clinical trials revealed that oral entinostat was well tolerated in patients with both solid tumors and hematologic malignancies [136,137,138]. ENCORE 301 (NCT00676663) was a Phase II randomized, double-blind, placebo-controlled study of the addition of entinostat to exemestane in patients with HR+ advanced breast cancer with disease progression after prior non-steroidal aromatase inhibitor [139]. The study demonstrated a significant improvement in PFS in the entinostat arm versus placebo (median 4.3 versus 2.3 months, p = 0.055), and an impressive improvement in OS was also observed in the entinostat arm versus placebo (28.1 versus 19.8 months, p = 0.036) [139, 140]. The follow-up randomized Phase III confirmatory study (E2112, NCT02115282) is ongoing [140]. On October 25, 2018, a press release by the Syndax Pharmaceuticals stated that the Phase III breast cancer trial E2112 failed to achieve its statistical hurdle for the co-primary end point of improvement in PFS [141]. However, the final data of the findings from the PFS analysis will not be available until report of the final OS results. Tucidinostat, an oral benzamide class of HDAC inhibitor, selectively inhibits HDAC1, HDAC2, HDAC3 and HDAC10 enzymes. The ACE study (NCT02482753) was a randomized, double-blind, placebo-controlled Phase III clinical trial of tucidinostat plus exemestane [142]. The clinical study revealed a significant improvement in PFS in the tucidinostat arm versus placebo (median 7.4 versus 3.8 months, p = 0.033). The following up for investigation of overall survival is ongoing. Serious adverse events were observed more common in the tucidinostat plus exemestane group (21%, 51 of 244 patients) than in the placebo plus exemestane group (6%, 7 of 121 patients).

Why are the PFS results from E2112 study seemingly divergent to the PFS results from both the ENCORE 301 study and the ACE study? In view of differences among the three clinical trials, the practice pattern, especially the exposure to previous systemic regimens, could be the major factor affecting the clinical outcomes. Patients enrolled in the E2112 study are more likely to have received previous CDK4/6 inhibitors. In China, neither CDK4/6 inhibitors nor everolimus was approved during the enrollment period (July 20, 2015 to June 26, 2017), thus only seven patients in the ACE study had previously received palbociclib and none of the 130 patients in the ENCORE 301 study had taken CDK4/6 inhibitors. The molecular mechanisms governing resistance to CDK4/6 inhibitor combination with endocrine therapy could be distinct from those facilitating resistance to anti-estrogen monotherapy [143]. Therefore, previous exposure to CDK4/6 inhibitors could modulate the therapeutic benefit with subsequent HDAC inhibitor treatment. Similarly, a Phase I/II clinical trial (NCT00258349) was developed to evaluate the response rate after treatment with vorinostat and trastuzumab in patients with HER2-overexpressing metastatic breast cancer with trastuzumab-resistant progressive disease. The results revealed that none of the patients in the primary analysis set responded to combination vorinostat and trastuzumab treatment [144]. Moreover, Kim et al. reported that pretreatment of various tumor cell lines with trichostatin A or vorinostat increased the cytotoxicity of chemotherapy, while administering the HDAC inhibitors after chemotherapy did not achieve the same results [145]. Hence, additional research is needed to determine the optimal treatment sequencing of HDAC inhibitors and the schedule of administration should ideally be modeled preclinically prior to the initiation of clinical trials. Overall, although both entinostat and tucidinostat have not been approved for clinical use by any regulatory agency for the management of HR+ advanced breast cancer, the results of the clinical trials (NCT00676663, NCT02115282 and NCT02482753) represented an important step forward in the development of epigenetic therapy for endocrine-resistant breast cancer.

Conclusions and future directions

Intrinsic and acquired resistance to hormonal therapy results in cancer recurrence and limits clinical benefit on HR+/HER2− advanced breast cancer. Recently, genomic mutations in the ESR1 gene were found in approximately 18% of endocrine-resistant HR+ breast cancers [7, 33,34,35,36]. Importantly, ESR1 mutations differentially affect the efficacy of ER antagonists [42,43,44,45]. Therefore, the ER signaling pathway for tumor progression remains to be elucidated further. While the ESR1 alterations offer beneficial advantageous insights into the genomic evolution of HR+ breast cancers under the selective pressure of drugs, pan-wild-type tumors with unknown mechanisms of ETs accounted for around 60% of patients. Therefore, more research data are required to provide evidence informing optimal sequencing of available therapies for the guidance to develop therapeutic approaches to overcome resistance. Although ESR1 mutations, MAPK alterations and PI3KCA aberrations were mutually exclusive at the level of individual cases in the prospective sequencing cohort according to the taxonomy, they could coexist in the metastatic tumors from one patient. Thus, multiple biologically distinct mechanism of ET resistance probably coexist in distinct tumor subclones in individual patients. Thereby, the effective overcome of ET resistance will be achieved by combination therapies that affect the cell cycle regulation, ER signaling and other compensatory mechanisms and alternative pathways. However, the phenocopy of coexistent mutations in individual cases makes it challenging to develop combination therapies which could uproot all ET resistance clones; hence, cross talk between these signaling pathways is required to be further investigated. Now, the complex mutational genomic landscape and the extensive genomic heterogeneity changes in ETs-resistant breast cancer have been revealed by large-scale genomics analyses [7, 91, 97, 146,147,148]. Novel essential factors contributing to endocrine resistance are being discovered at the preclinical level: for example, the nuclear envelope anchored protein LEM4, transcriptional factor FOXA, and non-coding RNA genes RMRP and NEAT1 [61, 149]. In parallel, genome sequencing efforts of thousands of uncultured tumors have revealed that more than 50% of human cancers harbor mutations in enzymes (HDACs and HATs, TETs and DNMTs, KDMs and KMTs) that are involved in chromatin organization [150,151,152]. The frequent existence of fascinating interplay between the genetic alterations and epigenetic abnormalities promote tumorigenesis and metastasis; for example, PI3K pathway regulates ER-dependent transcription in breast cancer through the epigenetic regulator KMT2D [28]. All these findings would influence clinical practice to personalize therapeutic regimens for individual patients or inform potential approaches to outcome resistance. Moreover, advances in genomic sequencing and other technologies that allow deeper understanding of the genetic alterations and epigenetic abnormalities of individual tumors and further investigation into the cross-talk between these signaling pathways have yielded a superabundance data both in the preclinical and clinical setting. And it is clear that these data require continued systematic mining to reveal many exciting discoveries to personalize therapeutic strategies for each patient with ER positive breast cancer.