Interaction between Estrogen Receptors and p53: A Broader Role for Tamoxifen?

Abstract

Tamoxifen is one of the most widely used anticancer drugs in the world. It is a safe drug with generally well-tolerated side effects and has been prescribed for the treatment of early-stage and advanced-stage or metastatic estrogen receptor α (ERα/ESR1)-positive breast cancer. Tamoxifen therapy also provides a 38% reduction of the risk of developing breast cancer in women at high risk. With the advent of newer medications targeting ERα-positive breast cancer, tamoxifen is now mainly used as adjuvant therapy for lower-risk premenopausal breast cancer and cancer prevention. It is widely accepted that tamoxifen as a selective estrogen receptor modulator exerts its therapeutic effect by competitively binding to ERα, leading to the recruitment of corepressors and inhibition of transcription of genes involved in the proliferation of breast cancer epithelium. As such, expression of ERα in breast tumors has been considered necessary for tumors to be responsive to tamoxifen therapy. However, ERα-independent effects of tamoxifen in various in vitro and in vivo contexts have been reported over the years. Importantly, the recent discovery that ERα and estrogen receptor β (ERβ/ESR2) can bind tumor suppressor protein p53 with functional consequences has provided new insights into the mechanisms underlying response to tamoxifen therapy and resistance. Furthermore, these findings have paved the way for broadening the use of tamoxifen by potentially repurposing it to treat triple negative (negative for ERα, human epidermal growth factor receptor 2, and progesterone receptor) breast cancer. Herein, we summarize these developments and discuss their mechanistic underpinnings and clinical implications.

Tamoxifen (brand names: Nolvadex, Soltamox) is considered the “gold standard” treatment for millions of breast cancer patients. It is approved by the US Food and Drug Administration to treat women and men diagnosed with hormone-receptor-positive, early-stage breast cancer after surgery (or chemotherapy and radiation) to reduce the risk of cancer recurrence. Numerous clinical studies have shown that tamoxifen, a selective estrogen receptor modulator (SERM), is highly effective in the treatment of breast cancer. The Early Breast Trialists Group meta-analysis of 55 clinical trials with 30 000 women reported that 5 years of tamoxifen reduced the risk of breast cancer recurrence by 47% and the risk of death from breast cancer by 26% (1). Extending tamoxifen treatment for 10 years further decreased the risk of recurrence and risk of death from breast cancer (2). Randomized clinical trials of tamoxifen vs other SERMS showed them to be equally or less effective than tamoxifen for treating advanced-stage or metastatic estrogen receptor α (ERα/ESR1)-positive disease enabling tamoxifen to remain as the lead SERM (3, 4). Moreover, a meta-analysis of clinical trials in women randomized to treatment with tamoxifen vs placebos showed that with tamoxifen there was a 38% reduction in breast cancer incidence irrespective of the age of the patients (5), and a long-term follow-up of the International Breast Cancer Intervention Study trial indicated continued risk reduction up to 15 years or more after the 5-year treatment period (6).

Besides targeting the estrogen receptor, aromatase enzyme, which catalyzes the conversion of androgens to estrogens through the “aromatization pathway,” has been an important target for endocrine therapy. Third-generation aromatase inhibitors (AIs) include steroidal (exemestane) and nonsteroidal (anastozole and letrozole) with greater specificity and fewer side effects (7). As AIs inhibit the conversion of androgens to estrogens, they cannot be used to treat breast cancer in premenopausal women unless they are also subjected to ovarian suppression. Currently, AIs are widely used in the treatment of ERα-positive breast cancer in postmenopausal women. A patient-level meta-analysis of 7030 women from 4 randomized trials to investigate AIs vs tamoxifen in premenopausal women with ERα-positive early-stage breast cancer treated with ovarian suppression showed that AIs rather than tamoxifen reduces the risk of breast cancer recurrence; however, longer follow-up is needed to assess any impact on breast cancer mortality (8). In the case of patients with locally advanced and metastatic endocrine-responsive postmenopausal breast cancer, a meta-analysis of phase III randomized controlled trials comparing first-line endocrine therapy with third-generation AIs and tamoxifen concluded that, in the first-line setting, AI instead of tamoxifen has a significant clinical benefit as it increases the duration of tumor control by prolonging progression-free survival; however, overall survival (OS) did not differ between the 2 arms (7). Besides the advantages of AIs, therapeutic resistance with various mechanisms including ERα mutations have been clinically challenging. Improvement in objective response rate and progression-free survival were reported when CDK4/6 inhibitors were combined with AI for first-line therapy or with fulvestrant as second-line therapy after progression or relapse on an AI (9-14).

Fulvestrant is a selective estrogen receptor degrader (SERD) that has been used as an endocrine therapeutic agent for the past several years (15). Unlike tamoxifen, fulvestrant-iduced conformational change of ERα disrupts both AF2- and AF1-mediated transcriptional activity. It acts exclusively as an ERα antagonist, whereas tamoxifen is also a partial agonist of ERα. Futhermore, ERα when complexed with fulvestrant is unstable, leading to accelerated degradation (16, 17). By binding to ERα with high affinity, fulvestrant impairs dimerization of the receptor, thereby blocking its nuclear localization (18). Several randomized trials have demonstrated the efficacy of fulvestrat as a single agent and in combination with various biologic and targeted therapy agents (19). Limitations of fuvestrant usage include intramuscular adminstation because of poor solubility, thereby limiting the volume and dose that can be administered.

New antiestrogen drugs have been developed to circumvent therapeutic resistance mechanisms including acquired ESR1 mutations. These include new SERDs, complete estrogen receptor antagonists, selective estrogen receptor covalent antagonists, and proteolysis-targeting chimerics (PRTOTACs) targeting ERα (19-21). Among them, elacistrant (RAD1901) is a new nonsteroidal oral SERM/SERD at an advanced stage of clinical development (22-24). It degrades ERα in a dose-dependent manner and inhibits estradiol-dependent ERα-mediated gene transcription, leading to inhibition of protumor signaling. In the second- and third-line treatment, this SERD had superiority over fulvestrant and AIs. Elasistrant demonstrated increased efficacy compared to standard of care endocrine monotherapy in the randomized phase III EMERALD trial, especially in tumors with ESR1 mutations (25). It has gained approval in the United States and Europe for the treatment of postmenopausal women or adult men with ERα⁺, human epidermal growth factor receptor 2 (HER2), ESR1-mutated advanced metastatic breast cancer with disease progression following at least 1 line of endocrine therapy (25-27). With the advent of such newer medications targeting ERα-positive breast cancer, tamoxifen is now mainly used as adjuvant therapy for lower-risk (cancer that tends to grow slowly, usually does not spread to other parts of the body, and may have a good chance of being cured) premenopausal breast cancer and cancer prevention (28).

The discovery of ERα and the finding that ERα-rich breast cancers were more likely to be responsive to endocrine ablation (29, 30), followed by the development of monoclonal antibodies against ERα (31) and molecular cloning of ERα (32, 33), paved the way for breast cancers to be classified as estrogen receptor-positive or estrogen receptor-negative based on the expression of ERα. It is widely accepted that tamoxifen exerts its therapeutic effect by competitively binding to ERα leading to recruitment of corepressors and the inhibition of transcription of genes involved in the proliferation of breast cancer epithelium. Therefore, expression of ERα in breast tumors has been considered necessary for these tumors to be responsive to tamoxifen therapy (34). However, ERα-independent effects of tamoxifen in various in vitro and in vivo contexts have been reported over the years (35).

A second estrogen receptor, termed estrogen receptor β (ERβ), was discovered in the mid-1990s. ERβ was originally identified in rat and subsequently in human and mouse (36-38). This discovery necessitated the reassessment of classical estrogen signaling models (39-42). Twenty years ago, Jensen et al reported that predominantly ERβ, not ERα, was coexpressed with proliferation marker Ki67 in epithelial cells of primary and locally recurring breast cancer (43). ERβ was shown to increase the efficacy of antiestrogens in breast cancer cells by its effects on apoptosis and cell cycle (44). Over the years, structural and functional comparisons of these 2 receptors have been extensively covered in other reviews (45-47).

Regarding domains required for ERα-p53 and ERβ-p53 interactions, both ERα and ERβ bind p53 with functional impact. ERβ binds to the C-terminal domain (CTD) of p53 (48-50) as does ERα (51). The domain of ERα from aa184-351 encompassing the DNA binding domain (DBD), hinge domain (H), and part of the ligand binding domain binds to the aa 363-393 CTD of p53. In the case of ERβ-p53 interaction, we mapped the region between aa 149-248, encompassing the DBD and H of ERβ required to bind CTD of p53 (49, 50). Bado et al mapped the binding to aa 211-530 that contains part of the DBD, the H domain, and the AF2 domain of ERβ to aa 296-393 of p53 (48) (Fig. 1).

Figure 1.

Domains required for ERα-p53, and ERβ-p53 interactions. The domains of ERα and p53 required for their direct interaction are shown (solid lines represent the domains determined by Das Laboratory (49, 50) and dotted lines represent the domains determined by Thomas Laboratory). Created in BioRender. Oturkar, C. (2025) https://BioRender.com/a11u356.

Abbreviations: CoAc, coactivator; CoR, corepressor; ERα, estrogen receptor α; ERβ, estrogen receptor β.

ERα and Tumor Suppressor Protein p53 Status in Patient Tumors

An inverse relationship between expression levels of ERα and p53 tumor suppressor protein as determined by immunohistochemistry (IHC) was reported more than 3 decades ago based on pathological studies of breast tumors (52, 53). Wild-type p53 protein is far less stable compared to mutant p53, and therefore, low levels of p53 in tumors had been considered as a surrogate for wild-type p53, and high levels of IHC staining for p53 is indicative of mutant p53. Since the advent of DNA sequencing technology, p53 mutational status in tumors has been determined by DNA sequencing in addition to immunohistochemical assessment. In general, ERα-positive breast cancer expresses wild-type p53, whereas p53 mutations are high in ERα-negative tumors, and p53 mutations are associated with worse overall and disease-free survival (54-57). More recently, mutual exclusivity between mutant ERα and mutant p53 in metastatic breast cancer has been reported (58-60). Furthermore, clinical studies showed that ERα-positive tumors expressing wild-type p53 were more responsive to adjuvant tamoxifen therapy leading to better overall survival and progression-free survival, compared to tumors with mutated p53 (61-63). Functional p53 deficiency, as assessed by gene expression readout, was shown to be predictive of resistance to tamoxifen in terms of shorter disease-specific survival and distant metastasis-free survival (64, 65). In a 12-year median follow-up of ERα-positive patients including those treated with tamoxifen in the International Breast Cancer Study Group Trials VIII and IX, wild-type p53 (as determined by IHC) was found to be associated with an increase in disease-free survival (66).

ERα-p53 Signaling Crosstalk

About 20 years ago, observations alluding to potential crosstalk between ESR1 and p53 signaling pathways started to emerge. For example, p53 was found to be required for hormonal protection against carcinogen-induced mammary cancer in rodents (67, 68). p53 was reported to bind ERα and repress in vitro transcriptional activity of ERα using reporter assays (69). In breast cancer cells, increased expression of ERα led to elevated levels of p53 and its inhibitor MDM2 (70), whereas overexpressed MDM2 functionally activated ERα (71). Importantly, ERα was shown to be capable of directly interacting with and functionally inactivating wild-type p53 bound to the promoters of its prototypic target genes in ERα-positive breast cancer cells and MCF-7 xenografts (51, 72-74).

The repression of wild-type 53 function by ERα was manifested in both transcriptional activation of p53-target genes such as CDKN1A (p21) and repression of p53-target genes BIRC5 (Survivin) and multidrug resistance gene 1 (MDR1). In other words, both transcriptional activation and transcriptional repression capabilities of p53 were suppressed by ERα. Furthermore, ERα repressed p53-dependent apoptosis in MCF-7 luminal breast cancer cells (74). Cooperation in cis between ERα and p53 bound to their cognate binding sites on FLT1 gene promoter to activate transcription has been reported as another mode of crosstalk between these 2 proteins (75). Genome-wide approaches showed that in response to doxorubicin treatment, estrogen receptor antagonized the proapoptotic function of p53 by targeting a set of proapoptotic genes directly regulated by p53, and both estradiol and tamoxifen promoted this antagonism (76). Besides the crosstalk by protein-protein interaction, ERα activated transcription of the p53 gene and vice versa (77-80). Based on various studies, there appears to be a positive feedback regulatory loop between ERα and p53 (80-82). Using the Cancer Genome Atlas breast invasive carcinoma patient cohort, we have shown that the expression of MDM4 and MDM2 is elevated in primary human breast cancers of luminal A/B subtypes and associates with ERα-positive disease. Additionally, ERα also positively regulates p53 expression, and endogenous MDM4 negatively regulates ERα expression and forms a protein complex with ERα in breast cancer cell lines and primary human breast tumor tissue. These findings implicate ERα as a central component of the p53-MDM2-MDM4 signaling axis in human breast cancer (80).

ERα-p53 Interaction and Response to Tamoxifen Therapy

As described in the earlier section, several studies on overall survival and progression-free survival of breast cancer patients treated with tamoxifen reported that ERα-positive tumors harboring wild-type p53 were more responsive to tamoxifen while those with p53 mutations were generally resistant (56, 57). However, the mechanistic basis for this observation remained unclear until the finding that ERα uses dual strategies to promote abnormal proliferation of breast cancer cells: (1) upregulating the transcription of proproliferation genes via estrogen response element (ERE) in their promoters (the conventional route) and (2) repressing p53-responsive antiproliferation genes by interacting with p53 and functionally suppressing it (72). Both these modes of ERα function are blocked by tamoxifen. Tamoxifen inhibits the ERE-mediated transcriptional activation by binding to ERα and enabling the recruitment of corepressors such as NCoR and SMRT (83, 84). Besides this well-known classical function of tamoxifen, it is also capable of downregulating ERα-p53 interaction and blocking transcriptional repression of p53 target genes. Thus, tumor suppressor functions of wild-type p53 are inhibited by ERα in breast cancer, and tamoxifen reactivates p53 by disrupting ERα-p53 interaction (72) (Fig. 2). This phenomenon in vivo in patient tumors was demonstrated in our randomized window-of-opportunity trial in patients with newly diagnosed stage I–III ERα+/HER2−/wild-type p53 breast cancer who were randomized to arms with or without tamoxifen treatment for 4 weeks prior to surgery (85). This trial revealed that the ERα-p53 interaction in tumors was inhibited by tamoxifen. Functional reactivation of p53 lead to transcriptional reprogramming that favors tumor-suppressive signaling, as well as downregulation of oncogenic pathways. These findings illustrating the convergence of ERα and p53 signaling during tamoxifen therapy provide a mechanistic explanation for previous clinical observations where ERα-positive breast cancer patients with tumors harboring wild-type p53 were more responsive to tamoxifen therapy as compared to those with p53 mutations. These observations stress the need for factoring p53 mutational status in developing new endocrine therapy strategies for breast cancer.

Figure 2.

Schematic model for protumorigenic gene expression program mediated by activated ERα and repressed p53 in response to estradiol and antitumorigenic gene expression program mediated by repressed ERα and activated p53 in response to tamoxifen. Created in BioRender. Oturkar, C. (2025) https://BioRender.com/v46f294.

Abbreviation: ERα, estrogen receptor α.

Crosstalk Among ERα, ERβ, and P53 in ERα-positive Breast Cancer Cell Lines and Tumors

In ERα-positive breast cancer cell lines, physical interaction between ERβ and p53 reduces ERα-p53 binding and antagonizes ERα-mediated proliferation and antiapoptotic signaling (82). In the presence of 17β-estradiol, ERβ activated ERα-repressed genes in a p53-dependent manner and ERβ competed with ERα for p53-mediated transcriptional activation. Furthermore, analysis of the Oncomine database showed that high expression of 172 genes that were targets of binding by both ERα and ERβ [as identified by chromatin immunoprecipitation sequencing (ChIP-Seq)] was significantly associated with better prognosis of breast cancer patients (82, 86). Based on RNA-sequencing, ChIP-Seq data sets, and ERβ ChIP assay, ERβ and p53 regulate a common set of gene targets and ERβ abrogated ERα binding to the prototypic p53-target gene CDKN1A (82). It is not clear if p53 and ERβ bind to the same chromatin region or if p53 is necessary for ERα and ERβ to bind the CDKN1A gene. Another study showed that MCF-7 cells expressing endogenous wild-type p53 and stably transfected exogenous ERβ showed increased sensitivity to a combination of cisplatin and tamoxifen (87). Knocking down ERβ in MCF-7 cells followed by cisplatin treatment resulted in a considerable decrease in transcription of p53-target genes such as BBC3 (PUMA) (87). Our studies using ChIP assay demonstrated depletion of ERβ in MCF-7 cells leading to increased p53 binding to CDKN1A gene promoter resulting in increased transcription (50). Importantly, ERα was in a complex with p53 at the p53 binding site on CDKN1A gene promoter leading to repression of CDKN1A transcription, and p53 was necessary for ERα to access this site (51), suggesting that ERα and ERβ have opposite effects on p53-mediated transcription. Thus, data from these 2 studies (50, 51) suggest that both endogenous ERα and ERβ are capable of binding and functionally repressing wild-type p53 in MCF-7 cells in the absence of any ligands or genotoxic agents. These apparently disparate findings from different laboratories on luminal breast cancer cells such as MCF-7 reflect the complexity of interplay among these 2 receptors and p53 when all 3 are endogenously expressed to different levels depending on the cellular milieu and the type of ligands and genotoxic exposure.

Both ERα and ERβ bind to the C-terminal domain of p53 (48, 50, 51). In MCF-7 cells, overexpression of exogenous ERβ resulted in decreased ERα-wild type p53 interaction (82, 87). However, whether the endogenous ERα and ERβ competes for binding to the C-terminal domain of p53 and, if so, the cellular signaling that dictates the increased binding capability of 1 receptor over the other remains unknown. Analysis of the genome landscape of ERα and ERβ binding in breast cancer cells by Liu et al showed that regions bound by the receptors have distinct properties (88), whereas global analysis of the binding sites by Grober et al indicated that the vast majority of genomic targets of ERβ can also bind ERα (89). Furthermore, the presence of coexpressed ERα and ERβ and their heterodimerization can influence chromatin accessibility and their binding specificities in the context of different ligands (86, 90-93).

ERβ-P53 Interaction in Triple-Negative Breast Cancer Cell Lines and Tumors

About 15% to 20% of breast cancers are classified as triple-negative [negative for ERα, HER2, and progesterone receptor (PR)] breast cancer (TNBC). Although TNBC cells and tumors lack expression of ERα, ERβ is expressed in a subset of these tumors. The use of multiple antibodies including several unvalidated ones and the use of nonstandardized methodologies for ERβ detection have contributed to variable detection and quantitation of ERβ levels in cells and tumors. Based on several current findings, ERβ is expressed in a significant number (70-75%) of TNBC tumors (94-96). However, the role of ERβ in breast cancer has been elusive (41, 97). As noted in the earlier section, in ERα-positive breast cancer cells, ERβ negatively regulates functions of ERα. In the case of TNBC, several studies have shown antiproliferative, antimetastatic, and anti-epithelial to mesenchymal transition (EMT) effects of ERβ (94, 98). Both antitumorigenic and protumorigenic functions were attributed to ERβ (95) and ERβ isoforms have been shown to have contrasting activities in TNBC (99). Cellular signaling context and interaction with other proteins have been proposed as a plausible mechanism. Wild-type vs mutant status of p53 in tumors was not considered in the interpretation of data from these studies. Only some of the divergent effects of ERβ could be attributed to the expression of specific isoforms and their cellular location (100). Although bifaceted functioning has been proposed to reconcile disparate findings on ERβ functions (101, 102), the mechanistic basis for such duality remained largely unknown until it was discovered that p53 status is an important determinant of pro- vs antiproliferative activity of ERβ1 in breast cancer cells (49, 50). In breast cancer cells, ERβ1-p53 interaction elicits an oncogenic vs tumor suppressor function depending on whether p53 is wild-type or mutant, respectively. As about 80% TNBC tumors express mutant p53, ERβ1 functions as a tumor suppressor in a majority of these tumors (50). Mukhopadhyay et al showed that depletion of ERβ1 in MDA-MB-231 TNBC cells resulted in decreased expression of p53-target genes such as CDKN1A and BBC3. Opposite effects were observed when mutant p53 in these cells was knocked out, followed by expression of exogenous wild-type p53, demonstrating that the function of ERβ1 is dependent on the p53 mutational status in this isogenic cellular context. Mechanistically, it was shown that ERβ1 sequestered mutant p53 from mutant p53-p73 complex thereby blocking mutant p53-p73 interaction resulting in reactivation of p73 enabling it to mediate antiproliferation and tumor suppressor functions (Fig. 3). One of the major gain-of-functions of mutant p53 is its ability to bind and inactivate tumor suppressor p73, a member of the p53 family (103-110). Consistent with the ERβ1 disrupting mutant p53-p73 interaction and consequent reactivation of p73 in TNBC cells, ChIP assays showed that depletion of ERβ1 decreased occupancy of p73 on CDKN1A and BBC3 gene promoters. The tumor-suppressive effect of ERβ1 in the context of mutant p53 in TNBC cells was corroborated by a retrospective analysis of Molecular Taxonomy of Breast Cancer International Consortium data. In this cohort, high levels of ERβ mRNA along with mutant p53 expression in tumors were associated with better prognosis with higher breast cancer-specific survival (50). On the other hand, patients with tumors harboring wild-type p53 with high ERβ expression were trending toward worse survival, although statistical significance was not obvious, apparently because of the very low number of patients in this category.

Figure 3.

Schematic model depicting disruption of mutant p53-p73 complex by ERβ in response to tamoxifen treatment, enabling reactivated p73 to activate transcription of antiproliferation genes. Mutant p53 sequestered by ERβ is unable to activate genes involved in regulating oncogenic pathways. Created in BioRender. Oturkar, C. (2025) https://BioRender.com/p07i687.

Abbreviation: ERβ, estrogen receptor β.

Separately, Bado et al reported ERβ binding to mutant p53 and p63, and silencing of p63 in MDA-MB-231 cells resulted in decreased ERβ binding to p53 (48). Mutant p53 can bind p63, inhibiting its activity and promoting cancer metastasis (111, 112). ChIP showed that ERβ1 and mutant p53 bind to mutant p53/p63 target genes involved in the regulation of metastasis such as ADAMTS9. Depending on the target gene, p63 either increased or decreased ERβ1-p53 interaction. Based on these observations, the authors proposed a model where mutant p53 along with p63 binds and represses some of the metastasis-inhibiting genes resulting in EMT and invasion of TNBC cells. ERβ1 recruited to these binding sites will attenuate the inhibition of p63 by mutant p53, leading to upregulation of these genes and suppression of EMT and invasion. It is possible that, depending on target genes, cellular context, presence and absence of particular ligands, and the drugs the cells were exposed to may facilitate differential binding of mutant p53 to p73 or p63 recruited to the target gene promoters.

Effect of Ligands

Bado et al showed that in TNBC cells overexpressing exogenous ERβ, ERβ1 agonists 17β-estradiol, 5α-androstane-3β,17β-diol, and diarylpropionitrile (DPN), but not tamoxifen, significantly increased the ERE-dependent transcription from a reporter plasmid (48). Notably, a different pattern was observed with effects of DPN and tamoxifen on transcription of endogenous genes regulating metastasis. 4-hydroxy tamoxifen upregulated expression of antimetastatic genes CCNG2, SHARP1, ADAMTS9, and GRP87, as well as prometastatic gene FST, whereas DPN downregulated ADAMTS9, GRP87, and FST. Reese et al demonstrated that activation of ERβ by estrogen elicited anticancer effects in TNBC by inducing cystatins, a family of secreted proteins that inhibit TGFβ signaling and suppress metastatic phenotypes (113). On the other hand, treatment of TNBC cells with ERβ agonist DPN increased the proliferation of these cells (114). The basis for these opposite results is not known.

Notably, Ma et al reported a switch from high to low ERα/ ERβ ratio during mammosphere formation from adherent cells, and ERβ was identified as a mediator of estrogen action in breast cancer stem cells (115). Mammosphere formation from both MCF-7 and MDA-MB-231 cells was increased in response to treatment with estradiol or DPN, whereas mammoshere formation from breast stem cells was reduced by DPN, suggesting that ERβ may function differently in breast cancer stem cells vs normal breast stem cells. The effect of endocrine therapies including tamoxifen on breast stem cells and the role of ERβ and p53 in therapeutic response need to be evaluated further in future studies.

Fulvestrant was reported to increase the level of ERβ in ERα+/ERβ+ MCF-7 breast cancer and in ERα−/ ERβ+ MDA-MB-231 TNBC cells. Tamoxifen and fulvestrant as single agents as well as the combination when administered in addition to estradiol suppressed the growth of ERα+/ERβ+ MCF-7 xenograft tumor. Monotherapy with fulvestrant, but not tamoxifen, was effective in reducing the growth of MDA-MB-231 TNBC cell-derived xenograft (116), However, the effect of fulvestrant plus tamoxifen combination on TNBC xenograft was not reported.

Repurposing Tamoxifen for TNBC?

The current standard of care of TNBC consists of chemotherapy regimens such as anthracyclines and taxanes with an evolving role for immunotherapy and drugs targeting DNA damage and repair. Advances in the treatment of metastatic TNBC have resulted in chemoimmunotherapy and antibody drug conjugates as available options. However, none of these regimens are curative, and response rates have been low; adverse effects are yet another challenge. Therefore, there is an unmet need for rationally designed therapies for TNBC. As ERα, the traditional target of tamoxifen, is not expressed in TNBC, tamoxifen therapy is not prescribed for TNBC. However, Gruvberger-Saal et al (117) and Honma et al (118) showed that high expression of ERβ in ERα-negative and TNBC cases, respectively, was significantly associated with better survival outcomes in patients treated with tamoxifen. In another study, ERβ along with a coregulator, steroid receptor RNA activator protein, was found to be predictive for benefit from tamoxifen therapy in patients with ERα-negative early breast cancer (119). On the other hand, there have been reports that ERβ1 protein expression was not associated with better survival outcomes (96, 120, 121). Lack of consideration of the p53 context of tumors could be one of the contributing factors for these inconsistent observations. A recent study that interrogated ESR2 RNA levels by RNA sequencing in a large population-based cohort of primary breast tumors from the SCAN-B study showed that high ESR2 RNA levels correlated with increased OS in TNBC patients in patients treated with endocrine therapy (122).

The finding from our laboratory that p53 status (wild-type vs mutant) is a determinant of pro- vs antitumorigenic effects, respectively, of ERβ1 in breast cancer provides a mechanistic basis for the functional dualism of ERβ in estrogen receptor-negative breast cancer and a new means of stratifying TNBC for increased therapeutic effect (50). Importantly, 4-hydroxy tamoxifen increased ERβ-mutant p53 interaction in TNBC cells while decreasing mutant p53-p73 interaction, causing p73 reactivation and transcriptional upregulation of antiproliferation genes CDKN1A and BBC3 (Fig. 3). Furthermore, both ERβ and p73 were necessary for 4-hydroxy tamoxifen to enhance transcription of these genes in these cells.

The clinical relevance of findings by Mukhopadhyay et al is supported by a recent proof-of-concept clinical observation in a patient with TNBC brain metastasis who had significant tumor regression following treatment with tamoxifen for a few weeks after other therapies failed (123). The metastatic tumor was ERβ-positive and had mutant p53. After tamoxifen therapy was successful in eliminating metastatic tumors in the brain, the patient was on immune checkpoint inhibitor therapy. After 5 years, the patient is “metastasis-free” as of July 2024 (personal communication). Currently, a phase II clinical trial (NCT06434064) in patients with metastatic TNBC to test the effect of a tamoxifen and chemotherapy combination is being set up at the Roswell Park Comprehensive Cancer Center.

The incidence of breast cancer brain metastasis in TNBC patients is 20% to 50%, with very poor prognosis, especially in premenopausal patients (124, 125). Breast cancer brain metastasis is a complex process involving the interaction between tumor cells and normal brain cells such as endothelial cells, macrophages, lymphocytes, astrocytes, and microglial cells in the tumor microenvironment (125, 126). Preclinical and clinical observations suggest the potential to use tamoxifen as monotherapy or in combination with other therapeutic agents for treating metastases to the brain. Over the past several years, there have been several clinical case reports on tamoxifen inhibiting brain metastases of both ERα-positive tumors as well as TNBC (127-132). These reports, along with our finding that both ERβ and mutant p53 are expressed in the metastatic tumor and the interaction between them was increased in response to tamoxifen therapy in the case of the patient we had reported on (123), should lead to basic and clinical studies aimed at understanding the role of tumor-intrinsic and -extrinsic ERβ and p53 signaling in brain metastasis for maximizing the antitumor effects of agents such as tamoxifen, especially in TNBC.

Completed and Ongoing Clinical Trials Targeting ERβ in TNBC

A multicenter, randomized, double-blind, prospective clinical trial with patients stratified by the HER2 status (NCT02062489), “Evaluation of Tamoxifen's Efficacy for ER/PR Negative, ER-beta Positive Operable Breast Cancer Patients,” sponsored by Sun Yat-Sen Memorial Hospital of Sun Yat-Sen University, China, is ongoing. The trial is designed to evaluate the effectiveness of tamoxifen as adjuvant therapy for ER (ER-α)/PR negative, ERβ positive operable breast cancer patients.

There are other clinical trials targeting ERβ with ligands other than tamoxifen for TNBC therapy. The following trials have been completed or are ongoing:

• A phase I trial, “A Pre-surgical Clinical Trial of Therapy with S-equol in Women with Triple Negative Breast Cancer” (NCT02352025) sponsored by The University of Texas Health Sciences Center at San Antonio was started in 2015 and completed in 2020. The primary objective of this study was to determine if S-equol, an ERβ agonist, is effective in decreasing the proliferation rate of TNBC. All stages of breast cancer were eligible. The study showed that S-equol is a well-tolerated oral ERβ agonist with inhibition of proliferation in patients with TNBC as measured by a decrease in Ki-67. However, it is important to be aware of certain off-target effects of ERβ-selective ligands. For example, among the several mRNAs translationally induced by Equol was the oncogene and eIF4G enhancer, c-myc (133).

• A phase 2 study of high-dose estradiol (10 mg by mouth 3 times a day) for targeting estrogen receptor beta in a phase II study in metastatic TNBC was undertaken by a Wisconsin Oncology Network study (NCT01083641) (134). There were no adverse events from estradiol. However, the study was closed after the first stage because of lack of efficacy.

• In another ongoing multi-institutional phase 2 study, estradiol is used for treating patients with ERβ positive (>25% moderate or strong nuclear staining), triple-negative locally advanced or metastatic breast cancer (NCT03941730; started at Mayo Clinic in 2019). The primary objective is to assess the antitumor activity of estradiol (2 mg 3 times daily).

• An interventional phase IV prospective multicenter randomized control clinical trial (NCT02089854) titled “Evaluation of Adjuvant Endocrine Therapy for Operable ER-beta Positive, ER-alpha/PR Negative, Her-2 Negative Breast Cancer Patients” sponsored by Sun Yat-Sen Memorial Hospital of Sun Yat-Sen University, China, is ongoing. The purpose of this study is to determine the effectiveness of adjuvant endocrine therapy (toremifene for premenopausal and anastrozole for postmenopausal patients) for operable TNBC patients. The disease-free survival and OS between the endocrine group and observation group will be compared to evaluate the effectiveness.

The rationale for the use of S-equol (NCT02352025) or estradiol (NCT01083641 and NCT03941730) to target ERβ appears to be based on the view that these agonists will activate ERβ-mediated gene expression changes leading to tumor suppression. The NCT02062489 and NCT02089854 trials appear to be based on the premise that ERβ is pro-oncogenic, and antiestrogenic agents such as tamoxifen, toremifene, or anastrazole will block ERβ signaling. In other words, the basis for these clinical trials appears to be that estradiol/DPN is an agonist of ERβ and tamoxifen is an antagonist. However, our preclinical data have shown that the mechanism underlying ERβ function in breast cancer is not simply based on the traditional agonist-antagonist concept. Instead, as described earlier, we have uncovered a critical vulnerability of TNBC dependent on augmenting the protein-protein interaction between ERβ and mutant p53 by tamoxifen. Estradiol, DPN, or 4-[2-phenyl-5,7-bis(trifluoromethyl)pyrazolo[1,5-a]pyrimidin-3-yl]phenol, a synthetic nonsteroidal ERβ-antagonist, do not increase this interaction (manuscript under preparation). This finding is consistent with the lack of response to estradiol in the trial by the Wisconsin Oncology Network (134).

Tamoxifen in Immunotherapy

Research on tumor endocrine therapy has been primarily focused on the tumor-intrinsic functions of estrogen receptors. However, it is becoming increasingly clear that estrogen signaling is important in the tumor-extrinsic stromal microenvironment (135, 136). The presence of estrogen receptors and the importance of estrogen signaling in cells of the immune system such as myeloid-derived suppressor cells (MDSCs), macrophages, dendritic cells, neutrophils, eosinophils, T-cells, and B-cells have been reported (135, 137-139). However, relative levels and specific functions of ERα vs ERβ in these cells remain unknown. Immunotherapy such as immune checkpoint inhibitor therapy frequently encounters blockage by protumorigenic proteins in various cells including MDSC. MDSCs express Erα, and estradiol signaling influences MDSC expansion in various tumor models including breast and ovarian cancer (140). The inhibition of estrogen signaling by removal of ovaries or by tamoxifen decreased the number of MDSCs and associated protumorigenic functions regardless of the tumor-intrinsic ERα status. These data suggest a new opportunity for using tamoxifen in combination with immunotherapy or chemoimmunotherapy agents. Patient stratification is necessary to determine who will be responsive to these therapeutic strategies (140, 141). Besides ERα, ERβ is also expressed in various immune cell types (142). However, the functional roles of these receptors in the tumor microenvironment and underlying mechanisms remain to be defined. A new generation of SERDS was reported to interact with ERα-positive immune cells in the tumor microenvironment such as MDSC, tumor-infiltrating lymphocytes, and other selected immune cell subpopulations such as eosinophils (138). SERD-induced inhibition of MDSCs and concurrent actions on CD8+ and CD4+ T-cells enhanced the interaction of immune checkpoint inhibitors with breast cancer cells in preclinical models, thereby leading to enhanced tumor killing of even aggressive TNBCs that lack tumor-intrinsic ERα expression (143). Pharmacological activation of ERβ increased the innate immunity to suppress TNBC metastases (144) and also overcame tumor resistance to immune checkpoint blockade therapy (145). Recently, Yuan et al reported a cell-autonomous ERβ function where T-cell receptor activation triggered ERβ phosphorylation leading to increased downstream signaling via nongenomic actions of ERβ. Furthermore, activation of ERβ by S-equol facilitated T-cell receptor activation that stimulated ERβ phosphotyrosine switch and enhanced anti-PD1 immune checkpoint therapy (146). In another study, estrogen-induced polarization (M1 to M2) of brain microglia dampened tumor-suppressive immune functions and promoted TNBC tumor stem cell growth. M2 microglia were found to have infiltrated brain metastases in premenopausal breast cancer patients (131). A similar observation was made in mice, and blocking estrogen signaling either by tamoxifen treatment or ovariectomy suppressed M2 microglial polarization, resulting in suppression of brain metastases. The pharmacological activation of ERβ in TNBC cells was found to induce upregulation of IL-1β, which triggered antitumor neutrophil chemotaxis, suggesting that activation of ERβ could augment innate immunity in addition to its tumor-intrinsic antitumor effects.

Mutant p53 was reported to suppress both cell-autonomous and non-cell-autonomous signaling to promote cancer cell survival and immune surveillance evasion (147, 148). Whether ERβ-p53 crosstalk within the tumor cell and/or in the tumor microenvironment has a role in tumor cell extrinsic estrogen signaling that impacts tumor growth and metastasis, and if so, whether it can be targeted by tamoxifen remains to be determined.

Conclusion

The finding that the wild-type vs mutant status of p53 is 1 of the determinants of pro- vs antiproliferative/tumorigenic functional duality of ERβ in breast cancer and tamoxifen can enhance the interaction between ERβ and p53 has opened up the possibility of repurposing tamoxifen as a single agent or in combination with chemotherapy and/or immunotherapy for TNBC. The approach, if successful, will lead to a therapy that is fundamentally better than current interventions in terms of effectiveness, time needed to reach the patients, and safety and is less financial toxicity to patients. In this regard, the clinical case of 1 TNBC patient whose brain metastatic lesions responded to tamoxifen is promising; however, the generality of this benefit needs to be validated through clinical trials. The benefit of repurposing tamoxifen as mono- or combination therapy may not be limited to TNBC and may be applicable to other cancers with ERβ and mutant p53 expression. Furthermore, exploiting the crosstalk between estrogen receptors and p53 (wild-type vs mutant) for stratifying luminal, HER2+, and TNBC patients to identify those who are likely to be responsive or resistant to therapy will be an attractive strategy. Whether different mutations in p53 and ERs differentially affect the interaction between these 2 proteins and the response to tamoxifen remains an open question. Also, if new-generation SERDS and SERMS affect estrogen receptor-p53 interaction, and if so, its functional consequences, need to be determined.

The presence of both ERα and ERβ capable of interacting themselves with functional impact makes it challenging to analyze mechanisms and consequences of estrogen receptor-p53 signaling crosstalk in luminal breast cancer. Rigorous studies are needed to decipher the functional consequences of crosstalk between the estrogen receptors and p53 (wild-type and mutant) and the effect of ligands and genotoxic agents in luminal A vs B and in HER2+ breast cancer. For example, cell line models where ESR1 and ESR2 are separately knocked down or knocked out (rather than overexpression under the nonstringent assumption that the endogenous protein is not expressed) may be used for molecular analyses such as RNA sequencing and ChIP-seq and response to drugs. In vivo experiments using ESR1 and ESR2 knockout mouse tumor models could complement the cell model studies.

The effect of posttranslational modifications of estrogen receptors and p53 on their interaction and response to chemo and endocrine therapeutic agents is an understudied area. A cautionary approach is warranted in interpreting data from overexpression of exogenous receptors and p53 as it may overwhelm the cellular machinery involved in posttranslational modifications, cellular localization, and balance of interaction among multiple protein partners.

Besides the effects of tamoxifen mediated by nuclear ERα and ERβ, other mechanisms of tamoxifen action have been reported. Mitochondrial and membrane signaling can also be targets of tamoxifen with or without direct involvement of estrogen receptors. Physiological and pathological roles of ERβ localized in mitochondria is an active area of research (149, 150). Tamoxifen was reported to regulate cell fate through mitochondrial ERβ in breast cancer (151). ERβ knockdown leads to uncoupling phenotype (152), whereas activation of ERβ decreases oxidative phosphorylation and disrupts mitochondrial complex activity (153). The role of crosstalk between ERs and p53 is not defined in these mitochondrial phenomena.

It should be noted that unlike in the case of ERα, several reports on the expression and functioning of ERβ in various cells and tissues and prognostic significance in cancer have been plagued with uncertainty in the specificity of certain ERβ antibodies, different methods used for preparing cell extracts and protein retrieval from tissues, lack of specificity of isoform detected, and different IHC cut-off values used for prognostic determination based on ERβ expression in patient tissues. It is prudent to be aware of this problem while assessing the quality and conclusions of various reports on ERβ. Even among reports on attempts to select validated antibodies, there appears to be no clear consensus on antibodies that are specific vs nonspecific in detecting ERβ under different cellular contexts and experimental conditions (154-157). For rigor and reproducibility of data, it is necessary to strive for defining standardized methods for validating isoform-specific ERβ antibodies and for detection and quantifying ERβ in tumor cells and tissues. Furthermore, rather than attributing all inconsistent data to the variable quality of different antibodies, it may be prudent not to overlook the fact that deficiencies in our understanding of the biology of ERβ in different tissue and cellular contexts may be contributing to the inconsistencies in analyzing and interpreting expression and function of ERβ.

Funding

The laboratory research was supported by grants from the National Cancer Institute (NCI) at the National Institutes of Health (NIH), USA (R21CA137635, R01CA079911, and R01CA251545 to G.M.D., and P30CA016056 involving the use of Roswell Park Comprehensive Cancer Center's Shared Resources the U.S. Department of Defense Breast Cancer Research Program (BC180932/W81XWH-19-1-0111) to G.M.D.), Roswell Park Alliance Foundation, USA to G.M.D., and Breast Cancer Coalition of Rochester to G.M.D.

Disclosures

There are no conflicts of interest.

Data Availability

Data sharing is not applicable to this article as no datasets were generated or analyzed during this study.