Latest generation estrogen receptor degraders for the treatment of hormone receptor-positive breast cancer

ABSTRACT

Introduction

The selective estrogen receptor degrader (SERD) and full receptor antagonist provides an important therapeutic option for hormone receptor (HR)-positive breast cancer. Endocrine therapies include tamoxifen, a selective estrogen receptor modulator (SERM), that exhibits receptor agonist and antagonist activity, and aromatase inhibitors that block estrogen biosynthesis but which demonstrate acquired resistance. Fulvestrant, the only currently approved SERD, is limited by poor drug-like properties. A key focus for improving disease management has been development of oral SERDs with optimized target occupancy and potency and superior clinical efficacy.

Areas covered

Using PubMed, clinicaltrials.gov, and congress websites, this review explored the preclinical development and clinical pharmacokinetics from early phase clinical studies (2015 or later) of novel oral SERDs, including giredestrant, amcenestrant, camizestrant, elacestrant, and rintodestrant.

Expert opinion

Numerous oral SERDs are in clinical development, aiming to form the core endocrine therapy for HR-positive breast cancer. Through property- and structure-based drug design of estrogen receptor-binding, antagonism, degradation, anti-proliferation, and pharmacokinetic properties, these SERDs have distinct profiles which impact clinical dosing, efficacy, and safety. Assuming preliminary safety and activity data are confirmed in phase 3 trials, these promising agents could further improve the management, outcomes, and quality of life in HR-positive breast cancer.

KEYWORDS:

• Amcenestrant
• breast cancer
• camizestrant
• elacestrant
• endocrine therapy
• fulvestrant
• giredestrant
• oncology
• rintodestrant
• SERD

1. Introduction

Breast cancer remains one of the most significant public health problems, with an increasing global incidence [1]. It is the most common cancer among women, representing 24.5% of all newly diagnosed cases, and is currently the leading cause of cancer death among women worldwide, accounting for 15.5% of all cancer deaths [2]. Breast cancer is a highly heterogeneous disease, distinguished by different molecular subtypes, risk factors, clinical behaviors, and responses to treatment [3]. Of the clinically annotated subtypes, hormone receptor (HR)-positive breast cancers are the most common [4], characterized by positive immunohistochemical staining for the estrogen receptor (ER) and/or the progesterone receptor (PR) [5]. Approximately 70–80% of all diagnosed breast cancers are ER-positive [4,6,7], of which 65% are also PR-positive [7]. Since 2007, there has been a steady increase in incidence rates of ER-positive breast cancer in Western countries, possibly linked to the obesity epidemic, and more widespread mammographic screening which preferentially detects slow-growing ER-positive cancers [2].

The role of hormones as drivers of breast cancer, specifically the steroid hormone estrogen, is well established [8]. Estrogen mediates its effects through nuclear estrogen receptors, particularly ERα [8]. As the majority of HR-positive cancers are dependent on ER signaling for tumor growth and progression, modulating estrogen synthesis and/or ER activity has been a key therapeutic strategy in disease management [9]. Endocrine therapies remain the mainstay of treatment options for HR-positive breast cancer [10–12]. These therapies include compounds which act by reducing the level of endogenous estrogens (e.g. aromatase inhibitors such as anastrozole, letrozole, and exemestane), a selective estrogen receptor modulator (SERM; i.e. tamoxifen) which acts by reducing the effects of estradiol via competitive binding of ER, or a selective estrogen receptor degrader (SERD; i.e. fulvestrant) which acts by fully antagonizing and degrading ER (Figure 1) [9]. Such treatments may also be useful in patients carrying truncating mutations in CHEK2 (e.g. 100delC), which have been associated with positive ER status [13].

Figure 1. Summary of the mode of action of the main classes of endocrine therapy. Endocrine therapies are the the mainstay of treatment options for HR-positive breast cancer. These therapies include compounds which act by reducing the level of endogenous estrogens (e.g. aromatase inhibitors such as anastrozole, letrozole, and exemestane), a selective estrogen receptor modulator (SERM; i.e. tamoxifen) which acts by reducing the effects of estradiol via competitive binding of ER, or a selective estrogen receptor degrader (SERD; i.e. fulvestrant) which acts by fully antagonizing and degrading ER.

AF1, activation function 1 domain; DBD, DNA-binding domain; E2, estrogen; ER, estrogen receptor; LBD, ligand-binding domain; SERD, selective estrogen receptor degrader; SERM, selective estrogen receptor modulator.

Despite the advances made in the treatment of HR-positive breast cancer with endocrine therapies, unmet needs remain. Disease recurrence typically occurs, with as many as 30–50% of patients relapsing in the adjuvant setting [14] and patients may subsequently fail multiple endocrine interventions. Resistance to endocrine therapy limits the use of these agents and remains a significant issue for optimal clinical management of patients whose tumor growth often continues to be driven by signaling through the ER [15]. Approximately 15–30% of patients will not benefit from standard-of-care treatments due to de novo resistance mechanisms, and 30–40% will acquire resistance during treatment [16,17]. Endocrine therapy resistance is a key driver of poor outcomes [17].

Several molecular mechanisms are believed to underlie therapeutic resistance (Figure 2), including hot-spot mutations in the ESR1 gene encoding ERα, which have attracted interest in recent years [16]. Hotspot mutations affecting the ER ligand-binding domain (LBD) (e.g. Y537S and D538G) occur in approximately 20–40% of endocrine-resistant breast tumors [18], and are usually acquired during disease progression after long-term treatment with aromatase inhibitors or tamoxifen [19]. In-frame ESR1 gene fusions (e.g. ESR1-DAB2, ESR1-GYG1, and ESR1-SOX9), in which the ligand-independent activation domain and DNA-binding domains are retained but the LBD is lost, have also been reported in ER-positive metastatic breast cancers (mBCs), and have been implicated as a mechanism of resistance to endocrine therapies [20]. These alterations promote ligand-independent ER activation, proliferation, and metastasis in the absence of estrogen [18,21], and patients with ESR1 mutations exhibit worse outcomes [22]. ESR1 mutations are associated with poor prognosis, and are predictive of resistance to aromatase inhibitors [23], which are often the preferred treatment in postmenopausal patients. In addition, tamoxifen, a standard-of-care in HR-positive breast cancer for many years, has a partial-agonist effect on the ER [24,25], lacking the ability to fully suppress the activation function 1 (AF1) domain [25], which has been linked to the development of resistance [24,25] and poorer prognosis.

Figure 2. Summary of main mechanisms of endocrine therapy resistance. Several molecular mechanisms are believed to underlie therapeutic resistance including hot-spot mutations in the ESR1 gene encoding ERα, which have attracted interest in recent years.

BCL2, B-cell lymphoma 2; CDK4/6, cyclin-dependent kinase 4/6; ER, estrogen receptor; Rb, retinoblastoma.

Another mechanism of endocrine-therapy resistance may be through the growth-promoting PI3K-AKT-mTOR and RAS/RAF/MEK/ERK pathways, which can drive reactivation of ER-mediated transcription in the absence of estradiol and drive ER signaling-independent resistance [26]. Indeed, genomic alterations in components/effectors of these pathways occur commonly in HR-positive mBCs [27,28]. Furthermore, HR-positive breast cancers frequently demonstrate mutations in the cyclin-dependent kinase 4/6 (CDK4/6)-cyclin D1 axis [28], which normally induces inactivation of the retinoblastoma protein and prompts release of E2F transcription factors, leading to progression of the tumor through the cell cycle [26]. To combat the above mechanisms of endocrine-therapy resistance, CDK4/6 cell cycle inhibitors (e.g. abemaciclib, ribociclib, palbociclib), PI3K inhibitors (alpelisib), and mTOR inhibitors (everolimus) have become available in recent years, and serve as combination partners for endocrine therapies [10,12].

In addition to treatment resistance, there are also serious adverse events associated with currently available treatment options. The partial-agonist activity of tamoxifen can lead to the development of thromboembolic disease [29], and has been linked to an increased risk of developing endometrial cancer [9,29]. Aromatase inhibitors have been reported to be associated with a potential for increased risk in adverse events including musculoskeletal symptoms and bone fractures [29,30].

Treatment adherence also remains a challenge for patients who may require therapy for many years. For example, up to 50% of patients do not adhere to the full duration of tamoxifen treatment due to factors including adverse events, belief that the clinical benefits do not outweigh the risks of the treatment, forgetfulness in taking medication, and socioeconomic factors [31–33]. Similarly, adherence to aromatase inhibitors is also suboptimal, with up to 50% of patients nonadherent to adjuvant therapy [34]. There is a clear medical need for therapies with improved efficacy that may prevent or overcome the resistance developed with current endocrine therapies, reduce/postpone the need for more toxic chemotherapy, optimize patients’ quality of life, and improve treatment outcomes [35].

1.1. Fulvestrant and the quest for orally available SERDs

Endocrine therapy-resistant breast cancer often remains dependent on signaling through ER for growth and survival [15]. Consequently, patients may still respond to subsequent therapies that target ER via a different mechanism, and through more potent inhibition of ER, such as SERDs [9,15,36]. Unlike SERMs, SERDs are high-affinity competitive antagonists of ER that immobilize and target ERα for proteasome-dependent degradation, and prevent ER signaling from both the LBD and the AF1 domain [9,36]. The medical need for a drug with full ER antagonistic activity led to development of the steroidal SERD fulvestrant [9,37], the first and only currently approved SERD for the treatment of HR-positive mBC [38–40]. Early preclinical assessments indicated that fulvestrant displayed enhanced anti-proliferative potential versus tamoxifen in breast cancer cells. Fulvestrant was devoid of estrogenic activity in the uterus and also effectively blocked the uterotropic action of both estradiol and tamoxifen in vivo [41]. Further experiments showed that fulvestrant was effective in inhibiting the growth of tamoxifen-resistant tumor cells [42]. Subsequently, fulvestrant demonstrated the clinical therapeutic value of SERDs as a drug class, showing efficacy in tamoxifen-refractory patients [43], and patients with ESR1-mutated HR-positive breast cancer who had progressed on prior aromatase inhibitors [22].

However, fulvestrant has several important limitations that have an impact upon more widespread use and potential efficacy. During early development, oral administration of fulvestrant was explored in animals and healthy volunteers, but was deemed unfeasible due to low aqueous solubility; so the focus switched to development of an intramuscular (IM) formulation [41,43,44]. Initial clinical pharmacokinetic (PK) studies in healthy volunteers, using an intravenous formulation, indicated extensive distribution of fulvestrant following administration, with plasma concentrations decreasing rapidly in a multiexponential fashion. The terminal elimination half-life (t_1/2) ranged from 13.5–18.5 hours [44]. Fulvestrant is highly bound (99%) to plasma proteins (mainly lipoproteins) and is extensively metabolized in vivo [44]. To provide adequate bioavailability and offer patient adherence advantages over available breast cancer treatments, a long-acting, oil-based formulation was developed for once-monthly IM injection [45]. Various studies showed that, following a single IM injection of the long-acting fulvestrant formulation at doses of up to 250 mg, the plasma concentrations declined slowly due to prolonged release of fulvestrant, with a mean time to C_max (t_max) ranging between 2 and 19 days [44]. For multiple-dose fulvestrant 250 mg once monthly, PK steady state was reached in approximately 6 months, as determined by trough concentrations (C_trough) [45,46].

During early phase 3 studies, fulvestrant 125 mg and 250 mg IM monthly doses were assessed versus anastrozole in patients with advanced breast cancer who had progressed on prior endocrine therapy [47,48]. The fulvestrant 125 mg IM monthly dose was stopped due to a lack of efficacy following planned interim analyses [47,48], and fulvestrant 250 mg IM monthly was subsequently approved based on comparable efficacy to anastrozole [38]. Unexpectedly, fulvestrant 250 mg IM monthly failed to demonstrate noninferiority versus tamoxifen when assessed in the first-line mBC setting [49]. As dose-dependent downregulation of ER with fulvestrant had been observed [46], it suggested that efficacy could potentially be improved with a higher dose level (500 mg) to increase fulvestrant in vivo exposure and by using a loading dose to enable exposure to reach steady-state levels more quickly. These observations led to the phase 3 CONFIRM study, which evaluated a fulvestrant 500 mg IM monthly dose that included a loading injection on day 14 versus the approved 250 mg IM monthly dose [50]. Based on the observed clinical efficacy improvement [50], the CONFIRM study established the current approved higher dosing regimen in the fulvestrant label [39,40]. In the subsequent phase 3 FALCON study, 500 mg IM fulvestrant demonstrated significantly longer progression-free survival in the first-line mBC setting, compared with anastrozole [51].

However, despite improved in vivo exposure and activity observed with the higher dose [9,39,40], steady-state concentrations for fulvestrant 500 mg take ~1 month to establish [46] and residual ER protein expression and function remains detectable. Although increasing the dose of fulvestrant to 750 mg IM monthly might further improve potency, no studies have been undertaken to compare long-term outcomes versus the established fulvestrant 500 mg regimen. The 750 mg IM dose would also require three 5 mL IM injections, which may be impractical [46]. Furthermore, although commonly used, IM injections can cause a variety of issues that have an impact on patients’ quality of life and care, including reduced drug efficacy and absorption when incorrectly administered, local injection site complications, patient anxiety, and nonadherence [52,53].

Consequently, the compelling medical need for orally bioavailable SERDs with increased in vivo potency to overcome the limitations of fulvestrant has driven a surge in research and development to identify new compounds to achieve superior efficacy. A number of new orally bioavailable SERDs have progressed into clinical trials over the past few years. Here, we conducted a literature search of PubMed, clinicaltrials.gov, and congress websites (2015 or later) in order to provide an overview of the physicochemical properties, preclinical data, and initial human PK findings of these compounds, and discuss their clinical relevance.

2. Overview of physicochemical properties of oral SERDs recently or currently in development

Table 1 summarizes some of the physicochemical properties of novel oral SERDs; all chemical structures are shown in Figure 3.

Figure 3. Chemical structures of marketed and investigational SERM/SERDs.

SERD, selective estrogen receptor degrader; SERM, selective estrogen receptor modulator.

Table 1. Summary of physicochemical properties and in vitro potencies for novel oral SERDs

	pKa	cLogP	LogD	Degradation MCF-7 DC50 (nM)/Sinf (%)	Proliferation MCF-7				Proliferation CAMA-1	Proliferation HCC1500
					No or 0.01 nM estradiol IC50 (nM)	0.1 nM estradiol		1 nM estradiol IC50 (nM)	1 nM estradiol IC50 (nM)	1 nM estradiol IC50 (nM)
						EC50 (nM)	IC50 (nM) or GI50 (nM)
GDC-0810	a 4.35, b 2.26 [61]a 4.3 [56]	7.0 [61] 6.2 [56]	3.8 [61]	0.74/96% [61] 0.7 [54] IC50 0.0949 [75]	IC50 0.032 [85]	17.2 [61]	IC50 2.5 [54] GI50 1.82 [75]	135.3 [61]	239.1 [61]	239 [61]
GDC-0927	a 6.12, b 7.43 [61]	5.4 [61]	2.9 [61]	0.03/97% [61] 0.1 [60]	0.1 [60]	0.31 [61]	–	5.3 [61]	2.8 [61]	2.1 [61]
Giredestrant (GDC-9545)	b 7.64 [61]	5.0 [61]	2.9 [61]	0.03/101% [61]	–	0.36 [61]	–	5.4 [61]	4.5 [61]	8.7 [61]
Amcenestrant(SAR439859)	a 4.31, b 7.93 [61]	8.4 [61]	3.7 [61] 3.25 [65]	0.2 [65]	1.1 [66]	–	–	80.5 [61]	63.7 [61]	179.0 [61]
Camizestrant(AZD9833)	b 8.01 [61] b 8.4/4.8 [70]	4.2 [61]	2.7 [61] 2.9 [70]	<1 [71]	–	–	–	17.6 [61]	11.5 [61]	27.2 [61]
Elacestrant (RAD1901)	b 9.78 [61]	6.8 [61]	3.6 [61]	0.7/83% [61] EC50 0.6 [81]	1.35 [82] 4.2 [81]	5.9 [61]	–	111.8 [61] 86.9 [82]	86.3 [61]	334.8 [61]
Rintodestrant (G1T48)	a 4.35 [61]	6.3 [61]	–	IC50 0.033 [75]	IC50 0.017 [85]	–	GI50 0.257 [75]	–	–	–

2.1. GDC-0810

GDC-0810 (formerly ARN-0810) was developed as an orally bioavailable SERD for treating HR-positive breast cancer [54,55]. It was identified by prospectively optimizing ERα degradation, antagonism, and PK properties of an indazole series of SERDs [54] and is characterized by a cinnamic acid side chain. GDC-0810 is a weak acid with an acidic pK_a of 4.3 and a logP value of 6.2 [56,57]. The intrinsic solubility of GDC-0810 is very low, and solubility below pH 6.5 was measured as below the detection limit (<0.06 μg/mL) [56]. GDC-0810 induces a distinct ERα conformation relative to that induced by currently approved therapeutics [58]. However, similar to SERMs, GDC-0810 also displayed mild agonistic activity in uterine models in vitro and in vivo [58], as well as inconsistent ER degradation [59]. Although reaching early-phase clinical trials, development GDC-0810 was discontinued based on the totality of available preclinical and clinical data.

2.2. GDC-0927

GDC-0927 (formerly SRN-927) was developed as a novel, nonsteroidal, orally bioavailable SERD designed to further improve potency over GDC-0810 [60]. The identification of GDC-0927 was based on the optimization of the ERα degradation efficacy of a series of ER modulators through side-chain substitution and the addition of a fluoromethyl azetidine group [60]. GDC-0927 has an acidic pK_a of 6.1 and a basic pK_a of 7.4 with a logP value of 5.4 [61]. By shifting away from the acrylic acid moiety in GDC-0810, GDC-0927 achieved increased potency and more consistent, complete suppression of ER signaling. GDC-0927 was shown to induce ERα conformations distinct from fulvestrant and 4-hydroxytamoxifen [60] and can immobilize ER [59]. However, development of GDC-0927 was subsequently halted due to low oral exposure and consequently high pill burden in clinical trials [61].

2.3. Giredestrant (GDC-9545)

Giredestrant is a highly potent, nonsteroidal, orally bioavailable selective estrogen receptor antagonist and degrader developed for the treatment of HR-positive breast cancer alone or in combination (e.g. with CDK4/6 inhibitors) [61,62]. Giredestrant was identified through prospective optimization of ER antagonism, ER degradation, and anti-proliferation in parallel with strategies to optimize absorption, distribution, metabolism, and excretion properties, to further improve on the profiles of GDC-0810 and GDC-0927 [61,63]. The tetrahydrocarboline core in giredestrant provides a better foundation for metabolic stability than those SERDs with a phenol moiety such as fulvestrant, GDC-0927, and elacestrant. A polar difluoropropyl alcohol in giredestrant (cLogP: 5.0) significantly attenuates lipophilicity compared with other SERDs such as amcenestrant (cLogP: 8.4), elacestrant (cLogP: 6.8), and rintodestrant (cLogP: 6.3), trending toward more drug-like properties [61]. The basic side chain in giredestrant enables a full antagonist and consistent degradation profile which has been challenging to achieve for ER ligands including GDC-0810, AZD9496, LSZ102, and rintodestrant which have an acrylic acid moiety. Giredestrant binds to the ER, outcompeting estradiol and immobilizing the receptor to suppress ER-mediated signaling and cellular proliferation; it also induces conformational changes that target the ER for degradation [62,64]. The low lipophilicity, high solubility, and high permeability make giredestrant a good candidate for oral administration [61].

2.4. Amcenestrant (SAR439859)

Amcenestrant is a nonsteroidal, orally bioavailable SERD [65,66] in development for the treatment of HR-positive, HER2-negative breast cancer as monotherapy or in combination with CDK4/6 inhibitors [67–69]. The tetrasubstituted alkene core of amcenestrant resembles that of GDC-0810 and tamoxifen, except that the two neighboring substitutions are cyclized into a seven-membered ring. Amcenestrant is a zwitterion with both an acidic and a basic moiety. The fluoropropyl pyrrolidine side chain in amcenestrant has one extra methylene group compared with the fluoropropyl azetidine moiety in giredestrant and camizestrant, rendering higher lipophilicity [66].

2.5. Camizestrant (AZD9833)

Camizestrant is a potent, orally delivered, nonsteroidal SERD that was developed for the treatment of HR-positive breast cancer. It was designed to further improve on the ER degradation and avoid the ER agonism observed with first-generation oral SERDs such as AZD9496 (the latter being subsequently discontinued in favor of continued development of camizestrant) [70,71]. Extensive structure-enabled chemical optimization of a series of tricyclic indazoles led to the identification of camizestrant. A stringent control of lipophilicity ensured that camizestrant has favorable physicochemical properties (logD: 2.9) similar to those of giredestrant. Given the most basic center with a pK_a of 8.4, camizestrant was found to be highly soluble in aqueous media (833 μM). Even though clearance was generally high and oral bioavailability was modest in rodent species, camizestrant was predicted to have moderate clearance and good oral bioavailability in humans [70].

2.6. Elacestrant (RAD1901)

Elacestrant is a nonsteroidal SERD with complex dose-related ER agonist/antagonist activity that is being developed for HR-positive breast cancer [72,73]. Elacestrant is composed of a tetrahydronaphthalene core and an ethylamine group for its pendent substitution. It is highly basic (pK_a: 9.8) and lipophilic (cLogP: 6.8; LogD: 3.6), possessing properties which are closer to those of the first SERM, tamoxifen (pK_a: 8.6; cLogP: 6.2; LogD: 4.2), than to those of the recent oral SERDs in development [61]. Elacestrant was identified in screens for compounds that manifest ER agonist activity in the CNS, and was claimed to cross the blood–brain barrier readily [74].

2.7. Rintodestrant (G1T48)

Rintodestrant is an orally bioavailable, nonsteroidal SERD that was developed through structure-guided investigations driven by activity in breast cancer cell lines. The chemical backbone of raloxifene was used during early development due to the favorable safety profile observed in clinical trials. Subsequent structural changes led to the identification of rintodestrant which contains a nonsteroidal backbone that incorporates an acrylic acid side chain [75]. The physiochemical properties of rintodestrant (pK_a: 4.4; cLogP: 6.3) [61] are close to those of GDC-0810 (pK_a: 4.3; cLogP: 6.2) [56] given the structural similarities. Rintodestrant is currently in clinical development for the treatment of patients with HR-positive breast cancer [76–78].

3. Summary of in vitro cell potency assays and non-clinical pharmacology

This section includes an overview of available in vitro potency data for each novel oral SERD from cell lines (including MCF-7 cells), anti-proliferation data in mouse xenograft models and patient-derived xenograft (PDX) models, and data from in vitro drug metabolism assays to assess the potential for drug–drug interactions. Table 1 summarizes some of the in vitro potency data of novel oral SERDs.

3.1. GDC-0810

3.1.1. Nonclinical pharmacology

In preclinical studies, GDC-0810 displayed robust activity in tamoxifen-sensitive and tamoxifen-resistant MCF-7 xenograft models and those that harbor ERα mutations [54,58]. In MCF-7 cells, GDC-0810 is an ER antagonist (IC₅₀ = 2 nM) and displays potency in ERα degradation (EC₅₀ = 0.7 nM) and in cell viability assays (IC₅₀ = 2.5 nM) [54]. However, while GDC-0810 displays a full antagonist profile in MCF-7 cells, it promotes partial agonism of ER in other breast cancer cell lines, including HCC1500, CAMA-1, and MDA-MB-134-IV cells, losing potency and anti-proliferative effect relative to the ‘true’ full ER antagonists in these cell lines [59]. GDC-0810 is also a partial agonist in uterine models in vitro and in vivo [58].

3.1.2. In Vitro drug–drug interaction assessment

In cytochrome P450 inhibition profiling, GDC-0810 had little to no inhibition against CYP1A2, CYP2D6, or CYP3A4 (IC₅₀ > 20 μM), modest inhibitory effect on CYP2C9 and CYP2C19 (IC₅₀ = 2.2 and 3.3 μM respectively), and potent inhibition of CYP2C8 (IC₅₀ < 0.1 μM) [54]. GDC-0810 has been shown to be a potent inhibitor of human organic anion-transporting polypeptides 1B1 and 1B3 (OATP1B1/1B3) in vitro [79]. However, a clinical study investigating the drug–drug interaction potential between GDC-0810 and pravastatin (OATP1B1/1B3 substrate) concluded that dose adjustments for pravastatin were not needed when co-administered with GDC-0810 [79].

Clinical development of GDC-0810 was ultimately halted based on the totality of available preclinical and clinical data, including inferior potency when considering other investigational agents in development and partial agonistic activity [59,61].

3.2. GDC-0927

3.2.1. Nonclinical pharmacology

GDC-0927 had high potency (IC₅₀ = 0.1 nM) and degradation efficiency (97%) in in vitro ERα degradation assays in MCF-7 cell lines [60]. In a tamoxifen-resistant breast cancer xenograft model, GDC-0927 induced tumor regression, together with robust reduction of intratumoral ERα levels [60]. GDC-0927 exhibited greater capacity for ER degradation versus GDC-0810 and AZD9496 (a now discontinued first-generation investigational oral SERD) across a range of cellular assays in multiple ER-positive cell lines. GDC-0927 also demonstrated greater potency in inhibiting proliferation of multiple ER-positive cell lines than fulvestrant, 4-hydroxy tamoxifen, AZD9496, or GDC-0810 [59]. In the HCI-013 PDX model, which expresses the ERα-Y537S mutant, GDC-0927 demonstrated greater anti-proliferative capacity as compared with GDC-0810 [59]. Unlike GDC-0810 and tamoxifen, GDC-0927 displays no agonist activity in rat uterine assays [60].

3.2.2. In vitro drug–drug interaction assessment

Cytochrome P450 inhibition profiling of GDC-0927 indicated little to no competitive inhibitory activity against CYP1A2, CYP2C19, CYP2D6, or CYP3A4 (IC₅₀ > 10 μM), and only modest inhibitory effect on CYP2C8 and CYP2C9 (IC₅₀ = 3.0 and 3.2 μM, respectively) [60]. GDC-0927 is a time-dependent inhibitor for CYP2C8, CYP2C9, CYP2C19, and CYP3A4 (Roche, data on file).

The bioavailability of GDC-0927 was low (<15%) resulting in low oral exposure [60], likely due to glucuronidation of the phenol groups in the molecule. Poor bioavailability and low oral exposure resulted in a high pill burden in the clinical setting, which ultimately limited dose escalation and clinical development [61], despite GDC-0927 exhibiting improved potency compared with GDC-0810 [59].

3.3. Giredestrant (GDC-9545)

3.3.1. Nonclinical pharmacology

Giredestrant was developed to address the distinct limitations of GDC-0810 and GDC-0927, aiming to improve on potency of the former and oral bioavailability of the latter [61,64]. In preclinical studies, giredestrant is a highly potent ER antagonist (IC₅₀ = 0.05 nM) and displays potent degradation efficiency (101%) in in vitro ERα degradation assays in MCF-7 cell lines [61]. Degradation and anti-proliferation potencies and degradation efficiency of giredestrant are superior to fulvestrant in both wild-type (DC₅₀ [nM]/S_inf: 0.06/107% versus 0.44/103%) and ERα-Y537S mutant (DC₅₀ [nM]/S_inf: 0.17/113% versus 0.66/109%) MCF-7 cells. Giredestrant is a more efficient degrader than fulvestrant and GDC-0927 in all tested cell lines in western blot assays [61].

Cellular viability assays in three ER-positive breast cancer cell lines (MCF-7, CAMA-1, HCC1500) indicate that giredestrant is more potent than fulvestrant and other SERDs in development including camizestrant, amcenestrant, and elacestrant [61]. Giredestrant (1 mg/kg) achieves the same efficacy (tumor regression) as GDC-0927 (100 mg/kg) in the HCI-013 PDX model at a 100-fold lower dose [61]. Based on the in vivo efficacy data demonstrating tumor regression in the PDX models, giredestrant was projected to be efficacious at a low dose (~1 mg daily [QD]) in humans. Giredestrant exhibits a full antagonist profile with a reduction in the uterine wet weight and no effect on the epithelium height of the endometrium in rat uterine assays [61]. Giredestrant demonstrates favorable nonclinical (in vitro and in vivo) safety profiles with a high safety margin (>190-fold higher than the projected human efficacious exposure) in the preclinical 4-week investigational new drug-enabling toxicity study in monkeys [61]. The low off-target toxicity potentially allowed clinical dosing to achieve greater target saturation.

3.3.2. In vitro drug–drug interaction assessment

In vitro, giredestrant exhibited a weak reversible inhibition of CYP3A4, did not block the function of other CYP enzyme isoforms including 2C9, 2C19, 2D6 (IC₅₀ > 10 μM), and was not a time-dependent inhibitor against CYP3A4, 1A2, 2C9, 2C19, and 2D6 [61].

3.4. Amcenestrant (SAR439859)

3.4.1. Nonclinical pharmacology

Amcenestrant has demonstrated potent antagonist and degradative properties against ER both in vitro and in vivo and binds with high affinity to human wild-type or mutant ERα (ERα-Y537S and ERα-D538G) [65,66]. Amcenestrant antagonizes mutant ERα with lower potency than wild-type ERα (EC₅₀ = 20 nM [wild type], 331 nM [ERα-Y537S mutant], 595 nM [ERα-D538G mutant]) [66]. Across a large panel of ER-positive cells, amcenestrant demonstrated broad and superior ER degradation activity as compared with SERD/SERM hybrids (i.e. GDC-0810 and AZD9496) including improved inhibition of ER signaling and inhibition of cell growth [66]. Amcenestrant effectively induces ERα degradation in MCF-7 cells at subnanomolar concentrations (DC₅₀ of 0.2 nmol/L) with maximal degradation levels of 98% comparable to the in vitro activity of fulvestrant [65,66]. In a tamoxifen-sensitive MCF-7 xenograft tumor model, amcenestrant exhibited dose-dependent tumor-growth inhibition and tumor regression was achieved with a 25 mg/kg twice daily (bid) oral dose [65,66]. In the HCI-013 PDX model, significant tumor regressions were achieved at oral doses of amcenestrant 12.5 and 25 mg/kg bid [66]. Amcenestrant did not have any agonist effect in rat uterine assays at doses of 25, 50, or 100 mg/kg daily [66]. No off-target activity (IC₅₀ ≤ 1 μM) was detected during in vitro selectivity assays [65].

It is notable that in the HCI-013 PDX model, giredestrant induces tumor regression at 1 mg/kg once daily with the AUC and C_max being 60- and 100-fold lower than those of amcenestrant at 25 mg/kg bid [61]. The ability of giredestrant to achieve regressions at a lower exposure in vivo, relative to amcenestrant, is likely due to its higher potency than amcenestrant.

3.5. Camizestrant (AZD9833)

3.5.1. Nonclinical pharmacology

Camizestrant is a potent degrader and antagonist of the ERα receptor (both EC₅₀ < 1 nM in MCF-7 cells) [71], delivering maximal ERα degradation equivalent to fulvestrant and to a greater extent than AZD9496, in all ER-positive cell lines tested, including MCF-7, CAMA-1, T-47D, and BT-474 [70,71,80]. Furthermore, in a wild-type and ERα-D538G mutant PDX model, camizestrant demonstrated combinatorial benefit with palbociclib [71]. In contrast to AZD9496, camizestrant does not cause ER agonism in the endometrial carcinoma cell line and does not cause an increase in the thickness of the endometrium in juvenile rats [71,80]. In vitro safety assays suggested low off-target toxicity and a predicted safety margin >250-fold higher than the predicted efficacious human free C_max [70].

3.5.2. Vitro drug–drug interaction assessment

In cytochrome P450 assays camizestrant showed no inhibition (IC₅₀ values >30 μM) against three isoforms (CYP1A2, CYP2C19, CYP2C9) and modest inhibition against CYP2D6 (IC₅₀ 4.3 μM) and CYP3A4 (IC₅₀ 2.2 μM) [70].

3.6. Elacestrant (RAD1901)

3.6.1. Nonclinical pharmacology

Elacestrant displays high affinity for ERα (IC₅₀ = 48 nM) in competitive receptor binding assays and is a potent antagonist and degrader (EC₅₀ of 0.6 nM) of the ERα receptor in MCF-7 cell lines [81]. Elacestrant induced a dose-dependent decrease in ER protein expression with similar ER degradation to fulvestrant across multiple ER-positive breast cancer cell lines, including MCF-7, T-47D, and HCC1428 [82]. A dose-dependent decrease in proliferation (IC₅₀ = 4.2 nM) was observed in MCF-7 cellular viability assays. Elacestrant at either 30 mg/kg or 60 mg/kg QD induced comparable tumor growth inhibition to that of tamoxifen and fulvestrant after 4 weeks of treatment in an MCF-7 xenograft model, but significantly greater growth inhibition at 90 mg/kg and 120 mg/kg doses [81]

Elacestrant has demonstrated potent antitumor activity in multiple PDX models expressing wild-type or mutant ERα (ERα-Y537S and ERα-D538G) at all doses tested (30 mg/kg, 60 mg/kg, and 120 mg/kg) [82,83]. Elacestrant inhibited tumor growth to a similar extent to fulvestrant in PDX models expressing wild-type ERα [82,83], but demonstrated greater anti-proliferative capacity as compared with fulvestrant in mutant ERα models [83] while also significantly inhibiting growth in models insensitive to fulvestrant [83,84].

Elacestrant at dose ranges between 0.3 and 100 mg/kg was reported not to affect uterine wet weight or epithelial thickness in rats significantly [81]. However, in mouse uterine wet weight assessments, a statistically significant increase in uterine weight was observed with a single low dose (0.3 mg/kg) but not at doses above 1 mg/kg [74]. The latter observation suggested that elacestrant may exhibit dose-dependent agonist/antagonist activity, with antagonist activity manifesting at the higher dose levels [74]. Indeed, in an MCF-7 xenograft model, a similar magnitude of tumor growth stimulation was observed following treatment with lower doses of elacestrant (1 mg/kg or 3 mg/kg) to that previously reported for partial ER agonists, but this was not observed at the higher 10 mg/kg dose [74]. It remains to be seen how this complex pharmacology will impact the clinical use of elacestrant.

3.7. Rintodestrant (G1T48)

3.7.1. Nonclinical pharmacology

Rintodestrant has shown potent activity in both wild-type and ERα mutant breast cancer cells [75,85]. Rintodestrant was found to downregulate ER in MCF-7 cells with an activity modestly more potent than steroidal and other SERDs including fulvestrant and AZD9496 [75]. Rintodestrant significantly inhibited estrogen-mediated growth of MCF-7 cells demonstrating approximately threefold higher potency when compared with fulvestrant [75]. However, in ER-positive BT474 cells, rintodestrant displays a weaker anti-proliferative activity than fulvestrant [75].

Rintodestrant, alone or in combination with CDK4/6 inhibitor lerociclib, demonstrated dose-dependent inhibition of growth of estrogen-dependent MCF-7 xenograft tumors and was more efficacious than fulvestrant in decreasing tumor growth in a PDX model harboring the ER-Y537S mutation as monotherapy (30 and 100 mg/kg) or in combination with lerociclib [75,85].

4. Overview of clinical drug exposure data in humans

This section provides an overview of available pharmacokinetic and pharmacodynamic data for novel oral SERDs from early clinical studies. [18 F]-fluoroestradiol positron emission tomography (FES-PET) imaging measures tumor uptake of radiolabeled estradiol and is used as a biomarker to infer ER occupancy by a competing drug and/or ER downregulation. Also, the clinical drug exposure and the selection of recommended Phase 2 dose (RP2D) are summarized in Table 2 if data are available or published.

Table 2. Summary of clinical drug exposure in humans

	GDC-0810 [55,79]	GDC-0927 [87]	Giredestrant (GDC-9545) [88,91]	Amcenestrant (SAR439859) [67,68,92]	Camizestrant (AZD9833) [80,93,95,96]	Elacestrant (RAD1901) [97]	Rintodestrant (G1T48) [76,77,100]
Absolute bioavailability	NA	NA	Moderate to high (TBD)	NA	40%*	10%	NA
Dose range tested (mg QD)	100–800	600–1400	10–250	20–600	25–450	10–1000	200–1000
Maximum mean exposure reached by oral administration	Cmax 25 μg/mL, AUC0–24102 ug.h/mL (600 mg QD fed)	Data not published	Cmax 1540 ng/mL, AUC0–24 25,800 ng.h/mL (250 mg QD)	Cmax 4380 ng/mL, AUC0–24 43,200 ng.h/mL (400 mg QD)	Cmax 63 ng/mL AUC0–24683 ng.h/mL (75 mg QD)	Cmax 543 ng/mL, AUC0–last 8327 ng.h/mL (1000 mg QD)^†	Cmax 472 ng/mL AUC0–242690 ng.h/mL(1000 mg QD)
Linear PK	No	Yes	Yes	Yes	Yes	No	No
Food effect	Yes	NA	No^‡	No	NA	Yes	Yes
MTD	Not reached^§	Not reached	Not reached	Not reached	Not reached^||	Not reached^¶	Not reached
RP2D	600 mg QD Fed	1400 mg QD	30 mg QD	400 mg QD** 200 mg QD**	75 mg QD	400 mg QD	800 mg QD Fed

* Predicted bioavailability value as reported in [70,80].

^†Values reported as arithmetic means.

‡ Data on file.

§ MTD was not reported; however, 800 mg QD was deemed intolerable due to gastrointestinal adverse events [55].

^||MTD was not reported however DLTs were observed at doses of 300 mg QD and 450 mg QD as monotherapy and at 150 mg QD in combination with palbociclib [93,95].

^¶Although the MTD was not reached, for doses higher than 500 mg QD, gastrointestinal adverse events were poorly tolerated leading to some treatment discontinuations [97].

** 400 mg QD is the RP2D as monotherapy, 200 mg QD is the RP2D in combination with palbociclib [67,68].

MTD, maximum tolerated dose; PK, pharmacokinetics; QD, every day; RP2D, recommended phase 2 dose.

4.1. GDC-0810

In a phase 1 dose escalation study, GDC-0810 was assessed at dose levels ranging from 100 mg QD to 800 mg QD in 41 postmenopausal patients with ER-positive, HER2-negative mBC [55]. GDC-0810 was rapidly absorbed with peak concentrations achieved within 1–3 hours. GDC-0810 exhibited linear PK with exposure increasing proportionally for doses up to 600 mg QD. The PK profile for GDC-0810 at 800 mg QD was similar to that of 600 mg QD; however, the presence of intolerable gastrointestinal adverse events and nonlinear PK observed with the 800 mg QD dose limited the choice of the RP2D. The food effect was explored for 600 mg QD and 800 mg QD doses in the phase 1 study and indicated that C_max and AUC_0–24 in the fed state were numerically higher than those in the fasted state [55]. The mean terminal t_1/2 following a single 600-mg dose of GDC-0810 under the fed condition was approximately 8 hours and the steady-state exposure reached a C_max of 25 μg/mL and AUC_0–24 of 102 μg.h/mL [55].

Of patients who underwent paired FES-PET scans, 80% (24 of the 30) achieved ≥90% decrease in FES avidity, including 1 of 3 patients receiving 200 mg QD, 2 of 4 patients receiving 400 mg QD, 14 of 16 patients receiving 600 mg QD, and 7 of 7 patients receiving 800 mg QD [86].

Since the exposure of 800 mg QD was similar to that of 600 mg QD and 800 mg QD was deemed intolerable due to gastrointestinal side effects, GDC-0810 600 mg QD administered with a meal was selected for future clinical trial assessment [55]. However, the development of GDC-0810 was ultimately discontinued based on the totality of available preclinical and clinical data.

A clinical study investigating the drug–drug interaction potential between GDC-0810 (in vitro data showed inhibition of OATP1B1/1B3) and pravastatin (OATP1B1/1B3 substrate) was conducted [79]. The AUC and C_max of pravastatin were increased by approximately 41% and 20%, respectively, following coadministration of GDC-0810 and pravastatin as compared with pravastatin given alone. This magnitude of drug–drug interaction is considered not clinically relevant and no dose adjustment is needed [79].

4.2. GDC-0927

In a phase 1 dose escalation study, GDC-0927 was assessed at three dose levels (600 mg QD, 1000 mg QD, and 1400 mg QD) in 42 postmenopausal patients with ER-positive, HER2-negative mBC [87]. GDC-0927 exhibited linear PK in humans and exposure was approximately dose-proportional with a t_1/2 of approximately 20 hours, supporting QD dosing [87]. FES-PET showed a complete or near-complete (>90%) suppression of FES uptake to background levels, including in patients with ESR1 mutations. Evidence of reduced ER levels and Ki67 staining (a biomarker for tumor proliferation) was observed in on-treatment biopsies [87]. GDC-0927 1400 mg QD was selected as the RP2D [87]. However, further development of GDC-0927 was halted in 2017 as a consequence of poor bioavailability and a high pill burden in clinical trials [61].

4.3. Giredestrant (GDC-9545)

In an ongoing phase 1 study (GO39932; NCT03332797), giredestrant was assessed at dose levels ranging from 10 mg QD to 250 mg QD in 111 postmenopausal patients with ER-positive, HER2-negative mBC [88]. Previous observations of low off-target toxicity with giredestrant during preclinical assessments against in vitro panels of human kinases, nuclear receptors, ion channels, transporters, and enzymes [61] are reflected by the clinical results where all tested dose ranges were well tolerated, no dose-limiting toxicities (DLTs) were reported, and no maximum tolerated dose (MTD) was reached [88].

Exposure to giredestrant increased dose-proportionally over the dose range of 10 mg to 250 mg. The maximum concentrations were achieved within 3 hours after oral administration [88]. The geometric mean t_1/2 ranged from 25.8 to 43.0 hours [88]. At 30 mg QD, the steady-state exposure reached a C_max of 231 ng/mL and AUC_0–24 of 3,850 ng.h/mL [88]. Giredestrant (given orally) achieved higher plasma molar concentrations than those of fulvestrant (given as IM injection), at all doses tested including at 10 mg QD [62]. A high-fat meal did not change total exposure of giredestrant as compared with dosing under fasted condition (Roche, data on file).

FES-PET showed that giredestrant (at all doses) appears to achieve a higher receptor occupancy than fulvestrant (500 mg dose), with 78% (11 of the 14) of patients showing >90% suppression of FES uptake, including patients with ESR1 mutations [62] compared with approximately 56% (9 of the 16) of patients treated with fulvestrant showing more than 75% suppression of FES uptake [89]. Giredestrant was able to achieve near-complete suppression of FES uptake (>90%) even at the lowest dose of 10 mg, demonstrating superior ER occupancy (or ER protein reduction) with giredestrant compared with fulvestrant.

In patients with HR-positive early breast cancer, giredestrant demonstrated similar levels of biological activity (decrease in ER transcriptional activity, ER protein levels, and Ki-67 expression) at 10 mg QD, 30 mg QD, and 100 mg QD doses [90]. The absence of an observed increase in efficacy at higher doses is consistent with the saturation of biological effects in the pharmacodynamics analysis [90].

Giredestrant has demonstrated encouraging antitumor activity in patients with HR-positive advanced or mBC, independent of whether patients had ESR1 mutations, or whether they had received prior treatment with chemotherapy, fulvestrant or CDK4/6 inhibitors [88,91]. Giredestrant is also well-tolerated as a single agent and in combination with palbociclib [88,91]. Adverse events were generally of grade 1–2 in severity, and no patients discontinued treatment with single-agent giredestrant due to adverse events [88,91]. Asymptomatic grade 1–2 sinus bradycardia was observed in patients receiving ≥100 mg single-agent giredestrant; at 30 mg giredestrant, sinus bradycardia was uncommon and assessed as unrelated to giredestrant [88,91].

The PK analysis and clinical data demonstrated no clinically relevant drug–drug interactions between giredestrant and palbociclib [91].

Based on the totality of available data on overall activity, safety, pharmacodynamics and PK data, giredestrant 30 mg QD dose was selected for subsequent phase 3 clinical development and is under evaluation in two ongoing phase 3 studies (NCT03916744, NCT04546009) [88].

4.4. Amcenestrant (SAR439859)

In a phase 1/2, first-in-human dose-escalation study, amcenestrant was assessed at dose levels ranging from 20 mg QD to 600 mg QD in 16 postmenopausal patients with ER-positive, HER2-negative mBC [67]. There were no DLTs at any dose levels and the MTD was not reached [67]. Amcenestrant was rapidly absorbed following oral administration, with a median t_max around 3 hours and an apparent terminal t_1/2 of around 8 hours [67]. After repeated QD administration, there was low to no accumulation and amcenestrant exhibited a dose proportional increase of exposure at dose ranges tested (20–600 mg) [67,92]. At 400 mg QD, the steady-state exposure reached a C_max of 4,380 ng/mL and AUC_0–24 of 43,200 ng.h/mL [93]. Food intake was reported to not have a major effect on exposure [67,92].

In FES-PET scans, ER occupancy generally exceeded >87% with plasma concentrations >100 ng/mL [67]; the 400 mg QD dose allows 90% of FES-PET signal inhibition during the dose interval [92].

Amcenestrant has shown a favorable overall safety profile (with no clinically significant cardiac findings) and an encouraging antitumor activity in patients who have not received a prior SERD, CDK4/6 inhibitors or mTOR inhibitors supporting development in earlier lines of therapy [68,92].

Amcenestrant may be a moderate inducer of cytochrome P450 3A4 [68]. A clinically relevant drug–drug interaction has been observed with amcenestrant in combination with palbociclib; co-administration of amcenestrant 200 mg QD or 400 mg QD was found to reduce palbociclib exposure by 23% and 60%, respectively [68].

Based on the totality of available data, amcenestrant 400 mg QD was selected as the RP2D as monotherapy [67] whereas the RP2D in combination with palbociclib was reduced to 200 mg QD due to the observed drug–drug interaction [68]. This drug–drug interaction is also expected with other CDK4/6 inhibitors (ribociclib and abemaciclib) since these are all primarily metabolized by CYP3A [94].

4.5. Camizestrant (AZD9833)

In an ongoing, phase 1 dose escalation study (SERENA-1), camizestrant was assessed at dose levels ranging from 25 mg QD to 450 mg QD in patients with HR-positive, HER2-negative advanced breast cancer [93,95]. The maximum concentration of camizestrant was achieved within 2–4 hours [93]. Camizestrant exposure was dose proportional with a median terminal t_1/2 of 11–13 hours across all doses [93]. After a single dose of 75 mg, the observed C_max and AUC_0–24 were 63 ng/mL and of 683 ng.h/mL, respectively [80]. Predicted bioavailability in humans is approximately 40% [70,80]. Camizestrant exposure was similar between that of monotherapy and in combination with palbociclib, and palbociclib exposure was not changed (compared with historical data) when given in combination with camizestrant, suggesting no drug–drug interactions between camizestrant and palbociclib [95].

Evidence of ER and PR reduction was observed by immunohistochemical staining of paired biopsies following treatment with camizestrant in all dose cohorts, indicating modulation of ER signaling [93].

DLTs were observed at doses of 300 mg QD and 450 mg QD as monotherapy and at 150 mg QD in combination with palbociclib [93,95]. The tolerability profile of camizestrant in combination with palbociclib was consistent with that observed as monotherapy, and the known tolerability profile of palbociclib [95]. All instances of camizestrant-related bradycardia were asymptomatic (grade 1–2). A DLT of grade 3 QTcF prolongation was observed at 300 mg QD, which was resolved with dose reduction [93]. Camizestrant will be studied at a dose of 75 mg QD in combination with palbociclib in a phase 3 study (SERENA-4) [96].

4.6. Elacestrant (RAD1901)

In two phase 1 studies (Study 001 and Study 004), elacestrant was assessed at dose levels ranging from 1 mg QD to 1000 mg QD in 140 healthy postmenopausal women [97]. Study 001 assessed a single ascending dose (SAD) of elacestrant from 1 to 200 mg, and multiple ascending doses of elacestrant from 10 to 200 mg QD. After a single oral dose under fasted conditions, elacestrant was absorbed with mean t_max ranging from 1.6 to 3.3 h. The mean t_1/2 ranged from 27.4 to 32.5 h. The absolute bioavailability was estimated at a fairly low percentage (10%) under the fasted state [97]. The bioavailability of elacestrant was likely limited by the low, pH-dependent solubility (≥5 mg/ml at pH 4.5 and 0.0174 mg/ml at pH 6.8) and low permeability of elacestrant [97]. A significant food effect on elacestrant PK was observed. At a 50 mg single dose, a high-fat meal delayed the elacestrant peak plasma concentration by approximately 2 hours and increased the C_max by approximately two-fold (3.4 ± 0.7 ng/mL versus 7.0 ± 1.5 ng/mL, respectively) and AUC_0–last by 1.6-fold (61.3 ± 11.7 ng.h/mL versus 96.8 ± 20.0 ng.h/mL) compared with the fasted state [97]. Elacestrant PK was found to be slightly non-linear and that exposure increased more than dose proportionality.

Study 004 was a dose-escalation study designed to assess the MTD of elacestrant [97]. Elacestrant was administered as oral capsules in the fed state at doses ranging from 200 mg QD to 1000 mg QD. The mean t_max ranged between 3.3 and 4.5 h and steady-state t_1/2 ranged from 37.5 to 41.6 h, in line with observations in study 001. Although investigators determined the MTD was not reached, gastrointestinal adverse events were poorly tolerated at doses higher than 500 mg QD, leading to some treatment discontinuations [97]. In an exploratory analysis, although a slightly positive trend was observed between elacestrant concentration and QTcF values, elacestrant was deemed to have a low risk on QT prolongation [97].

Among healthy postmenopausal women in Study 004, FES-PET scans of the uterus showed 83% reduction in FES uptake at elacestrant dose 200 mg QD and 92% at dose at 500 mg QD [97]. In a separate phase 1 study in 16 postmenopausal patients with HR-positive, HER2-negative, advanced breast cancer, elacestrant 200 mg and 400 mg QD had a median reduction in tumor FES uptake of approximately 89% [73].

Elacestrant has shown an acceptable safety profile and demonstrated single-agent activity in a phase 1 study of 57 heavily pretreated, postmenopausal patients with HR-positive, HER2-negative, advanced breast cancer [72]. Most patients (50 of the 57) were treated with a 400 mg QD dose and no DLTs were reported [72].

The totality of available data on the PK profile, ERα occupancy and safety data for elacestrant, supported further assessment of the RP2D 400 mg QD dosing regimen [72,73,97]. Elacestrant is currently in phase 3 development (NCT03778931) in patients with HR-positive, HER2-negative advanced or mBC [98].

4.7. Rintodestrant (G1T48)

In an ongoing, phase 1 study, rintodestrant was assessed at dose levels ranging from 200 mg QD to 1000 mg QD in postmenopausal patients with HR-positive, HER2-negative advanced or mBC [76]. Less than dose-proportional increases in rintodestrant exposure were observed at doses above 400 mg [76]. Absorption of rintodestrant was rapid with a median T_max of approximately 2 hours and the apparent mean terminal t_1/2 was estimated to be approximately 16 hours [76]. Following a high-fat meal, AUC_0–24 increased by approximately 21%, and T_max was delayed by 1 hour relative to fasted conditions, suggesting food increased exposure in patients with relatively low exposure in the fasted state [76]. Food intake resulted in less variability in C_max (fed: 46.3% versus fasted: 63%) and AUC_0–tau (fed: 34.7% versus fasted: 46.1%) [76].

Of patients who underwent FES-PET scans (9 of the 26), median maximum standard tumor uptake values were decreased substantially after 4 weeks of treatment with rintodestrant, ranging from 70% decrease at 200 mg QD up to 92% decrease at 1000 mg QD indicating target engagement with the ER at all tested doses [76]. In a subsequent analysis it was reported that at doses ≥600 mg QD, the mean reduction FES-PET in the maximum standard uptake value was 89% [99].

No DLTs were observed at any dose level and the MTD was not determined [76,77]. Rintodestrant has demonstrated a favorable safety profile and preliminary evidence of antitumor activity in patients following progression after several lines of endocrine therapy, including in patients with tumors harboring ESR1 mutations [76,77]. Rintodestrant exposure was as predicted when given in combination with palbociclib, and palbociclib exposure was comparable to historical data suggesting no drug–drug interactions between rintodestrant and palbociclib [100].

In the absence of an MTD, 600 and 1000 mg QD with food were selected for further clinical development in an expansion cohort in part 2 of the study based on available safety, PK, pharmacodynamics, and preliminary antitumor activity data [76]. Based on subsequent safety, efficacy, and ER degradation data, 800 mg QD was selected as the optimal dose for further assessment in part 3 of this study which is evaluating rintodestrant in combination with palbociclib in patients in the mBC setting [77,78].

5. Conclusions

For decades, endocrine therapies have formed the therapeutic backbone for the management of HR-positive breast cancer. A progressive evolution in our understanding of the molecular mechanisms underlying endocrine therapy resistance and encouraging clinical efficacy of fulvestrant helped fuel interest in the development of orally bioavailable SERDs with greater potency. Recent advances in structural design and optimization of competitive ER binding, ER antagonism, and ER degradation in parallel with strategies to optimize absorption, distribution, metabolism, and excretion properties has led to the prospective development of a novel series of SERD compounds. The improved profiles of these molecules are such that the community is now able to test the hypothesis that overcoming the limitations of fulvestrant may improve outcomes for women with ER-positive breast cancer. However, challenges remain for developing an oral SERD, such as the need for high dose (pill burden) and an unfavorable PK profile (including nonlinear PK and significant food effects), as well as safety concerns with DLTs or drug–drug interactions in combination with CDK4/6 inhibitors, which have been observed with oral SERD candidates. A number of these candidates are currently in early-stage clinical development or progressing toward more advanced stages of clinical development. If the results of phase 3 trials confirm the preliminary safety and efficacy results of early clinical studies, these investigational agents could become important additions to the therapeutic arsenal against HR-positive breast cancer.

6. Expert opinion

Despite the availability of effective treatment options for HR-positive breast cancer, key unmet needs remain to be addressed, including acquired or de novo resistance to endocrine therapies [16,17], to help further optimize the management and outcomes of patients. The treatment of HR-positive breast cancer is entering an exciting era, attributable to the advances in structural design and optimization of new investigational ER-targeted agents. It is hoped that these new agents will help overcome disease resistance with current standard-of-care treatment options and, through a combination of optimized mechanistic, potency, and drug-like properties, may translate into superior efficacy in the clinic. After many years without success, the possibility of having one or more oral SERDs as available treatment options appears a step closer, with encouraging data reported for many candidates in early-phase clinical trials. These latest generation oral SERDs are aiming to form the core treatment of endocrine-sensitive mBC, as single agents or in combination with other therapies, such as CDK4/6 inhibitors. As well as the oral SERDs discussed here, ARV-471 and ZB716 have also entered clinical trials, but there are insufficient publications to warrant a detailed review.

Fulvestrant is currently the only approved SERD available for the treatment of HR-positive advanced or mBC [39,40]. It has demonstrated efficacy in multiple clinical trials either as a single agent or in combination with CDK4/6 inhibitors. However, clinical development of fulvestrant has been limited by several factors. While the development of a long-acting formulation made fulvestrant monthly administration feasible [45], the bioavailability issue and the need for IM injection have been indicated as a potential basis for suboptimal clinical efficacy [46] and patient nonadherence in real-world clinical practice. In particular, it has been suggested that a shorter time to reach steady state as well as a higher in vivo exposure could have improved the efficacy of fulvestrant. Although a loading dose was used to bring initial fulvestrant exposure closer to steady state, a higher dose of fulvestrant (e.g. 750 mg IM monthly administration) has never been studied in large randomized clinical trials to test the assumption, likely due to feasibility reasons [46]. To address the limitation of fulvestrant, a serious effort has been made to overcome the bioavailability challenge through the development of oral SERDs which may also help to improve convenience for patients by moving away from the needs for IM injections toward a preferable oral administration option. While these goals appear to have largely been achieved by these new oral SERDs (amcenestrant, camizestrant, elacestrant, giredestrant, and rintodestrant), we need to await the outcomes of larger phase 3 trials to fully understand the risk–benefit profiles of these agents under different clinical settings.

The focus on optimization of the ER degradation and ERα antagonism has enabled either comparable or improved potency versus fulvestrant against either wild-type or mutant ERα in vitro and in vivo. Notably, based on preclinical data from cellular viability assays, giredestrant displays potency superior to that seen with other SERDs in development including camizestrant, amcenestrant, and elacestrant [61]. The ability of giredestrant to achieve target saturation, as well as saturation of pathway inhibition, at low exposures in vivo is enabled by its high potency for ER binding, antagonism, and degradation, which has been demonstrated in vitro [61]. As a result, giredestrant has the lowest RP2D dose (30 mg QD) relative to all the other investigational agents entering later-stage clinical development.

The bioavailability of many of these investigational agents may ultimately lead to improved efficacy over existing treatments and provide options for managing disease resistance in the clinic. However, any possible efficacy advantages will need to be balanced against the emerging safety profiles of these agents and other factors that may have an impact on a patient’s quality of life, as the risk–benefit ratio will be an important consideration in the future use of these agents. Many of these agents have shown a favorable safety profile in early clinical studies to date. Of note, giredestrant has combined high potency with a large safety margin at the lowest dose (30 mg QD) of SERDs in development [61]. The potential for drug–drug interactions reported for some SERDs in development will also require further investigation regarding their impact and will be an important consideration for future combination with other agents. In addition to oral route of administration, pill burdens, therapeutic window, and food requirement associated with drug administration are also key considerations when seeking to optimize clinical management of a disease that has historically been associated with adherence issues due to long-term treatment requirements [31–34]. Once-a-day dosing, low pill burden, a wide therapeutic window, and an absence of any administration restrictions for the patient are therefore desirable attributes for these new SERDs.

The breast cancer community has great expectations for these innovative oral SERDs under investigation, which could provide a new wave of treatment options able to further enhance patient outcomes and quality of life.

Article Highlights

• Despite the success of current endocrine therapies in treating hormone receptor-positive (HR-positive) breast cancer, acquired or de novo resistance limits their use.

• The clinical impact of fulvestrant, the only currently approved selective estrogen receptor degrader (SERD) for HR-positive breast cancer, has been limited by poor drug-like properties and an inconvenient method of administration (i.e. intramuscular injection).

• Property- and structure-based drug design and optimization of estrogen receptor (ER)-binding, antagonism, degradation, anti-proliferation, and pharmacokinetic (PK) properties have recently led to the development of innovative oral SERDs.

• The unique physicochemical properties of each oral SERD are evident through their different potencies, presence or absence of ER-agonist activity, PK profiles (including nonlinear PK and food effects), requirements for higher dosing, and potential for drug–drug interactions.

• Giredestrant (GDC-9545) has displayed potency superior to other SERDs in development based on preclinical data which, combined with a high safety margin, have translated into having the lowest recommended phase 2 dose (30 mg once a day) relative to all the other investigational agents entering later-stage clinical development.

• It is hoped that, if the results of phase 3 trials confirm the preliminary safety and efficacy results of early clinical studies, these investigational agents may lead to superior efficacy in the clinic.

This box summarizes key points contained in the article.

Author Contributions

All authors were involved in conception of the work and drafting the article or revising it critically for important intellectual content. All authors approved the final version to be published and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Declaration of interest

All authors are employees of Genentech, Inc., and hold stock options in F. Hoffmann-La Roche Ltd and received research support in the form of third-party medical writing assistance for this manuscript, provided by F. Hoffmann-La Roche Ltd. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Acknowledgments

Support for third-party writing assistance for this manuscript, furnished by Martin Cadogan, PhD, of Health Interactions, was provided by F. Hoffmann-La Roche Ltd, Basel, Switzerland.

Additional information

Funding

This work was supported by F. Hoffmann-La Roche Ltd.