Accelerating drug development in breast cancer: new frontiers for ER inhibition

Abstract

The estrogen receptor (ER) is an important driver in the proliferation, tumorigenesis, and progression of breast cancers, and targeting ER signaling at different levels is a successful strategy in the control of hormone receptor positive (HR+) breast cancer. Endocrine therapy has been the treatment of choice for HR+ breast cancer in the early and advanced stages with multiple agents, including selective estrogen receptor modulators (SERMS), selective estrogen receptor degraders (SERDs), and aromatase inhibitors (AIs), which vary in their mechanisms of action and pharmacokinetics. Combination strategies also employ cyclin dependent kinase 4 and 6 and phosphatidylinositol 3-kinase to maximize the benefits of endocrine therapy. This paper reviews the clinical development of SERDs and other novel ER inhibitors, as well as combination strategies to overcome mechanisms of ER pathway escape. It also assesses the advantages of newer oral ER inhibitors with increased bioavailability, improved therapeutic index, better administration, and increased efficacy, as well as discussing future directions in the field.

Introduction

Approximately two-thirds of breast cancers express the estrogen receptor (ER), a key driver of proliferation, tumorigenesis, and progression [1]. Interfering with ER signaling via endocrine therapy, a successful strategy in the control of hormone receptor positive (HR+) breast cancer, is the mainstay of treatment for HR+ breast cancer in early and advanced stages of disease. Agents used in clinical practice include selective estrogen receptor modulators (SERMs), selective estrogen receptor degraders (SERDs), and aromatase inhibitors (AIs), which vary in their mechanisms of action and pharmacokinetics (PKs). More recently developed combination strategies involve cyclin dependent kinase 4 and 6 (CDK4/6) and phosphatidylinositol 3-kinase (PI3K) inhibitors, which are used to maximize and prolong the benefits of endocrine therapy. When disease progression occurs on first line therapy with CDK4/6 inhibitors, there are a lack of effective treatment options in the second line setting and beyond, especially in patients whose tumors do not harbor targetable alterations, with poor expected response rates to endocrine monotherapy. Novel, orally administered endocrine agents such as oral SERDs have been developed to overcome the pharmacodynamic and pharmacokinetic limitations of standard endocrine therapies [2].This review provides an overview of the pharmacology and clinical data of oral SERDs, along with our perspective on next-generation endocrine therapy and future directions in the field.

Pharmacological evolution from SERM to SERD

ER pathway and “old fashioned” endocrine agents

The addiction of breast cancer cells to estrogen is mediated by the ER pathway, which drives their survival and progression by regulation of the transcriptional process. The classical mechanisms of action of the ER pathway involves estrogen binding to the receptors, dimerization, and subsequent binding to DNA regulator regions called estrogen response elements, located in the promoters of specific genes. ERs can also regulate gene expression without directly binding to the DNA. This secondary mechanism of action occurs through protein–protein interactions with other transcription factors. Additionally, less common membrane-associated ERs mediate nongenomic actions of estrogens, which can affect cytoplasmic protein function and lead to gene expression modulation [3].

Endocrine therapy was developed to impair at different levels the ER pathway, one of the main players in the breast cancer cells survival and progression. The aromatization of A-cycle steroids leads to the conversion of androstenedione to estradiol in extra-ovarian tissue (e.g., adipose tissue), which is the main source of estrogen in post-menopausal women. By inhibiting this conversion, AIs reduce circulating estrogens and deprive the ER stimulus. In the adjuvant setting, AIs are the standard-of-care therapy for postmenopausal women and are often given in combination with ovarian function suppressors for premenopausal women at high risk of recurrence. AIs are approved to treat advanced breast cancer in combination with CDK4/6 inhibitors in the first-line setting and in combination with everolimus, an mTOR inhibitor, in subsequent lines of therapy [4].

Unlike AIs, SERMs and SERDs directly inhibit the ER. The two ER isoforms, ERα and Erβ, have different structures, affinities for circulating estrogens, and tissue distributions. All therapeutic and diagnostic strategies currently target ERα [5]. Tamoxifen, an SERM, is the forerunner of ER inhibitors and an antagonist of genes relying on the AF-2 region for ER-mediated transcription. Crystallography has shown that when tamoxifen binds the ER, the repositioning of helix 12 prevents binding of co-activators, thereby preventing AF-2-mediated transcription and inhibiting AF-2 activation. However, tamoxifen also functions as an agonist of genes whose transcription is driven by the AF-1 domain [6]. Tamoxifen has tissue selectivity because of the predominance of AF-1- or AF-2-activated genes in specific tissues (e.g., uterine and breast, respectively) and has thus been reclassified as a modulator of ERα rather than a pure ER antagonist. Tamoxifen competes with estrogens in ER binding and inhibits transcription in breast tissue, but acts as an ER agonist in other tissues, including endometrial tissue, resulting in a 2- to 3-fold increased risk of endometrial cancer (higher in post-menopausal patients), and an increased risk of venous thromboembolic disease. This toxicity profile combined with the inferior efficacy compared to AI, makes tamoxifen a second choice therapy for post-menopausal patients in both early and metastatic setting [7, 8]. The US Food and Drug Administration (FDA) and European Medicine Agency (EMA) approved indications for tamoxifen include treatment of advanced HR+ breast cancer, adjuvant treatment of other HR+ tumors, and risk reduction in patients with ductal carcinoma in situ (DCIS).

Fulvestrant, a SERD, is a pure ER antagonist that, when bound to the ER, reduces dimerization and transcriptional activation while accelerating degradation through a proteasome-dependent system. The proportion of ER-fulvestrant complex that migrates into the nucleus is transcriptionally inactive, with both AF-1 and AF-2 functions disabled. Fulvestrant has shown efficacy in patients who previously progressed on tamoxifen [9–11] and is the only FDA/EMA approved SERD for the treatment of metastatic HR+ breast cancer [12]. Fulvestrant monotherapy is approved as a 500 mg monthly intramuscular injection in treatment-naive patients based on the FALCON trial [13] and as second-line therapy after progression on tamoxifen or AI [10]. Fulvestrant is also approved in combination with the CDK4/6 inhibitors palbociclib, abemaciclib, and ribociclib in the first- or second-line setting, and in combination with the PIK3CA inhibitor alpelisib in PIK3CA-mutant advanced breast cancer after progression on first-line therapy. However, the lack of oral bioavailability of fulvestrant, and its modest clinical activity as monotherapy after disease progression on CDK4/6 inhibitors and alpelisib limits its clinical use in later lines.

CDK4/6 inhibitors in combination with endocrine therapies were established as first-line therapy for the treatment of metastatic HR+ disease after several studies demonstrated good rates of progression-free survival (PFS) and overall survival (OS). However, there are limited non-chemotherapy options after patients receive CDK4/6 inhibitors. Recent data suggests that in patients who progress on first line therapy with CDK4/6 inhibitors, a PFS of <2 months is observed with single agent fulvestrant [14–16]. Therefore, there is an unmet need to understand the mechanisms of resistance to endocrine therapy and find newer optimal endocrine agents that can overcome the pharmacologic liabilities and toxicity issues of currently approvents endocrine therapies. This in turn could facilitate sequential lines of endocrine therapy and delay time to chemotherapy administration.

Mechanisms of resistance to endocrine therapy

Endocrine therapy resistance is a complex phenomenon from both the clinical and molecular perspectives. The phenotype of a resistant cell can include mechanisms of ER reactivation or ER escape. Mechanisms of resistance to tamoxifen are commonly related to ER reactivation [17]. Prolonged exposure to tamoxifen can select for cancer cell subclones that express coregulators able to bind the otherwise inactive tamoxifen-ERα complex, resulting in ER agonism and resistance. Mutations in ESR1, the gene encoding ERα, are a well-established driver of resistance to endocrine treatment, through ER reactivation [18]. ESR1 mutations are rare in primary breast cancers but prevalent in metastatic cancers, occurring in 25% to 40% of patients previously treated with AIs, implying that these mutations are acquired [19, 20]. ESR1 mutations induce a conformational change that confers ligand independent aberrant ER transcriptional activity and decreases proteolytic degradation, thereby rendering AIs ineffective, with more variable effects on SERMs and SERDs [21, 22]. A combined analysis of the SoFEA and EFECT trials revealed that patients with ESR1 mutations detected on cell free DNA (cfDNA), in 30% of the overall study population, had improved PFS and OS when treated with fulvestrant versus exemestane, suggesting that unlike AIs, SERDs have activity in patients harboring ESR1 mutations [23]. The ESR1 mutations Y537S (14%) and D538G (36%) are the most prevalent mutations causing resistance to AIs. These mutations fall within the encoded ligand-binding domain of ESR1, and also reduce binding to tamoxifen and fulvestrant, thereby promoting potential resistance [18, 21, 24–26]. In clinical studies, while fulvestrant has been shown to be effective in patients whose tumors progressed on AI [27–29], data suggests that the development of an ESR1 Y537S mutation is a mechanism of acquired resistance to fulvestrant, thereby resulting in limited therapeutic activity [30]. Pre-clinically, Toy et al [31] have demonstrated that Y537S-mutant cells are less sensitive to fulvestrant and require higher drug levels to achieve antitumor activity compared to other ESR1 mutations. In cohort A of plasmaMATCH trial, higher dose fulvestrant did not confirm an increased response in patients with HR+/HER2- metastatic breast cancer carrying ESR1 mutations detected by cfDNA[32]. Recent updates presented at ASCO 2022 Annual Meeting identified a specific mutation, ESR1 F404, as a mechanism of acquired resistance to fulvestrant in patients with prior ESR1 activating alterations [33]. The in vitro modelling suggests that the presence of an ESR1 F404 interferes with fulvestrant-ER binding, whereas sensitivity to tamoxifen and novel SERDs appeared to be preserved. Overall, data suggests that the ESR1 mutation-based phenotype of resistance to AI or fulvestrant is variable and appears to depend on the type of mutation, the function of the involved domain, clonality, variant allele frequency in cfDNA and the presence of co-mutations in other driver pathways [34–36]. Given the limitations of currently approved endocrine therapies, a primary research focus is the development of oral SERDs with higher bioavailability and greater ER targeting and degradation capabilities [37].

New therapeutic strategies combine endocrine therapies with inhibitors of other prominent pathways associated with endocrine resistance and escape of ER-dependent cancer cells, including pathways involved in cell growth and cell cycle progression such as CDK4/6 and PI3K/Akt/mTOR. For instance, tumors with mutations in PIK3CA, reported in >40% of metastatic HR+ breast cancers, are vulnerable to PI3K inhibitors. Combination therapy is more effective than PI3K inhibitor monotherapy because inhibition of the PI3K pathway leads to a rebound of ER-dependent transcriptional activity [38]. Recently, new resistance of mechanisms have been identified, including activating ERBB2 mutations, NF1 loss-of-function mutations, and alterations in MAPK pathway genes (EGFR, KRAS) and in ER transcriptional regulators (MYC, CTCF, FOXA1, and TBX3) [39–43]. In a cohort of 692 pretreated breast tumors, Razavi et al. [44] reported that 22% of tumors had ER pathway escape alterations, which were mutually exclusive with ESR1 mutations and were associated with a shorter duration of response to subsequent hormonal therapies. These findings support the investigation of oral SERDs not only as monotherapy but also in combination strategies in early drug development.

New oral SERDs: mechanisms of action and clinical development

The development of oral SERDs has been challenging and the hope of improving efficacy with a more favorable mode of administration has largely fallen short. Challenges with earlier agents were due to the lack of competitive bioavailability and unacceptable toxicity profiles. Nonetheless, many oral SERDs and new ER inhibitors are under development, with several ongoing phase III trials. Metcalfe et al., demonstrated that ER degradation is the primary mechanism of action of both fulvestrant and oral SERDs; ER inhibition slows nuclear ER translocation and immobilizes ER in the nucleus, thereby inhibiting transcription and accelerating ER degradation [45].

The molecular interaction with ER differs across SERDs such that oral SERDs can be categorized based on their structure and mechanism of action (Figure 1). Some SERDs are non-steroidal molecules with an ER-binding site and a side chain of an acrylic acid, while others have an amino base side chain, inducing different conformational changes in ER that affect transcriptional activity and the vulnerability of the ER to degradation.

Figure 1: Interaction of new or traditional ER inhibitors with a breast cancer cell.

Estrogen receptors binds to estrogens, dimerize, and translocate to the nucleus. In the nucleus, the dimers bind to estrogen response element (ERE) of several genes and induce transcription, promoting cancer survival and proliferation. SERMs including tamoxifen, bazedoxifene, and lasofoxifene bind the ER, working as antagonists of the transcription process in the breast cancer cell and as agonists in other tissues (bone and endometrium). Fulvestrant works by inhibiting dimerization and translocation in the nucleus as well as by proteasomal degradation. In the nucleus, the transcription is not active in the absence of ER stimulus. Fulvestrant is partially active also on ESR1-mutant receptor (dotted line). ESR1, if not inhibited, is able to enhance the transcription in a ligand-independent fashion. Oral SERD, SERCA, and CERAN have the same effect as fulvestrant on ER, resulting in the inhibition of transcription, but they are more potent on ER as well as ESR1, with a higher rate of degradation of the receptors. The elements that confer more potency to these agents compared to fulvestrant is reported in the figure. PROTAC is composed of a ligand that binds to the E3 ubiquitin ligase and a ligand that binds to the target protein through a linker, which can induce the polyubiquitination and proteasomal degradation of ER in the cancer cells.

SERDs with acrylic acid side chains

SERDs with an acrylic acid side chain, including GW5638 (etacstil), with its active metabolites GW7604, GDC0810, AZD9496, G1T48, and LSZ102, were the first potent antiestrogen agents developed without cross-resistance to tamoxifen (Table 1). The crystallographic analysis of the ERα-GW7604 complex showed that the interaction of a carboxylic acid with the peptide backbone of ER induced a conformational change that exposed a hydrophobic surface of the receptor that targeted it for proteolytic degradation [46]. As monotherapy, these agents have antitumor activity in both endocrine-sensitive and resistant preclinical models and in ESR1-mutated tumors [31, 37, 47–49]. Further preclinical studies demonstrated potent antitumor activity of these agents when combined with CDK4/6 inhibitors. For example, G1T48 (rintodestrant), in combination with the novel CDK4/6 inhibitor G1T38 (lerociclib), has greater tumor growth inhibition compared to either agent as monotherapy [50, 51].

Table 1:

Pertinent clinical data from phase I trials of oral SERDs and other novel compounds

Investigational drug +/− other agents	Type of inhibitor	Phase	N	ESR1 mut (%)	Prior CDK4/6i (%)	Prior SERD (%)	ORR (%)	CBR (%)	Median PFS	*Any grade TEAE (%)	Grade ≥3 TEAE or DLT (%)	Reference
Acrylic Acid Side Chains
AZD9496	SERD	I	45	NA	15.6	55.6	2	8	NA	Diarrhea (35.6), Fatigue (31.1)	15	Hamilton, CCR 2018
Rintodestrant (G1T48)	SERD	I (part A and B)	67	45	70	64	5	30	2.6–3.6	Hot flush (24), Fatigue (21)	NA	Aftimos, SABCS 2020
Rintodestrant (G1T48) + palbociclib	SERD + CDK4/6i	I (part C)	40	NR	0	15	5	61	NR	Neutropenia (88)	Neutropenia (43)	Maglakelidze, ASCO 2021
LSZ102	SERD	I: Arm A	77	42	58	60	1.3	9.1	1.8	Nausea (61), Diarrhea (55)	NA	Jhaveri, CCR 2021
LSZ102 + ribociclib	SERD + CDK4/6i	I: Arm B	78	39	34.6	60.3	16.9	35.1	6.2	Nausea (50), Diarrhea (39)	NA	Jhaveri, CCR 2021
LSZ102 + alpelisib	SERD + PIK3CAi	I-Arm C	43	25	65	46	7	20.9	3.5	Nausea (60), Diarrhea (70)	NA	Jhaveri, CCR 2021
Basic Side Chain
Elacestrant (RAD1907)	SERM / SERD	I	50	50	52	52	19.4	42.6	4.5	Nausea (50), Dyspepsia (32)	41.7	Bardia et al., JCO 2021
Giredestrant (GD-9545)	SERD	I	111	NR	64	21	15	50	7.2	Fatigue (21), Arthralgia (17)	5	Jhaveri et al., ASCO 2021
Giredestrant (GDC9545) + palbociclib	SERD	I	48	NR	0	7	33	81	NR	Neutropenia (50)	50	Lim et al., ASCO 2020
Amcenestrant (SAR439859)	SERD	I (part A+B)	59	28	63	46.8	8.5	33.9	NR	Hot Flush (16), Constipation (9.7), Arthralgia (9.7)	NA	Linden, SABCS 2020; Campone et al., ASCO 2020
Amcenestrant (SAR439859) + palbociclib	SERD + CDK4/6i	I (part C+D)	39	NR	5.1	7.7	32.4	73.5	Not reached	Nausea (17.9), Fatigue (17.9%)	46.2	Chandarlapaty et al., ASCO 2021 and SABCS 2021
Camizestrant (AZD9833)	SERD	I	98	46	50	53	10	35.3	NR	Visual disturbances, Bradycardia, Asthenia, Anemia, QTcF prolongation	DLT: QTcF, vomiting); visual disturbance	Hamilton et al., ASCO 2021 Baird et al., SABCS 2021
Camizestrant (AZD9833) + palbociclib	SERD	I	25	11	17	20	5.9	28	NA	Infections, Neutropenia	NA	Oliveira et al., ASCO 2022
Imlunestrant (LY3484356)	SERD	I	141	38	92	51	12	55	4.3	Nausea (35), Fatigue (25),	NA	Jhaveri, ASCO 2021 and 2022
Others
D0502	SERD	I	16	NR	NR	NR	10	50	NA	NA	NA	Osborne, SABCS 2020
ZN-c5	SERD	I	56	41	70	46	5	38	3.8	Nausea, Fatigue, Arthralgia	No DLT	Kalinsky, SABCS 2021
OP1250	CERAN	I/II	32	16	92.5	67.9	9	21	NA	Nausea(59), Fatigue(35), Constipation (24)	No DLT	Patel, SABCS 2021
HB36545	SERCA	II	94	62	87	73	16.7	40.3	3.8	Sinus bradicardia (35), Anemia(19), Fatigue(16), Diahrraea (12)	Sinus bradicardia G2 (5) and QT prolungation G2/3	Hamilton ASCO 2021 and SABCS 2021
ARV471	PROTAC	I	60	NA	100	50	6	41	NA	Nausea (24), Fatigue (12), Vomiting (10)	No grade 3 AE	Hamilton et al., SABCS 2021

^*TEAE: the two most common TEAE occurring in ≥15% of all patients.

ASCO: American Society of Clinical Oncology; CBR: clinical benefit rate; CCR: Clinical Cancer Research; CDK4/6i: CDK4/6 inhibitor; CERAN: Complete ER Antagonists; DLT: dose limiting toxicity; eBC: early breast cancer; JCO: Journal of Clinical Oncology; mutn: mutation; NA: not applicable; NR: not reported; ORR: overall response rate; PROTAC: PROteolysis TArgeting Chimeras; PFS: progression free survival; SABCS: San Antonio Breast Cancer Symposium; SERCA: Selective Estrogen Receptor Covalent Antagonists; SERD: selective estrogen receptor downregulator; TEAE: treatment emergent adverse event.

Early-phase clinical trials of oral SERDs with acrylic acid side chains showed less promising efficacy and tolerability results, so most of these compounds have not been further developed beyond phase I. For instance, GDC0810, a direct successor of GW5638, was less active than fulvestrant in a phase II study that was terminated early (NCT02569801). In a phase I/Ib trial of patients who had progressed on prior endocrine therapy, the majority of whom had received prior fulvestrant, CDK4/6 inhibitor, and chemotherapy, LSZ102 monotherapy resulted in an overall response rate (ORR) of only 1%, which increased to 17% and 7% when LSZ102 was given in combination with ribociclib or the PI3K inhibitor alpelisib, respectively [52]. LSZ102 monotherapy was associated with high rates of gastrointestinal (GI) toxicities experienced: any grade nausea (61%) and diarrhea (55%). Notably, the toxicity profile did not significantly differ in the combination arms [53].

Rintodestrant was evaluated in a phase I trial as monotherapy and in combination with palbociclib (NCT03455270). Preliminary results from 14 patients treated with G1T48 alone demonstrated reductions in ¹⁸F-fluoroestradiol positron emission tomography uptake ranging from 70% to 98% after 4 weeks of treatment. However, among a larger group of 67 patients treated with monotherapy, 20 patients were on study >24 weeks but only 3 patients experienced partial response (PR), with a median PFS of 2 months. Similar response rates were observed when G1T48 was combined with palbociclib [54, 55]. The toxicity profile appears to be favorable, with <10% of patients experiencing grade 1–2 nausea, vomiting, and neutropenia. Results from the other combination cohorts are awaited.

AZD9496 was designed with a novel binding motif of substituted aryl indole to improve affinity for either wild-type or mutant ER, but showed low response rates and some toxicity. Approximately 90% of patients experienced drug-related adverse events, including 40% with any grade diarrhea or other GI toxicity [56]. In February 2021, AZD9496 development was halted in favor of its more potent and tolerable successor AZD9833 (camizestrant), developed with an amino side chain instead of an acrylic acid side chain. Camizestrant rapidly progressed through phase I testing and is currently under investigation in phase II and phase III clinical trials, as discussed below.

SERDs with basic amino side chains

In addition to unfavorable toxicity profiles, SERDs with acrylic acid side chains were found to inconsistently inhibit ER. To overcome these challenges, SERDs with basic amino side chains were developed in an attempt to maximally and consistently degrade ER [2], including RAD1901 (elacestrant), GDC9545 (giredestrant), SAR439859 (amcenestrant), AZD9833 (camizestrant) and LY3484356 (imlunestrant) (Table 1). These agents have demonstrated good oral bioavailability and significant antitumor activity in both treatment-naive and endocrine-resistant preclinical models, including those with ESR1 mutations and those resistant to CDK4/6 inhibition, demonstrating growth inhibition comparable to fulvestrant [45, 57–65]. Notably, these agents may treat brain metastases from HR+/HER2- breast cancers, as evidence suggests that elacestrant crosses the blood-brain barrier. While it is unclear whether ER inhibition is as effective in the central nervous system (CNS) as in other disease sites [66], a phase Ib/II trial of elacestrant plus abemaciclib in patients with active CNS metastases is ongoing (NCT04791384).

The maximum tolerated dose (MTD) was not achieved in any trial of novel oral SERDs with basic amino side chains, resulting in the recommended phase II dose (RP2D) being chosen from efficacy, PK, pharmacodynamic, and safety data from phase I studies. Ongoing phase III trials will determine if these agents can replace traditional endocrine therapies in both early-stage and advanced disease (Table 2).

Table 2:

Ongoing trials in breast cancer

Drug	Clinical Trials Identifier	Phase	Experimental arm	Control arm	Population characteristics
First-line metastatic disease
Amcenestrant (SAR439859)	NCT04478266 (AMEERA-5)	III	SAR439859 + palbociclib	Letrozole + palbociclib	No previous treatment for metastatic disease; no previous SERDs
Camizestrant (AZD9833)	NCT04964934 (SERENA-6)	III	AZD9833 + palbociclib/abemaciclib	AI + palbociclib/abemaciclib	MBC ER+/HER2- on AI + palbo/abema with detectable ESR1 mutation without radiological POD on imaging
-	NCT04711252 (SERENA-4)	III	AZD9833 + palbociclib	Letrozole + palbociclib	De novo Stage IV or relapse after 24 mo of adjuvant AI and/or within 12 mo after discontinuation
Giredestrant (GDC9545)	NCT04546009 (persevERA)	III	GDC9545 + palbociclib	Letrozole + palbociclib	De novo Stage IV or relapse after 24 mo of adjuvant AI and/or within 12 mo after discontinuation
Imlunestrant (LY3484356)	NCT04188548 (EMBER-1)	Ia/Ib	LY3484356 + abemaciclib +/− AI or LY3484356 + everolimus or LY3484356 + alpelisib or LY3484356 + trastuzumab +/− abemaciclib	-	MBC ER+/HER2- or ER+/HER2+
ZN-c5	NCT04514159	I	ZN-c5 + abemaciclib	-	No prior CDK4/6 I
-	NCT03560531	I/II	ZN-c5 +/−palbo	-	MBC ER+/HER2-
Beyond first-line metastatic disease
Elacestrant (RAD1901)	NCT04791384	Ib/II	Elacestrant + abemaciclib	-	Progressing brain mets in patients pretreated with ET and maximum 2L chemo
Rintodestrant (G1T48)	NCT03455270	I	Rintodestrant +/− palbociclib	-
Amcenestrant (SAR439859)	NCT04059484 (AMEERA-3)**	II	Amcenestrant	ET monotherapy (any AI, fulv or Tam)	post CDK4/6i, secondary endocrine resistance*
-	NCT03284957 (AMEERA-1)	I/II	SAR439859 + /-everolimus or palbociclib or abemaciclib or alpelisib	-	secondary endocrine resistance*
Camizestrant (AZD9833)	NCT03616587 (SERENA-1)	I	AZD9833 +/− everolimus or palbociclib or abemaciclib or alpelisib or capivasertib	-	Preexposure to 1 line ET (different from SERD), only 1 line of chemo allowed,
-	NCT04214288 (SERENA-2)	II	AZD9833	Fulvestrant	Preexposure to ET, no limits of lines; only 1 line of chemo allowed
Giredestrant (GDC9545)	NCT04576455 (acelERA BC)**	II	GDC9545	ET monotherapy (any AI or fulv)	Progression after 1–2 L of ET
-	NCT04802759	Ib/II	GDC9545 + inavolisib (PIK3CAi), ipatasertib (CDK7i), abemaciclib, ribociclib, everolimus	-	MBC ER+/HER2 progression after 1–3L of ET including CDK4/6i
Imlunestrant (LY3484356)	NCT04188548 (EMBER)	Ia/Ib	LY3484356 +/− AI or everolimus or abemaciclib or alpelisib or capivasertib or trastuzumab or pertuzumab	-	MBC ER+/HER2- or ER+/HER2+ and metastatic endometrial cancer
-	NCT04975308 (EMBER-3)	III	LY3484356	AI or fulvestrant	POD after AI +/− CDK4/6 I
D-0502	NCT03471663	Ia/Ib	D-0502 +/− palbociclib	-	MBC ER+/HER2-
Zn-c5	NCT04176757	I	Zn-c5	-	MBC ER+/HER2-
Zb-716	NCT04669587	1/II	ZB716 + palbociclib		MBC ER+/HER2-
Neoadjuvant setting
Camizestrant (AZD9833)	NCT04588298 (SERENA-3)	II	AZD9833	Fulvestrant	eBC ER+/HER2-, palpable tumor of any size, or T ≥ 1.0 cm
Elacestrant (RAD1901)	NCT04797728 (ELIPSE)	II	Elacestrant	-	eBC ER+/HER2-, T ≥1.5 cm (cT1c-3), N0
Imlunestrant (LY3484356)	NCT04647487 (EMBER-2)	II	LY3484356	-	eBC ER+/HER2-, stage I-III
Giredestrant (GDC9545)	NCT03916744	II	GDC9545 +/− palbociclib	Anastrozole + /-palbociclib	eBC ER+/HER2-, stage I-III
Amcenestrant (SAR439859)	NCT01042379 (I-SPY)	II	SAR439859 alone SAR439859 + AI SAR439859 + abemaciclib	Standard therapy	eBC ER+/HER2-, stage I-III
Adjuvant setting
Giredestrant (lidERA)	NCT04961996 (GDC9545)	III	Giredestrant alone	Physician choice endocrine treatment	eBC ER+/HER2- after definitive surgery
Amcenestrant (AMEERA-6)	NCT05128773 (SAR439859)	III	Amcenestrant alone	Tamoxifen	eBC who have discontinued adjuvant AI due to treatmentrelated toxicity
New ER targeting agents
H3B6545 (SERCA)	NCT04288089	I	H3B6545 + palbociclib	-	MBC ER+/HER2- after 1L
ARV471 (PROTAC)	NCT04072952	I/II	ARV-471 (PROTAC) +/− palbo	-	MBC ER+/HER2-
OP1250 (CERAN)	NCT04505826	I/II	OP-1250	-	MBC ER+/HER2-
Bazedoxifene (SERM/SERD)	NCT02448771	I/II	Bazedoxifene + palbociclib	-	MBC ER+/HER2- after 1L
Lasofoxifene	NCT03781063 (ELAINE)	II	Lasofoxifene	Fulvestrant	MBC ER+/HER2- with ESR1 mutn after POD to CDK4/6i + AI
	NCT04432454 (ELAINE-2)	II	Lasofoxifene + abemaciclib	-	MBC ER+/HER2- with ESR1 mutn after POD to CDK4/6i + AI in 1st or 2nd line

^*Secondary endocrine resistance: progression while on ET after at least 6 months of treatment for MBC, or relapse while on adjuvant endocrine therapy but after the first 2 years, or with a relapse within 12 months after completing adjuvant ET.

AI: aromatase inhibitor; eBC: early breast cancer; ET: endocrine therapy; fulv: fulvestrant; L: lines; MBC: metastatic breast cancer; mo: months; mutn: mutation; N: nodal status; POD: progression of disease; T: tumor size; Tam: tamoxifen; WoO: window of opportunity

^**Active, not recruiting. Press release on primary enpoint is available. See reference [78] for acelERA trial and [88] for AMEERA-3 trial, respectively.

Elacestrant is the agent furthest along in development, with early clinical trials demonstrating its safety, tolerability, good oral bioavailability, and ability to penetrate the blood-brain barrier [66, 67]. In a phase I study of 57 heavily pretreated patients by Bardia et al., the ORR was 19%, and response rates were 15% in those treated with prior SERD, 17% in those treated with prior CDK4/6 inhibitors, and 33% in those with an ESR1 mutation. The clinical benefit rate (CBR) at 24 weeks was 43%, including 57% in those with an ESR1 mutation and 30% in those treated with prior CDK4/6 inhibition [68, 69]. The reported toxicities were predominantly GI (nausea [50%], dyspepsia [32%], and vomiting [20%]), the overwhelming majority of which were grades 1–2. The selected RP2D was 400 mg daily.

The randomized EMERALD study was the first phase III of an oral SERD; 477 patients previously treated with CDK4/6 inhibitors for HR+ metastatic breast cancer, were randomized to elacestrant versus endocrine monotherapy (fulvestrant or an AI). One prior chemotherapy line was permitted. The primary endpoints were PFS in the overall population and PFS in patients with tumors harboring ESR1 mutations. Both primary endpoints were met. In the intention-to-treat analysis, there was a 30% reduction in the risk of progression or death versus fulvestrant or AI and a 45% reduction in the risk of progression or death in patients harboring ESR1 mutations [15]. The 12-month PFS was 22.3% in the elacestrant arm versus 9.4% in the control arm with a HR of 0.7 (0.55–0.88). A greater magnitude of benefit was shown in the subgroup of patients with tumors harboring an ESR1 mutation (12-month PFS: 26.6% versus 8.2% for ESR1 mutated and wildtype respectively). Although the absolute PFS benefit was small in this study, (2.8 versus 1.9 months in the overall population and 3.8 versus 1.9 months in the ESR1-mutant population), the PFS curves between these populations significantly diverge at 6 months and 12 months after an initial dramatic drop in both treatment arms [15]. Similar outcomes were reported in the subgroup analysis of patients who received prior endocrine therapy only [70]. One limitation of the EMERALD study was that 20–25% of the patients had received prior chemotherapy. In addition, approximately 30% of the study population had received prior fulvestrant and 80% had received prior AI, which may account for the initial dramatic drop seen in the PFS curves and the short PFS observed in both arms. These findings suggest that elacestrant benefit may be driven by a group of patients with endocrine sensitive tumors. The most common any grade adverse effects observed in ≥15% of patients treated with elacestrant were nausea (35%), vomiting (19%), and fatigue (19%). This study represents the first positive phase III trial of an oral SERD, and approval of elacestrant is currently under review by the FDA. Additional data on the elacestrant efficacy will be provided by the ongoing phase Ib/II trial evaluating elacestrant in combination with CDK4/6 inhibitor in patients with progressing brain metastasis (NCT04791384).

Giredestrant has been shown to be active as monotherapy and in combination with CDK4/6 inhibitors. In a phase I trial, 111 patients, >60% of whom had previously been treated with CDK4/6 inhibitors, received monotherapy, and 48 patients naive to CDK4/6 inhibitors were treated with giredestrant plus palbociclib. The CBR was 50% with monotherapy (PR, 13%) and 81% with combination therapy (PR, 33%) [71, 72]. Grade ≥3 toxicities were observed infrequently (neutropenia [33%], fatigue [2%], and diarrhea [2%]). The phase II acelERA breast cancer trial evaluating giredestrant versus endocrine monotherapy in patients previously treated with 1–2 prior lines of systemic therapy for metastatic disease (NCT04576455) failed to demonstrate superiority of giredestrant over standard of care (Hoffmann-La Roche press release, April 2022 [73]). The persevERA study is an ongoing phase III trial evaluating the efficacy of giredestrant combined with palbociclib in the first-line metastatic setting (NCT04188548).

Giredestrant has also been studied in early-stage breast cancer with encouraging results. The first trial of predominantly luminal A breast cancers, evaluated by PAM50 criteria, studied single agent neoadjuvant giredestrant with an interim analysis, showing that giredestrant reduced Ki67 in 78% of tumors and caused complete cell cycle arrest in 55% of tumors (NCT03916744) [74]. The coopERA trial is a randomized, phase II neoadjuvant trial of giredestrant versus anastrozole, with the addition of palbociclib in both arms after 14 days of treatment. The final analysis on 201 evaluable patients on the lead-in phase of the study (without palbociclib) has shown a statistically significant reduction in Ki67 after 14 days of treatment with giredestrant compared to AI (relative reduction in geometric Ki67 of −75% vs −67%, respectively) [75]. However, giredestrant combined with palbociclib did not show an improvement in terms of pCR rate compared to the control arm. The lidERA randomized phase III trial is currently investigating the benefits of giredestrant compared to standard-of-care endocrine therapy in early-stage HR+, human epidermal growth factor receptor 2 negative (HER2-) breast cancer after definitive surgery (NCT04961996).

In the ongoing phase I SERENA-1 study (NCT03616587), camizestrant is being studied as monotherapy and in combination with palbociclib or abemaciclib (CDK4/6 inhibitors), everolimus (mTOR inhibitor), or capivasertib (AKT inhibitor). Among 98 patients treated with camizestrant monotherapy, in those previously treated with fulvestrant (53%) and CDK4/6 inhibitors (50%), the ORR and CBR were 10% and 35%, respectively. In the CDK4/6 inhibitor-naive cohort including 25 patients receiving camizestrant (RP2D:75 mg) plus palbociclib, the ORR and CBR were 5.9% and 28%, respectively. Of the patients treated with camizestrant monotherapy, 46% had baseline cfDNA-detected ESR1 mutations, of whom 50% achieved PR or stable disease at 24 weeks and 85% had reductions or loss of mutant ESR1 with treatment [76]. Efficacy data from the other combination arms are awaited. Any grade toxicities reported in >15% of patients were visual disturbances (53%), bradycardia (45%), and nausea (18%) [77, 78]. The RP2D of 75 mg has been brought forward to the phase II setting, with the ongoing SERENA-2 study (NCT04214288) evaluating camizestrant versus fulvestrant, and to the phase III setting, with the SERENA-4 and SERENA-6 studies (NCT04711252; NCT04214288) evaluating camizestrant plus CDK4/6 inhibitor in the first-line metastatic setting.

Amcenestrant is another new molecule under investigation as monotherapy and in combination with targeted therapies. In the phase I/II AMEERA-1 study, among 59 evaluable patients treated with monotherapy, the ORR was 9% and the CBR was 34%, increasing to 64% among patients without prior SERD, CDK4/6 inhibitors, or mTOR inhibitors [79]. Among 39 patients treated with amcenestrant plus palbociclib, the ORR was 32% and CBR at 24 weeks was 73%, with a median PFS of 15 months (range 11–22.3) [80, 81]. The most frequent any-grade toxicities occurring in ≥15% of patients were fatigue and nausea; 13% of patients experienced grade ≥3 toxicities with amcenestrant alone [79, 82]. The RP2D of 400 mg daily is now being investigated in the first-line setting in combination with CDK4/6 inhibitors (NCT04478266). A recent press release reported that the phase II AMEERA-3 trial (NCT04059484) which studied patients progressing on endocrine therapy +/− CDK4/6 inhibitors (mandatory for 80% of the population) did not met its primary endpoint of PFS superiority favoring amcenestrant over the control arm (Sanofi press release, March 2022) [83]. Similarly, the AMEERA-4, a window of opportunity trial for patients with I-III early breast cancer showed antiproliferative activity of amcenestrant, measured as changes in Ki67 from baseline to day 14, but there were no differences conmpared to the control arm of AI [84]. More recently, the phase III AMEERA-6 trial evaluating amcenestrant versus tamoxifen as adjuvant treatment for patients intolerant to AI is open to enrollment (NCT05128773).

The EMBER study, the first-in-human trial of imlunestrant, showed encouraging results in 114 heavily pretreated patients with metastatic HR+/HER2- breast cancer with a median PFS of 4.3 months (range 3.6–7.1 months). Interestingly, in 45 patients who received imlunestrant as second line therapy after CDK4/6 inhibitors, the median PFS was 6.5 months (range 3.6–8.3 months), longer than the expected PFS with currently available standard endocrine therapy[14–16]. Predominantly grade 1–2 adverse events were reported, with nausea and diarrhea being the most frequently reported toxicities [85, 86]. Results of the other combination arms with everolimus, abemaciclib, trastuzumab and pertuzumab are awaited. The dose of 400 mg daily is the RP2D and has been used in the design of phase I-III trials. The EMBER-3 phase III trial has been designed to evaluate the efficacy of imlunestrant alone versus imlunestrant with abemaciclib versus AI/fulvestrant in the second line setting (NCT04975308). Prior chemotherapy and fulvestrant are not permitted. Other ongoing trials are investigating the activity of imlunestrant in early-stage disease given in neoadjuvant (EMBER-2, NCT0464748) and adjuvant setting (upcoming EMBER-4). Several other molecules, including D-0502, Zb-716, ZN-c5, and SHR9549, demonstrated promising preclinical data and are under investigation in phase I and II clinical trials (Table 2).

Innovation in ER inhibition: next steps beyond SERDs

Other strategies aimed at targeting ER inhibition include PROteolysis Targeting Chimeras (PROTACs), selective estrogen receptor covalent antagonists (SERCAs), complete ER Antagonists (CERANs), and new oral SERM/SERD hybrids that are direct successors of tamoxifen. These new molecules were developed with the hope of increasing the potency of ER inhibition/degradation. SERCAs inactivate both wild-type and mutant ER receptors by targeting a cysteine residue (C530), thereby enforcing a unique antagonist conformation [2]. H3B6545, a SERCA, showed anti-tumor activity both as monotherapy and in combination with CDK4/6 and mTOR inhibitors in the preclinical setting [87]. H3B6545 is now being investigated in phase I/II trials both as monotherapy (NCT03250676; NCT04568902) and in combination with palbociclib (NCT04288089). Preliminary results demonstrated that H3B6545 monotherapy resulted in an ORR of 30% among patients with tumors harboring clonal ESR1 Y537S mutations, compared to 17% in the overall population, with CBR of 70% and 40%, respectively. Cardiovascular toxicities included grade 1 asymptomatic bradycardia (36%), grade 2 symptomatic bradycardia (5%), and grade 2/3 QT prolongation (3%) [88].

Unlike partial agonists, CERANs bind the ligand binding pocket, thereby blocking ER transcriptional activity mediated by both AF1 and AF2 to achieve complete antagonism. OP1250 is a first-in-class CERAN that showed an ORR of 9% when given as monotherapy in a phase I trial. Low-grade nausea was reported in 60% of the population, but there were no dose-limiting toxicities observed (NCT04505826) [89].

PROTACs are bifunctional small molecules that link an ER binding ligand with an E3 ligase to induce polyubiquitination and degradation of the ER complex. ARV471 is a first-in-class PROTAC targeting the ERα receptor and showed promising results when compared to fulvestrant, and in combination with palbociclib in the preclinical setting [90, 91]. A phase I/Ib clinical trial is currently evaluating ARV471 alone and in combination with palbociclib (NCT04072952). Preliminary results have shown favorable toxicity profile: there were no cases of bradycardia, no grade 4 toxicities, and only 2 of 50 patients experienced grade 3 adverse events. The CBR was 40% in 47 evaluable patients, with responses reported in patients previously treated with fulvestrant, CDK4/6 inhibitors, and oral SERDs [92]. These results suggest intriguing activity of PROTACs after progression on oral SERDs, which may be important in the future development of this new class of drug.

SERM/SERD hybrids have demonstrated potent Erα antagonism and degradation. Lasofoxifene is a next-generation non-steroidal SERM that was developed to treat vulvovaginal atrophy and osteoporosis. In preclinical models, lasofoxifene monotherapy was more effective than fulvestrant at inhibiting primary tumor growth and reducing disease progression. In xenograft models, the addition of a CDK4/6 inhibitor to lasofoxifene improved its activity in tumor suppression and metastases prevention [93]. This strong preclinical data supports the phase II ELAINE and ELAINE2 trials that are investigating lasofoxifene alone and in combination with abemaciclib, respectively (NCT03781063, NCT04432454).

Bazedoxifene (BZA) is another SERM/SERD hybrid with strong antagonist and SERD profiles in breast tissue and concomitant agonist properties in the bone, but without agonist properties in the endometrium [62, 94–96]. With long-term safety data in thousands of patients, BZA is already approved in the US as a hormone replacement therapy and is approved in Europe for the prevention of osteoporosis [62, 94, 96–99]. In breast cancer, BZA demonstrated good oral bioavailability and improved PK profile compared to fulvestrant, as well as potent anti-tumor activity in both AI and SERM-resistant tumors and in ESR1-mutated cells [94, 97, 98, 100]. Preliminary results from a phase Ib/II study of BZA with palbociclib has shown an ORR of 8%, CBR of 36%, and PFS of 4 months in a heavily pretreated population. Final results from this study are expected towards the end of 2022 (NCT02448771) [101]. Additionally, BZA conjugated with estrogens is under investigation as primary prophylaxis for patients at high risk of developing breast cancer and in patients with DCIS (NCT04821141; NCT02694809).

Future directions

The drug development platform of oral SERDs has rapidly progressed from preclinical studies to phase III trials both in late and early-stage settings. The phase III EMERALD trial has recently provided proof-of-principle for activity of oral SERDs compared to physician’s choice endocrine therapy after progression on CDK4/6 inhibitors in the metastatic setting, particularly in patients whose tumors harbor ESR1 mutations. The oral bioavailability of these agents, in addition to their favorable toxicity profiles, may be particularly pertinent when considering their role in combination with other targeted therapies, including CDK4/6 inhibitors and agents targeting the PI3K/Akt/mTOR pathways. Although GI toxicities including nausea, vomiting and diarrhea occur frequently, they are typically low grade and are rarely observed as grade 3–4 toxicities. There are no major overlapping toxicities between SERDs and targeted therapies. Diarrhea has been observed mainly for SERDs with acrylic acid side chains and therefore may present challenges when given in combination with abemaciclib. However, this combination might be still potentially manageable and tolerable with dose reductions and/or alternative schedules. Hence, multiple phase III trials are ongoing and evaluating the role of oral SERDs plus CDK4/6 inhibitors in the first-line metastatic setting. The PADA1 trial demonstrated that switching from an AI plus CDK4/6 inhibitor to fulvestrant plus CDK4/6 inhibitor in the first-line setting doubles PFS in those patients who develop a cfDNA-detected ESR1 mutation prior to radiological progression of disease [102]. This supports the idea that oral SERDs may be beneficial when used early, especially in patients refractory to currently approved endocrine therapies and in those who develop ESR1 mutations. The ongoing phase III SERENA-6 trial (NCT04964934) will evaluate if switching from standard endocrine therapy to an oral SERD at first appearance of an ESR1 mutation can prevent clinical disease progression in patients on first-line therapy with CDK4/6 inhibitors. The current challenge is how to best use these agents to optimize treatment outcomes for patients. To that end, we eagerly await results from ongoing studies of other oral SERDs to determine if these agents should be best utilized as single agents or in combination with targeted therapies post-progression on CDK4/6 inhibitor therapy or may be most effective in the first-line metastatic setting in combination with CDK4/6 inhibitors.

Despite oral SERDs not yet obtaining regulatory approval for metastatic disease, trials in the early-stage setting are already in progress (Table 2). Given their high potency and possibility for dose reduction, they have the potential to increase patients’ adherence and tolerance of treatment both in the metastatic but more importantly in the early stage setting where non-adherence is an issue [103]. Abemaciclib was recently FDA approved as adjuvant therapy for high-risk, early-stage ER+ breast cancers based on the results of the MonarchE trial [104]. However, despite the significant reduction in both disease-free and relapse-free survival with 2 years of adjuvant abemaciclib, this agent is associated with high rates of GI toxicities. Given the excellent tolerability and relatively low rates of toxicities seen with newer-generation oral SERDs, these agents may be particularly promising in early-stage disease to change the landscape of adjuvant endocrine therapy. However, there are many unanswered questions, including whether oral SERD monotherapy might be sufficient treatment for high-risk disease, and whether it is necessary to combine oral SERDs with abemaciclib in the adjuvant setting. With respect to the optimal sequence in early-stage disease, questions remain about whether oral SERDs should be used as upfront therapy versus tamoxifen/AI, whether they should be used as part of a switch strategy after patients have been treated with at least 2 years of standard-of-care endocrine therapy, or whether oral SERDs may function as extended endocrine therapy.

In conclusion, oral SERDs and other new ER inhibitors represent a promising advancement in drug development for ER+ breast cancers. There is high expectation in the scientific community that these agents may be soon included in the treatment paradigm for metastatic HR+/HER2- breast cancer. The big question ahead of us remains how best to optimally sequence these agents in order to improve outcomes for our patients. While these agents are active against ESR1 mutations, other agents are needed that will tackle the ligand binding domain mutation independent mechanisms of endocrine resistance such as ARID1A mutations and ESR1 fusion. With the currently available data, single agent SERDs have not been shown to provide a prolonged and meaningful benefit post CDK4/6 inhibitor therapy. Better results are anticipated when SERDs are used in earlier lines and/or as a backbone for combination therapies. Our hope is that the introduction of novel endocrine therapy options with increased potency and favorable toxicity profiles into clinical practice may contribute to prolonging the chemotherapy-free interval as much as possible for our patients, thereby facilitating more targeted and personalized treatment approaches, especially in the context of endocrine resistance.

Highlights.

• New ER inhibitors show improved pharmacokinetics and pharmacodynamics compared to fulvestrant

• Composition of the side chain of oral SERDs affects their efficacy for efficacy

• Five oral SERDs are in phase III trials in the advanced setting and two in the adjuvant setting

• New ER inhibitors/degraders under investigation: SERCA, CERAN, PROTAC, and SERM/SERD hybrid

• Oral route and potency render the agents ideal endocrine backbone for combination therapies