Systemic therapy for hormone receptor-positive/human epidermal growth factor receptor 2-negative early stage and metastatic breast cancer

Abstract

Hormone receptor (HR)-positive and human epidermal growth factor receptor 2 (HER2)-negative breast cancer is defined by the presence of the estrogen receptor and/or the progesterone receptor and the absence of HER2 gene amplification. HR-positive/HER2-negative breast cancer accounts for 65%–70% of all breast cancers, and incidence increases with increasing age. Treatment varies by stage, and endocrine therapy is the mainstay of treatment in both early stage and late-stage disease. Combinations with cyclin-dependent kinase 4/6 inhibitors have reduced distant recurrence in the early stage setting and improved overall survival in the metastatic setting. Chemotherapy is used based on stage and tumor biology in the early stage setting and after endocrine resistance for advanced disease. New therapies, including novel endocrine agents and antibody-drug conjugates, are now changing the treatment landscape. With the availability of new treatment options, it is important to define the optimal sequence of treatment to maximize clinical benefit while minimizing toxicity. In this review, the authors first discuss the pathologic and molecular features of HR-positive/HER2-negative breast cancer and mechanisms of endocrine resistance. Then, they discuss current and emerging therapies for both early stage and metastatic HR-positive/HER2-negative breast cancer, including treatment algorithms based on current data.

INTRODUCTION

The current clinically relevant classification schema of breast cancer distinguishes three main subtypes of breast cancer, including (1) hormone receptor (HR)-positive/human epidermal growth factor receptor 2 (HER2)-negative, (2) HER2 gene amplified/receptor overexpressed (HER2-positive), and (3) HR-negative/HER2-negative or triple-negative breast cancer (TNBC). HR-positive breast cancer is defined as the presence of the estrogen receptor (ER) and/or the progesterone receptor (PR). HR-positive/HER2-negative breast cancer is the most common subtype of breast cancer, accounting for 65%–75% of all breast cancers, and is the subject of this article. In this review, we first discuss the pathologic and molecular features of HR-positive/HER2-negative breast cancer. Then, we review current treatment strategies and promising novel therapies that are under investigation for patients with HR-positive/HER2-negative early stage and metastatic breast cancer (MBC), including endocrine therapies (ETs), targeted therapies, chemotherapy, antibody-drug conjugates (ADCs), and immunotherapy. For each topic, we summarize important preclinical and clinical data, discuss implications for clinical practice, and highlight future research directions.

FEATURES OF HR-POSITIVE/HER2-NEGATIVE BREAST CANCER AND MECHANISMS OF ENDOCRINE RESISTANCE

Epidemiology and risk factors for HR-positive/HER2-negative breast cancer

Breast cancer is the most common cancer in women (excluding skin cancer), accounting for approximately one third of all female cancers. According to the National Cancer Institute’s Surveillance, Epidemiology, and End Results database, there were approximately 281,550 new cases of invasive breast cancer and 43,600 breast cancer-related deaths in the United States in 2021, representing 14.8% of all cancer cases and 7.2% of all cancer deaths.¹ According to data from the World Health Organization, there were 2.3 million women diagnosed with breast cancer and 685,000 breast cancer-related deaths globally in 2020.² Approximately 12.9% of women (one in eight) will be diagnosed with breast cancer during her lifetime.¹ The vast majority of breast cancer cases occur in females (approximately 99%), but there are approximately 2500 new cases of breast cancer diagnosed in males annually in the United States. Breast cancer subtypes vary by race/ethnicity. Incidence rates of HR-positive/HER2-negative breast cancer are highest in White women (138 cases per 100,000), with rates 23% higher than in Black women (112 cases per 100,000) and 45% higher than in Hispanic and Asian/Pacific Islander women (94–97 cases per 100,000).³ In the United States, the median age at diagnosis of HR-positive/HER2-negative breast cancer is approximately 61 years, which is slightly older than the median age at diagnosis of HER2-positive breast cancer (56 years) and TNBC (57 years).⁴ There is variation in breast cancer incidence worldwide, which is thought to be caused by differences in reproductive patterns and other hormonal factors, as well as screening and early detection.⁵ Risk factors for HR-positive breast cancer include germline genetic mutations, family history, increased age, obesity, heavy alcohol intake, decreased physical activity, early menarche, late first full-term pregnancy, and late menopause.⁶ Exogenous use of female hormones also plays a role in breast cancer incidence, especially the use of hormone-replacement therapy.^7-9

Most patients are diagnosed with breast cancer that is localized to the breast and axillary lymph nodes (94%–97%), but 3%–6% of patients in high-income countries have metastatic disease at the time of diagnosis, termed de novo metastatic disease.¹⁰ In addition, 10%–30% of patients with early stage breast cancer later develop systemic recurrence. The evolution of disease from early stage to metastatic disease is marked by the acquisition of new mutations and altered gene expression, which leads to increased proliferation, formation of metastases, and development of treatment resistance.

Of note, the definition of ER/PR positivity has been an evolving area of research. Prior studies demonstrated that assessing ER positivity by immunohistochemistry (IHC) was superior to use of the ligand-binding assay,¹¹ so IHC is now the preferred testing method. Currently, the American Society of Clinical Oncology (ASCO) and the College of American Pathologists recommend that ER should be considered positive if ≥1% tumor cells demonstrate positive nuclear staining with an IHC assay. More recently, studies evaluating intrinsic subtypes have demonstrated that many, but not all, tumors with ER <10% are more like TNBC both clinically and biologically,¹² resulting in the inclusion of low ER-positive disease in some TNBC clinical trials.

Pathologic and molecular features of HR-positive/HER2-negative breast cancer

Breast cancer typically arises from benign ductal or lobular breast tissue by undergoing cumulative genetic alterations until one cell accumulates enough mutations to proliferate clonally and uncontrollably. In HR-positive/HER2-negative breast cancer, estrogen binding to ER stimulates receptor-regulated transcription, which promotes tumor cell growth and proliferation.

There is significant heterogeneity among HR-positive breast cancer. Tumors vary in terms of levels of ER and PR, histologic grade, histologic subtype (ductal, lobular, cribriform, tubular), degree of proliferation (measured by Ki-67 and other markers), patterns of gene expression, and differences in genomic alteration.¹³ There has been significant effort to elucidate the biological heterogeneity; and, in 2009, Parker et al. introduced a clinically applicable gene expression-based test known as PAM50 (Prosigna), which identifies four main intrinsic molecular subtypes of breast cancer: luminal A, luminal B, HER2-enriched, and basal-like.^14-16 Luminal A breast cancers tend to be strongly ER-positive and PR-positive, with a lower histologic grade, lower expression of proliferation-related genes, and generally have a better prognosis. In contrast, luminal B breast cancers often have lower ER expression, lower or negative PR expression, a higher histologic grade, higher expression of proliferation cluster genes (e.g., MKI67) and cell cycle-associated genes (e.g., CCNB1, MYBL2), and generally have a worse prognosis.¹⁷ Luminal B tumors may also be both HR-positive and HER2-positive. Basal-like tumors are generally triple-negative, although highly proliferative HR-positive tumors may fall into this subtype. In a combined data set among 9258 patients with early stage HR-positive/HER2-negative disease, 54.5% were luminal A, 39.9% were luminal B, 5.8% were HER2-enriched, and 2.2% were basal-like.¹⁸ Multiple studies have demonstrated that these subtypes are associated with response to ET and chemotherapy as well as prognosis in the early stage and metastatic settings.^{14, 19, 20}

There are several known germline mutations that confer an increased risk for breast cancer. BReast CAncer gene 1 (BRCA1) and BReast CAncer gene 2 (BRCA2) are tumor suppressor genes that encode proteins involved in DNA double-strand break repair via the homologous recombination repair pathway. In a prospective cohort, the cumulative breast cancer risk to age 80 years was 72% for BRCA1 carriers and 69% for BRCA2 carriers.²¹ Patients with a BRCA1 mutation are more likely to develop TNBC, whereas patients with a BRCA2 mutation are more likely to develop HR-positive breast cancers. Other common germline mutations that increase the risk of breast cancer include CHEK2 (approximately1% of cases), ATM (approximately 1%–2%), and PALB2 (1.0%–2.5%), most of which moderately increase the risk for HR-positive breast cancer.²² The prevalence of hereditary mutations in HR-positive/HER2-negative breast cancer is highest in patients who are diagnosed with breast cancer when they are younger than 40 years old (15%).²³ Guidelines from the National Comprehensive Cancer Network (NCCN) recommend assessing for germline BRCA1/BRCA2 mutations in all patients with recurrent breast cancer or MBC cancer to identify candidates for PARP inhibitor therapy. For patients with early stage breast cancer, the NCCN guidelines recommend germline genetic testing for patients aged 50 years or younger, any patient with TNBC, those with a strong family history of breast cancer, those who would be candidates for adjuvant PARP inhibitor therapy, and all male patients with breast cancer, among others.²⁴ Alternatively, there are some groups, such as the American Society of Breast Surgeons, that recommend genetic testing for anyone with a personal history of breast cancer.²⁵

The prognosis for HR-positive/HER2-negative breast cancer depends on anatomic stage (tumor size and nodal status), tumor grade, and genomic information.^{26, 27} Several prognostic and predictive gene expression assays have been developed that both predict the risk of breast cancer recurrence and predict benefit from chemotherapy, such as the 21-gene Recurrence Score (RS) (Oncotype Dx) and the 70-gene signature (MammaPrint; Agendia), which are discussed in the early stage section of this review (see below). Breast cancers in premenopausal women tend to have lower levels of ER, a higher histologic grade, and more aggressive genomic signatures compared with breast cancers in postmenopausal women, and the prognosis is generally worse in younger women because of these features.²⁸

In HR-positive/HER2-negative breast cancer, the risk of recurrence persists for a long duration of time: approximately 50% of risk occurring in the first 5 years, and there is a persistent risk of late recurrences (>10 years). In an analysis of data from the Early Breast Cancer Trialists' Collaborative Group of over 60,000 patients with early stage HR-positive breast cancer who had received 5 years of adjuvant ET in the 1990s, the risk of late distant recurrence was strongly correlated with the original TN status: the 20-year risk of recurrence ranged from 13% in patients with small, lymph node-negative tumors to 41% in patients with four to nine positive axillary lymph nodes.²⁹ However, the risks of recurrence observed in that study likely overestimate the recurrence risk today because contemporary adjuvant therapies and determination of receptor status have improved in the intervening years. Indeed, in an updated analysis presented at the San Antonio Breast Cancer Symposium (SABCS) in 2019, the 20-year recurrence risk was approximately 25% less in women who were diagnosed after 2000 compared with those who were diagnosed before 2000. Given the extended risk of recurrence in HR-positive disease, there has been a great deal of interest in whether prolonging ET will improve outcomes, which we discuss further in the early stage section of this review (see below).

Endocrine therapy and mechanisms of endocrine resistance

Because estrogen drives the proliferation of HR-positive/HER2-negative breast cancers, it is critical to reduce the ability of estrogen to stimulate cancer cell growth and promote cancer cell survival. Endocrine treatment strategies include decreasing estrogen production, modulating signaling through the ER, and/or antagonizing and degrading the ER itself (Figure 1). Despite the initial efficacy of ET in most patients, malignant cells acquire new mutations and altered gene expression to develop endocrine resistance over time.

FIGURE 1

Currently available endocrine therapies and targeted therapies that combine with endocrine therapies for HR-positive/HER2-negative breast cancer. The growth factor receptor and estrogen receptor signaling pathways are shown here. When constitutively active, these pathways can promote the transcription of genes that promote cancer growth and survival. Endocrine therapies and targeted agents can inhibit steps in this pathway, as shown. AI indicates aromatase inhibitor; CDK4/6, cyclin-dependent kinase 4/6; ER, estrogen receptor; HR, hormone receptor; HER2, human epidermal growth factor receptor 2; MAPK, mitogen-activated protein kinases; MEK, mitogen-activated protein kinase; mTORi, mammalian target of rapamycin inhibitor (member of the phosphatidylinositol 3-kinase–related kinase family of protein kinases); P, phosphorylation; PI3K, phosphatidylinositol 3-kinase; SERD, selective estrogen receptor degrader; SERM, selective estrogen receptor modulator; TSC, tuberous sclerosis. Figure created with Biorender.com.

ET resistance can be de novo or acquired. De novo resistance exists before treatment, and these patients do not respond to first-line ET. In contrast, other tumors may initially respond to ET but later lack response because of acquired resistance. Mechanisms of ET resistance have been described in preclinical and clinical studies, as reviewed previously.³⁰ The most common mutations in HR-positive breast cancer are activating mutations in phosphatidylinositol 3-kinase (PIK3CA) catalytic subunit α (PIK3CA), occurring in approximately 35% of metastatic tumors.³¹ Mutations may occur in other genes in this pathway including AKT1 and phosphatase and tensin homolog (PTEN). Other pathologic mutations include acquired mutations in HER2, BRCA, TP53, and others. In addition, upregulation of FOXA1, c-myc, and cyclin D and epigenetic reprogramming of ER transcription can reduce the efficacy of hormone therapy and promote cancer cell growth and metastasis. The selective pressure from hormone therapy agents, especially aromatase inhibitors (AIs), can lead to acquired mutations in the ER ligand-binding domain, which are present in approximately 25% of recurrent or progressive HR-positive cancers.^32-34 Specifically, gain-of-function mutations in the ER gene ESR1 can lead to constitutive activity of ER in the absence of estrogen and alter transcriptional patterns in the ER pathway, affecting the effectiveness of specific ETs. ESR1 mutations are heterogeneous and represent a complex area of ongoing study, with efforts to identify agents that are effective in tumors with ESR1 mutations.

In addition, cancers that are initially HR-positive can lose ER or PR expression. Because of this evolution, as well as the need to confirm tumor origin, it is important to obtain a biopsy at the time of cancer recurrence to confirm breast cancer histology and receptor status to guide treatment options. Tumor mutations may be identified in either primary (archival) tumor or in biopsies obtained from a metastatic site; acquisition of mutations with recurrence makes evaluation in a metastatic tumor sample preferable. It is now possible to detect cell-free tumor DNA in blood, which has revolutionized the ability to identify tumor alterations with disease progression. Understanding altered signal transduction pathways and genetic driver mutations that promote cancer cell proliferation and resistance has led to the identification of therapeutic targets, such as PIK3CA and others, which we discuss in the subsequent sections in more detail.

THERAPY FOR EARLY STAGE HR-POSITIVE/HER2-NEGATIVE BREAST CANCER

Early stage breast cancer is defined as cancer involving the breast and regional lymph nodes, including the axillary, supraclavicular, and internal mammary nodes, without involvement of distant sites; it is further defined using a staging system that considers tumor size and extent of axillary nodal involvement (stage I, II, and III). Early stage breast cancer is treated with curative intent, with the goal to reduce the risk of future recurrence. A multimodality approach is used, including surgery, radiation, and systemic therapies. Systemic therapy can be administered in either the neoadjuvant (before surgery) or adjuvant (after surgery) setting, as discussed in more detail below; options include ET, chemotherapy, and targeted therapies. The mainstay of treatment for HR-positive/HER2-negative early stage breast cancer is ET, including tamoxifen, AIs, and ovarian function suppression (OFS). Chemotherapy is administered as needed based on biology and extent of disease, as discussed below. Targeted therapies are also sometimes recommended, such as CDK4/6 inhibitors in patients with high-risk disease or the poly(ADP-ribose) polymerase (PARP) inhibitor olaparib for patients with a pathogenic BRCA1 or BRCA2 mutation. In this section, we discuss systemic therapy options for early stage breast cancer, including promising areas of research. Tables 1 and 2 ^35-52, summarize key completed and ongoing clinical trials in early stage HR-positive/HER2-negative breast cancer, respectively, as described in the text below. Figure 2 depicts a general treatment approach to HR-positive/HER2-negative early stage breast cancer based on current available data.

TABLE 1. Phase 3 trials of adjuvant therapies for early stage hormone receptor-positive/human epidermal growth factor receptor 2-negative breast cancer.

Trial	Agent	No. of patients	Randomized	Arms	DFS rate (p)	OS rate or HR for breast cancer mortality (p)	References
Adjuvant endocrine therapy
ATAC	Anastrozole	9366	Yes	TAM vs. anastrozole	10-year DFS, 93.0% vs. 93.9% (p = .01)	HR, 0.95; 95% CI, 0.84–1.06 (p = .4)	Cuzik 2010^35
Combined SOFT/TEXT	OFS	5738	Yes	TAM/OFS vs. AI/OFS	12-year DFS, 80.5% vs. 75.9%; HR, 0.79; 95% CI, 0.70–0.90	12-year OS, 89.1% vs. 90.1%; HR, 0.93; 95% CI, 0.78–1.11	Regan 2022,^46 Pagani 2014^36
Extended adjuvant endocrine therapy
ATLAS	Extended tamoxifen	6846	Yes	5 vs. 10 years of TAM	12.6-year DFS, 79.2% vs. 82.0% (p = .002)	12.6-year OS, 78.9% vs. 81.4% (p = .01)	Davies 2013^37
aTTOM	Extended tamoxifen	6953	Yes	5 vs. 10 years of TAM	9-year DFS, 80.7% vs. 83.3% (p = .003)	9-year OS, 75.5% vs. 73.8% (p = .1)	Gray 2013^38
MA.17	Extended AI (letrozole)	1918	Yes	5 additional years of AI vs. placebo after initial 5 years of TAM	5-year DFS, 95.0% vs. 91.0% (p = .01)	5-year OS, 93.0% vs. 94.0% (p = NS)	Goss 2003,2016^39, 40
NSABP B-33	Extended AI (exemestane)	1598	Yes	5 additional years of AI vs. placebo after initial 5 years of TAM	4-year DFS, 91.0% vs. 89.0% (p = .07)	4-year OS, 97.8% vs. 98.3% (p = NS)	Mamounas 2008^41
IDEAL	Extended AI (letrozole)	1824	Yes	2.5 vs. 5 years of extended letrozole	6.6-year DFS, 82.0% vs. 83.4% (p = .49)	6.6-year HR, 1.04; 95% CI, 0.78–1.38 (p = .79)	Blok 2018^42
NSABP B-42	Extended AI (letrozole)	3966	Yes	5 years of extended letrozole vs. placebo	10-year DFS, 76.1% vs. 72.1% (p = .011)	10-year OS, 86.1% vs. 85.5% (p = .77)	Mamounas 2020^43
TEAM	Sequential AI (exemestane)	9779	Yes	AI monotherapy vs. sequential therapy (TAM→AI)	10-year DFS, 67.0% vs. 67.0% (p = .39)	10-year OS, 74% vs 73% (p = ·74)	van de Velde 2011,^44 Derks 2017^45
SOLE	Extended intermittent AI (letrozole)	4884	Yes	Extended continuous vs. intermittent AI (letrozole)	7-year DFS, 81.4% vs. 81.5%; HR, 1.03; 95% CI, 0.91–1.17	7-year OS, 93.7% vs. 94.3% (p = .16)	Jerusalem 2021^46
Chemotherapy
USORT 9735	TC	1016	Yes	TC × 4 vs. AC × 4	7-year DFS, 81.0% vs. 75.0% (p = .033)	7-year OS, 87.0% vs. 82.0% (p = .032)	Jones 2009^47
ABC	TaxAC	2125	Yes	TaxAC × 6 vs. TC × 6	4-year DFS, 90.7% vs. 88.2% (p = .04)	4-year OS, 95.2% vs. 94.8% (p = .60)	Blum 2017^48
Plan B	EC-T	3198	Yes	EC-T × 4 vs. TC × 6	5-year DFS, 89.6% vs. 89.9% (p = NS)	5-year OS, 94.5% vs. 94.7% (p = NS)	Nitz 2019^49
Adjuvant CDK4/6 inhibitors
PENELOPE-B	Palbociclib	1250	Yes	1 year of AI + palbociclib vs. AI	3-year DFS, 81.2% vs 77.7% (p = .525)	3-year OS, 93.6% vs. 90.5% (p = .420)	Loibl 2021^50
PALLAS	Palbociclib	5760	Yes	2 years of AI + palbociclib vs. AI	3-year DFS, 88.2% vs. 88.5% (p = .51)	Data not mature	Mayer 2021^51
MonarchE	Abemaciclib	5637	Yes	2 years of AI + abemaciclib vs. AI	4-year DFS, 85.8% vs. 79.4% (p = .01)	Data not mature	Johnston 2022^52

• Abbreviations: ABC, advanced breast cancer; AC, doxorubicin/cyclophosphamide; AI, aromatase inhibitor; ATAC, Arimidex, Tamoxifen, Alone, or in Combination trial; CI, confidence interval; CDK4/6, cyclin-dependent kinase 4/6; DFS, disease-free survival; EC-T, epirubicin/cyclophosphamide then docetaxel; HR, hazard ratio; NA, not available; NS, not significant; NSABP, National Surgical Adjuvant Breast and Bowel Project; OFS, ovarian functional suppression; OS, overall survival; TAM, tamoxifen; TaxAC, doxorubicin/cyclophosphamide + taxane; TC, docetaxel/cyclophosphamide.

TABLE 2. Key ongoing phase 3 clinical trials for early stage hormone receptor-positive/human epidermal growth factor receptor 2-negative breast cancer.

Trial name (NCT identifier)	Enrollment criteria	Agent	Phase	Planned/enrolled n	Primary end point
Neoadjuvant
CheckMate 7FL (NCT04109066)	High-risk, stage II–III, ER+/HER2− BC	Nivolumab vs. placebo + neoadjuvant AC/T and adjuvant endocrine therapy	3	506	pCR and EFS
Postneoadjuvant
SASCIA (NCT04595565)	HER2− BC with residual disease after neoadjuvant chemotherapy	Sacituzumab govitecan vs. TPC, defined as capecitabine or platinum-based chemotherapy × 8 or observation)	3	1200	IDFS
Adjuvant therapies
NATALEE (NCT03701334)	Stage II–III, ER+/HER2− BC	Ribociclib + ET vs. ET alone	3	5101	IDFS
lidERA (NCT04961996)	ER+/HER2− early stage BC after chemotherapy and radiation as indicated	Giredestrant vs. TPC ET	3	4100	IDFS
EMBER-4 (NCT05514054)	ER+/HER2− early stage BC after 2–5 years of standard ET	Imlunestrant vs. TPC ET	3	6000	IDFS

• Abbreviations: AC/T, doxorubicin/cyclophosphamide + taxane; BC, breast cancer; EFS, event-free survival; ER, estrogen receptor; ER+ estrogen receptor-positive;/ ET, endocrine therapy; HER2−, human epidermal growth factor receptor 2-negative; IDFS, invasive disease-free survival. NCT, ClinicalTrials.gov identifier; pCR, pathologic complete response rate; TPC, treatment of physician's choice.

FIGURE 2

HR-positive/HER2-negative early stage breast cancer current treatment paradigm. Shown is a treatment paradigm for HR-positive/HER2-negative early stage breast cancer based on current data. *Consider 2 years of adjuvant abemaciclib for patients who have HR-positive/HER2-negative breast cancer with high-risk clinical and pathologic factors, such as a Ki-67 score ≥20%, four or more positive axillary lymph nodes, or from one to three positive axillary lymph nodes with either grade 3 disease or tumors measuring >5 cm. **Consider 1 year of adjuvant olaparib for patients who have stage III HR-positive/HER2-negative breast cancer with a germline BRCA1 or BRCA2 mutation. Also consider the addition of adjuvant bone-modifying agents for high-risk patients. − indicates negative; +, positive; AI, aromatase inhibitor; CDK, cyclin-dependent kinase; ET, endocrine therapy; G, grade, HR, hormone receptor; HER2, human epidermal growth factor receptor 2; pN, pathologic lymph node status; OFS, ovarian function suppression.

Endocrine therapy

Tamoxifen

Tamoxifen is a selective ER modulator (SERM), which is a nonsteroidal triphenylethylene derivative that binds to the ER.⁵³ It has an antiestrogenic effect on mammary epithelium, which makes it useful for the prevention and treatment of breast cancer.⁵⁴ Several early trials demonstrated that adjuvant tamoxifen reduced breast cancer recurrence and death, particularly in patients who received it for 5 years.⁵⁵ In a combined analysis of 55 trials, recurrence rates among women with HR-positive tumors who completed 1, 2, or 5 years of tamoxifen were reduced by 21%, 28%, and 50%, respectively; similarly, 1, 2, or 5 years of tamoxifen resulted in reductions in mortality by 14%, 18%, and 28%, respectively.⁵⁶ These trials established the role of tamoxifen in the treatment of early stage HR-positive breast cancer.

In terms of side effects, tamoxifen can have a proestrogenic effect on the uterine epithelium, which can increase the risk of endometrial cancer with long-term use. In one study from the National Surgical Adjuvant Breast and Bowel Project (NSABP), the incidence of endometrial cancer among individuals taking 20 mg of tamoxifen daily was 1.6 per 1000 patient-years compared with 0.2 per 1000 patient-years among those taking placebo.⁵⁷ Thus individuals treated with tamoxifen should have an annual gynecologic examination and should be advised to report any changes in vaginal bleeding. In addition, tamoxifen can also increase the risk of thromboembolic events,⁵⁶ so it should be avoided in patients with risk factors for venous thromboembolism.

Aromatase inhibitors

AIs are another class of hormone therapy used in the management of HR-positive breast cancer. AIs have antiestrogenic activity by inhibiting the peripheral conversion of adrenally synthesized androstenedione to estradiol through inhibition of the aromatase enzyme.⁵⁸ The first AI to become available in the late 1970s was aminoglutethimide; it was demonstrated to be effective as second-line therapy after tamoxifen in postmenopausal women with HR-positive MBC, but it required concomitant corticosteroid use, so toxicity limited its use.⁵⁹ Next, a steroidal AI called formestane became available; it was more selective than aminoglutethimide but showed no significant benefits over the progestogen megestrol acetate⁶⁰ and had to be given intramuscularly, leading to local site reactions.⁶¹ Most recently, third-generation AIs, which include the nonsteroidal agents anastrozole, letrozole, and fadrozole (Japan only) and the steroidal compound exemestane, became available, and these are the AIs used today. Of note, AIs do not sufficiently suppress estradiol synthesis in the ovaries and thus are not adequate to be used alone in premenopausal women, but they can be used in combination with OFS in this setting.

In the Arimidex, Tamoxifen, Alone, or in Combination (ATAC) trial (ClinicalTrials.gov identifier NCT00849030), postmenopausal women with early stage HR-positive breast cancer were randomized to receive anastrozole or tamoxifen.⁶² In this trial, at a median follow-up of 10 years, anastrozole was superior to tamoxifen in terms of disease-free survival (DFS) (hazard ratio, 0.86; 95% confidence interval [CI], 0.78–0.95; p = .003), time to recurrence (hazard ratio, 0.79; 95% CI, 0.70–0.89; p = .0002), and time to distant recurrence (hazard ratio, 0.85; 95% CI, 0.73–0.98; p = .02).³⁵ However, there was no statistically significant difference in overall survival (OS) between groups at 10 years of follow-up (hazard ratio, 0.95; 95% CI, 0·84–1.06; p = .4).³⁵

Next, randomized trials were designed to evaluate the efficacy of AIs in premenopausal women. The SOFT trial (ClinicalTrials.gov identifier NCT00066690) randomly assigned 3066 premenopausal women to 5 years of adjuvant tamoxifen alone, tamoxifen/OFS, or exemestane/OFS for 5 years; the TEXT trial (ClinicalTrials.gov identifier NCT00066703) was similar and enrolled 2672 premenopausal women but only contained two arms, both of which required OFS: tamoxifen/OFS and exemestane/OFS arms for 5 years. The 12-year follow-up of these combined studies was presented at the 2021 SABCS and demonstrated that the absolute reduction in distant recurrence was 1.8% higher with exemestane/OFS versus tamoxifen/OFS (88.4% vs. 86.6%; p = .003).⁶³ At 12 years, OS was excellent in both groups: 90.1% in patients on exemestane/OFS versus 89.1% in patients on tamoxifen/OFS, which was not statistically different. Importantly, there were meaningful relative reductions in distant recurrence and death with the use of OFS (in combination with either ET) vs. tamoxifen alone, with absolute reductions at 12 years more clinically substantial (approximately 10%) for those at higher clinical risk. The benefit of OFS was particularly important for women younger than 35 years.

In terms of side effects, AIs can cause arthralgias, increased risk of osteoporosis, hot flashes, vaginal dryness, skin dryness, and mood disturbance. Arthralgias can be managed with increased activity/exercise, acupuncture, and nonsteroidal anti-inflammatories as needed. Patients on AIs should undergo routine bone density scans to monitor for bone thinning, and bone-strengthening medications should be started if there is accelerated bone loss.

Sequencing and duration of endocrine therapy

Because HR-positive breast cancer is more likely to have late recurrences, there have been multiple studies that have evaluated the optimal duration of ET. First, the ATLAS (registry number ISRCTN19652633) and aTTom trials (registry number ISRCTN17222211) evaluated the optimal duration of tamoxifen. The ATLAS trial randomized women who successfully completed 5 years of tamoxifen to continue tamoxifen for up to 10 years or stop at 5 years. In this trial, patients who continued tamoxifen for 10 years had a significant reduction in breast cancer recurrence (18% vs. 21%; p = .002) and overall mortality (19% vs. 21%; p = .01).³⁷ The aTTom trial randomized patients from the United Kingdom who successfully completed 5 years of tamoxifen to continue for an additional 5 years or stop at year 5. At 9 years of median follow-up, continuation of tamoxifen reduced breast cancer recurrence (21% vs. 25%) and breast cancer mortality (13% vs. 15%).³⁸ However, there was a higher incidence of endometrial cancers (ATLAS event ratio, 1.74 [95% CI, 1.30–2.34]; and aTTom relative risk [RR], 2.20 (95% CI, 1.31–2.34]) and thromboembolic events (ATLAS event rate ratio, 1.87; 95% CI, 1.13–3.07) with longer duration of tamoxifen treatment.

There have been several studies evaluating whether the addition of an AI after completion of 5 years of tamoxifen improves outcomes. MA.17 (National Cancer Institute of Canada Clinical Trials Group) was a large randomized trial that evaluated whether extended adjuvant therapy with letrozole after the completion of 5 years of standard tamoxifen therapy could prolong DFS in postmenopausal women with HR-positive early stage breast cancer.³⁹ The 5-year DFS rate was 95% in the extended AI arm compared with 91% in the placebo arm.⁴⁰ The NSABP-33 trial also studied the extension of hormone therapy, randomizing patients who received 5 years of tamoxifen to receive 5 years of exemestane versus placebo.⁴¹ The trial was terminated early because of results from the MA.17 trial, which showed benefit from the extension of hormone therapy with letrozole, so participants were unblinded and could cross over to the exemestane arm. With an intention-to-treat analysis, the NSABP B-33 trial demonstrated a nonsignificant reduction in DFS at 4 years (89% vs. 91%; p = .07) and a significant reduction in relapse-free survival (95% vs. 94%; p = .004).⁴¹

There have also been several trials comparing the sequential use of tamoxifen and AIs. The TEAM trial (ClinicalTrials.gov identifier NCT00279448) was a phase 3 trial that randomized individuals to exemestane for 5 years or tamoxifen for 2–3 years followed by exemestane for 5 years. This trial demonstrated no significant difference in the 10-year DFS rate (67% vs 67%; hazard ratio, 0.97; 95% CI, 0.88–1.08; p = .60) between individuals who received exemestane or sequential therapy.⁴⁴ This was also confirmed in the Breast International Group BIG 1-98 trial (ClinicalTrials.gov identifier NCT00004205), which was a phase 3 trial that randomized patients to 5 years of tamoxifen monotherapy, 5 years of letrozole monotherapy, or 2 years of treatment with one agent followed by 3 years of treatment with the other.⁶⁴ After a median follow-up of 12.6 years, there was a nonsignificant 9% relative reduction in DFS with letrozole monotherapy compared with tamoxifen (hazard ratio, 0.91; 95% CI, 0.81–1.01).⁶⁵

In terms of evaluating the duration of ET, the IDEAL trial (EudraCT number 2006-003958-16) was a phase 3 trial that randomized patients to either 2.5 or 5 years of letrozole after receiving any adjuvant ET for 5 years.⁴² After 5 years of initial adjuvant ET, this trial did not show a benefit between 5 versus 2.5 years of additional adjuvant letrozole at 6.6 years of follow-up in terms of DFS (hazard ratio, 0.92; 95% CI, 0.74–1.16) or OS (hazard ratio, 1.04; 95% CI, 0.78–1.38). In the NSABP B-42 trial, patients who had previously received 5 years of an AI or sequential treatment of tamoxifen followed by an AI for 5 years were randomized to receive 5 years of extended letrozole or placebo. In the 10-year analysis of this study presented at the 2019 SABCS, extended adjuvant letrozole resulted in a statistically significant increase in DFS versus placebo (hazard ratio, 0.84; 95% CI, 0.74–0.96; p = .011).⁴³

The question of continuous versus intermittent administration of AI was evaluated in the phase 3 SOLE trial (ClinicalTrials.gov identifier NCT00553410), which compared continuous versus intermittent letrozole in women with ER-positive, lymph node-positive breast cancer who had completed 5 years of adjuvant hormone therapy. At a median follow-up of 7 years, the DFS and OS were not statistically different between groups.⁴⁶

Several prognostic tests have been evaluated to help determine which patients will benefit from extended ET. For example, the Breast Cancer Index is a genomic assay that analyzes the activity of 11 genes to help predict the risk of breast cancer recurrence and the predicted benefit from extended ET in patients with HR-positive/HER2-negative early stage breast cancer. The Breast Cancer Index was first evaluated in the MA.17 trial population and was found to be predictive of benefit from extended hormone therapy in a subgroup of HR-positive patients who remained disease-free after 5 years of tamoxifen,⁶⁶ and it was validated in the aTTom and IDEAL trials.^{67, 68} An additional strategy has been to include clinicopathologic features, such as tumor size, grade, and patient age, to guide extended therapy, such as with the Clinical Treatment Score post-5 years for postmenopausal patients^{69, 70} and others, although further clinical validation is needed. Of note, interesting data from the IDEAL trial suggest that patients with a MammaPrint high-risk score do not derive benefit from extended ET, likely because of the early risk of recurrence (before 5 years) and reduced sensitivity to ET, whereas patients with MammaPrint low-risk scores do derive benefit.⁷¹ In general, extended ET is recommended for patients with a higher risk of recurrence (e.g., N0–N1 disease with high-risk features or N2-positive disease), but treatment duration should be balanced with patient characteristics, comorbidities, preferences, and quality of life.

Chemotherapy

Adjuvant chemotherapy substantially reduces the risk of breast cancer recurrence and death in some patients with HR-positive/HER2-negative early stage breast cancer who have clinically and/or genomically high-risk disease.⁴⁸ Several studies have demonstrated the benefit of chemotherapy in the adjuvant setting and defined the preferred regimens. A meta-analysis of 194 trials from the 1990s demonstrated that 6 months of anthracycline-based polychemotherapy reduced the annual breast cancer death rate by about 38% in women younger than 50 years and by 20% in those between ages 50 and 69 years. Anthracycline-based chemotherapy was shown to be more effective than cyclophosphamide, methotrexate, fluorouracil chemotherapy with regard to recurrence and breast cancer mortality.⁵⁵ Another 15-ear meta-analysis of 123 randomized trials conducted by the Early Breast Cancer Trialists' Collaborative Group demonstrated a 33% reduction in 10-year breast cancer mortality with chemotherapy regimens that used an anthracycline, a taxane, and an alkylator compared with no chemotherapy.⁷²

Because of toxicity concerns with anthracyclines, four cycles of docetaxel/cyclophosphamide (TC) were compared as an alternative to four cycles of doxorubicin/cyclophosphamide (AC).⁷³ At a median of 7 years of follow-up, both DFS and OS were higher with TC compared with AC (DFS: 81% vs. 75%; p = .33; OS: 87% vs. 82%; p = .032).⁴⁷ However, AC was thought to be a weak comparator to TC, so subsequent trials compared AC plus a taxane versus TC. The ABC trials were a series of three trials that evaluated the efficacy AC with a taxane versus TC in patients with node-positive or high-risk lymph node-negative, HER2-negative breast cancer. After a median follow-up of 3.3 years, six cycles of AC plus a taxane had higher invasive DFS (IDFS) compared with six cycles of TC (90.7% vs. 88.2%; p = .04).⁴⁸ In addition, the West German Study Group PlanB trial assessed the efficacy of anthracycline-based regimens with taxanes versus TC. In this multicenter phase 3 trial, patients with node-positive disease or high-risk, node-negative disease were randomized to receive four cycles of epirubicin/cyclophosphamide (EC) followed by four cycles of docetaxel versus six cycles of TC. After a median follow-up of 60 months, there was no statistically significant difference between the EC/docetaxel and TC arms with regard to DFS (89.6% vs. 89.9%) or OS (94.5% vs. 94.7%).⁴⁹ In general, anthracycline-based chemotherapy regimens are still preferred for patients with high-risk early stage HR-positive/HER2-negative disease, but nonanthracycline-based regimens, such as TC for four to six cycles, can be considered an appropriate strategy for patients with lower risk disease (e.g., zero to three positive lymph nodes, smaller tumors), those with a cardiac history in whom anthracyclines may pose a greater risk of cardiac toxicity (e.g., patients with a history of myocardial infarction, uncontrolled hypertension, and/or those who are older), and/or patients who are unwilling to accept the risks associated with anthracycline-based therapy (e.g., cardiac toxicity, secondary leukemia).

Two methods of molecular profiling of tumors, the MammaPrint gene expression score and the Oncotype Dx breast RS test, have been established for clinical use to help determine the risk of recurrence and need for chemotherapy. The MammaPrint (i.e., 70-gene signature) is a diagnostic assay that uses expression levels of the 70 genes to assess distant recurrence risk in early stage breast cancer, providing a binary low-risk or high-risk prediction of breast cancer recurrence.⁷⁴ The prognostic and predictive nature of the MammaPrint gene expression score was first validated in several retrospective studies.^{75, 76} Subsequently, it was evaluated in the prospective Microarray in Node Negative Disease May Avoid Chemotherapy (MINDACT) trial (ClinicalTrials.gov identifier NCT00433589), which included 6693 patients aged 18–70 years with histologically confirmed primary invasive breast cancer and up to three positive lymph nodes who underwent risk assessment by clinical criteria and the 70-gene profile; patients with discordant clinical and genomic predictions were randomly assigned to receive chemotherapy or not.⁷⁷ In this study, patients with high clinical risk and low genomic risk did not have a significant difference in distant metastasis or death based on whether or not they received chemotherapy, with a 5-year distant metastasis-free survival rate that was similar with and without chemotherapy (95.9% vs. 94.4%, respectively; hazard ratio for distant metastases or death, 0.78; 95% CI, 0.50–1.21); these data suggest this assay may identify patients who can safely avoid chemotherapy. An exploratory analysis of patients with HR-positive/HER2-negative disease showed different effects of chemotherapy administration on 8-year distant metastasis-free survival according to age, with a larger absolute benefit in younger women: 93.6% versus 88.6% with and without chemotherapy, respectively, in women aged 50 years or younger, and 90.2% versus 90.0%, respectively, in women older than 50 years.⁷⁸ However, this is an underpowered exploratory analysis and it remains unclear whether the benefit of chemotherapy is caused primarily by chemotherapy-induced OFS.

The Oncotype RS was also validated in several retrospective studies^79-81 and then prospectively validated in the TAILORx study (ClinicalTrials.gov identifier NCT00310180) in 9719 women with HR-positive/HER2-negative, node-negative disease.⁸² In this study, women with an RS score ≤10 received ET alone, those with a score between 11 and 25 were randomized to receive ET alone versus chemotherapy plus ET, and those with a score ≥26 received chemotherapy. Among the 6700 women with an RS from 11 to 25, the IDFS outcomes were similar between those who received ET alone versus those who received chemotherapy plus ET (83% vs. 84%, respectively, at 9 years; hazard ratio, 1.08; 95% CI, 0.94–1.24). In a subgroup analysis of women aged 50 years or younger, chemotherapy plus ET resulted in a lower rate of distance recurrence than ET alone if the RS was 15–20 (percentage-point difference, 1.6% at 9 years) than if the RS was 21–25 (percentage-point difference, 6.5% at 9 years).⁸²

However, it is not clear whether the benefit of chemotherapy is caused in part by OFS, so some clinicians instead recommend ET alone with OFS in premenopausal patients, and this is typically a shared decision between the patient and provider in these cases.

Another prospective trial, RxPONDER (ClinicalTrials.gov identifier NCT01272037) evaluated the role of the RS in patients with HR-positive/HER2-negative, node-positive disease. In this study, women with HR-positive/HER2-negative early stage breast cancer who had from one to three positive lymph nodes and an RS <25 were randomized to receive chemotherapy plus ET versus ET alone.⁸³ Among postmenopausal women, IDFS at 5 years was 91.9% in the ET-only group and 91.3% in the chemotherapy plus ET group, with no chemotherapy benefit (hazard ratio for invasive disease recurrence, new primary cancer, or death, 1.02; 95% CI, 0.82–1.26; p = .89). In contrast, among premenopausal women, IDFS at 5 years was 89.0% with ET only and 93.9% with chemotherapy plus ET (hazard ratio, 0.60; 95% CI, 0.43–0.83; p = .002), with a similar increase in distant relapse–free survival (hazard ratio, 0.58; 95% CI, 0.39–0.87; p = .009).⁸⁴ However, a major concern in the interpretation of the RxPONDER results in premenopausal women is the very low percentage of patients with node-positive disease who received OFS (just 17%) combined with the lack of a comparator arm with just OFS plus ET to address the question of whether the benefit from chemotherapy in premenopausal women could be attributed to its effect of OFS alone; therefore, again, it is important to interpret these data with this in mind. Future studies are planned to more directly compare chemotherapy versus OFS/AI in patients with intermediate-risk RS scores, such as with the NSABP trial BR009, which randomizes patients with N0–N1 disease who have an RS of 16–26 to receive chemotherapy plus OFS/AI for 5 years versus OFS/AI for 5 years.

Targeted therapies

In the metastatic setting, CDK4/6 inhibitors have been shown to improve PFS when used in combination with AIs, as discussed in the metastatic section below. Therefore, it was hypothesized that these agents may reduce the risk of recurrence in patients with high-risk early stage HR-positive/HER2-negative breast cancer. Several studies have evaluated the use of CDK4/6 inhibitors in the neoadjuvant setting. In the phase 2 PALLET trial (ClinicalTrials.gov identifier NCT05163106), the addition of palbociclib to letrozole in the neoadjuvant setting in high-risk postmenopausal patients significantly decreased the cell proliferation rate (Ki-67) but did not increase the clinical response rate over 14 weeks.⁸⁵ In the neoMONARCH study (ClinicalTrials.gov identifier NCT02441946), the addition of abemaciclib to anastrozole in the neoadjuvant setting also decreased the Ki-67 rate after 2 weeks of treatment compared with anastrozole alone.⁸⁶ There have also been studies evaluating ET plus CDK4/6 inhibitors versus chemotherapy in the neoadjuvant setting. In the phase 2 NeoPAL study (ClinicalTrials.gov identifier NCT02400567), patients with HR-positive/HER2-negative, node-positive early stage breast cancer were randomized to either letrozole and palbociclib versus 5-fluorouracil, epirubicin, and cyclophosphamide followed by docetaxel. In this study, patients randomized to letrozole and palbociclib had lower pathologic complete response (pCR) rates (3.8% vs. 5.9%) but similar clinical response rates (75%).⁸⁷ The CORALLEEN trial (ClinicalTrials.gov identifier NCT03248427) had a similar design and evaluated letrozole plus ribociclib versus chemotherapy (AC/paclitaxel), with a primary end point of PAM50 low risk of relapse at the time of surgery; 85% had a high risk of relapse score in the letrozole/ribociclib arm versus 89% in the chemotherapy arm.⁸⁸ Currently, the use of neoadjuvant CDK4/6 inhibitors remains investigational, and the optimal end points to predict long-term outcomes are unknown.

There have also been several trials evaluating the role of adjuvant CDK4/6 inhibitors in high-risk, early stage HR-positive/HER2-negative breast cancer. The PENELOPE-B trial (ClinicalTrials.gov identifier NCT01864746) evaluated the addition of adjuvant palbociclib to ET in patients who had residual disease after treatment with taxane-containing neoadjuvant chemotherapy (NACT). In this trial, 1 year of adjuvant palbociclib did not improve IDFS (hazard ratio, 0.93; 95% CI, 0.74–1.17).⁵⁰ Similarly, the PALLAS trial (ClinicalTrials.gov identifier NCT02513394) studied the addition of 2 years of adjuvant palbociclib to ET in individuals with early stage HR-positive/HER2-negative breast cancer. After a median follow up of 24 months, there was no difference in IDFS in the palbociclib plus AI group compared with the AI-alone group (88.2% vs. 88.5%; p = .51).^{51, 89,} The monarchE trial (ClinicalTrials.gov identifier NCT03155997) randomized individuals with high-risk, HR-positive/HER2-negative breast cancer (defined as four or more positive nodes, one to three positive nodes and tumor size >5 cm, histologic grade 3, or Ki-67 index ≥ 20%) to adjuvant ET with or without abemaciclib. At 42.0 months of follow-up, the addition of abemaciclib to ET resulted in a 33.6% reduction in developing an IDFS event (hazard ratio, 0.71; p = .0009) and a 34.1% reduction in the risk of developing a distant recurrence-free survival event (hazard ratio, 0.659; p < .0001).⁵² Interestingly, the abemaciclib treatment benefit deepened over time, with lower piecewise hazard ratio in terms of IDFS and distant recurrence-free survival in year ≥3 relative to earlier timepoints.⁵² The Ki-67 index was found to be prognostic, although the benefit of abemaciclib was present regardless of the Ki-67 index.⁹⁰ Diarrhea, fatigue, neutropenia, and leukopenia were the most common adverse events (AEs). This trial led to the US Food and Drug Administration (FDA) approval of abemaciclib as the first CDK4/6 inhibitor approved for the adjuvant treatment of early stage breast cancer in October 2021. Finally, the NATALEE trial (ClinicalTrials.gov identifier NCT03701334) is an ongoing phase 3 study evaluating the efficacy of 3 years of adjuvant AI with or without ribociclib in patients who have early stage, HR-positive/HER2-negative breast cancer⁹¹; results from this study are expected in 2023.

For individuals with a germline BRCA1 or BRCA2 mutation, the addition of adjuvant poly(ADP-ribose) polymerase (PARP) inhibitors also has been studied. PARP inhibition causes cell death because of the accumulation of irreparable DNA damage and induces PARP trapping at sites of DNA damage. The phase 3 OlympiA trial (ClinicalTrials.gov identifier NCT02032823) recently demonstrated that 1 year of adjuvant olaparib can improve 3-year IDFS (85.9% vs. 77.1%) in patients with high-risk, HER2-negative early stage breast cancer who have a germline BRCA1 or BRCA2 mutation.⁹² Data presented at the March 2022 European Society for Medical Oncology (ESMO) Virtual Plenary demonstrated that the OS benefit derived with olaparib met the significance threshold in the overall population (stratified hazard ratio, 0.68; 98.5% CI, 0.47–0.97; p = .009).⁹³ It is important to note that HR-positive individuals only constituted about 18% in the olaparib arm and 17% in the placebo arm. Based on these results, in February 2022, the FDA approved the use of adjuvant olaparib for patients with BRCA-mutated, HER2-negative, high-risk early breast cancer who have previously been treated with chemotherapy before or after surgery.

Other therapies

There has also been interest in evaluating the addition of bone modifying agents such as bisphosphonates and the anti-RANK ligand agent denosumab to reduce recurrence risk in patients with HR-positive/HER2-negative early stage breast cancer. The Austrian Breast and Colorectal Cancer Study Group ABCSG-12 trial (ClinicalTrials.gov identifier NCT00295646) evaluated the addition of the bisphosphonate zoledronic acid to OFS with tamoxifen or anastrozole. After a median follow-up of 94.4 months, the relative risks of disease progression (hazard ratio, 0.77; 95% CI, 0.60–0.99; p = .042) and of death (hazard ratio, 0.66; 95% CI, 0.43–1.02; p = .064) were reduced with the addition of zoledronic acid but were no longer significant at the predefined significance level.⁹⁴ In a meta-analysis that included data from 18,766 women in 26 clinical trials of 2–5 years of bisphosphonates, at a median follow-up of 5.6 years, adjuvant bisphosphonates lead to marginal reductions in recurrence (RR, 0.94; 95% CI, 0.87–1.01; p = .08), distant recurrence (RR, 0.92; 95% CI, 0.85–0.99; p = .03), and breast cancer mortality (RR, 0.91; 95% CI, 0.83–0.99; p = .04) but a more convincing reduction in bone recurrence (RR, 0.83; 95% CI, 0.73–0.94; p = .004).⁹⁵ Bisphosphonate treatment had no apparent effect on any outcome in premenopausal women but reduced the risk of recurrence (RR, 0.86; 95% CI, 0.78–0.94; p = .002), distant recurrence (RR, 0.82; 95% CI, 0.74–0.92; p = .0003), bone recurrence (RR, 0.72; 95% CI, 0.60–0.86; p = .0002), and breast cancer mortality (RR, 0.82; 95% CI, 0.73–0.93; p = .002) in postmenopausal women.⁹⁵ In the phase 3 ABCSG-18 study (ClinicalTrials.gov identifier NCT00556374), 3425 postmenopausal women with early stage HR-positive breast cancer were randomized to receive denosumab 60 mg or placebo administered every 6 months. At a median follow-up of 8 years, there were 309 versus 368 DFS events (hazard ratio, 0.83; 95% CI, 0.71–0.97) in the denosumab versus the placebo group, respectively, resulting in an absolute 9-year DFS benefit of 3.5 percentage points (79.4% vs. 75.9%)⁹⁶; however, no trial to date has demonstrated an OS benefit. Based on these data, the NCCN guidelines state that adjuvant bisphosphonate therapy should be considered for risk reduction of distant metastasis for 3–5 years in postmenopausal patients (natural or induced) with high-risk, node-negative or node-positive tumors.⁹⁷ Both bisphosphonates and denosumab are associated with electrolyte disturbances, such as hypocalcemia, atypical fractures, and osteonecrosis of the jaw, so dental clearance is recommended before initiation, and the duration of therapy should be limited.

Finally, there are several novel therapies that are demonstrating promising results in the metastatic setting, so they are now being studied in the early stage setting. For example, in the phase 2 I-SPY2 trial (ClinicalTrials.gov identifier NCT01042379), the addition of pembrolizumab to NACT has been evaluated in patients with HER2-negative early stage breast cancer. In the 40 individuals who had ER-positive disease in this arm, the estimated pCR rate with pembrolizumab was 30% compared with 13% in the control group.^{98, 99} In particular, patients with an immune positive response–predictive subtype molecular signature had a higher RR with the addition of pembrolizumab (4% vs. 69%).²⁰ Novel ETs are also under investigation in the neoadjuvant and adjuvant settings in clinical trials. For example, the EMBER-4 study (ClinicalTrials.gov identifier NCT05514054) evaluates imlunestrant versus standard ET in patients with early breast cancer who have already received at least 2–5 years of standard of care ET.

Neoadjuvant versus adjuvant therapy

In early stage breast cancer, treatment can be administered before surgery (neoadjuvantly) or after surgery (adjuvantly). There appears to be no difference in DFS or OS in adjuvant versus NACT administration for early stage HR-positive/HER2-negative breast cancer.¹⁰⁰ However, NACT administration provides the ability to assess response to therapy and adjust therapy if needed, provides prognostic information, and may allow for breast-conservation surgery in some cases. The goal of NACT is to shrink the tumor and, ideally, to achieve a pCR in which there is no tumor detected in the breast or axillary lymph nodes. The pCR rate is relatively low for ER-positive tumors after NACT—approximately <5%–15% in most studies^{20, 101}—reflecting the relative chemotherapy resistance of most patients who have HR-positive/HER2-negative disease compared with those who have HER2-positive disease and TNBC. Notably, pCR rates are low, but the association with DFS varies based on biologic subtype, with little association for luminal A tumors but with increasing association for more highly proliferative luminal B and basal-like HR-positive tumors. Another way to quantify residual disease after NACT is the residual cancer burden, which categorizes patients from residual cancer burden 0 to residual cancer burden III based on the amount of residual disease and cellularity. Several studies have demonstrated that both the pCR rate and the residual cancer burden are prognostic of DFS and OS in neoadjuvant studies.^102-104

Neoadjuvant ET (NET) has also been evaluated in several clinical trials. The P024 study was a large multi-institutional trial of postmenopausal women with HR-positive/HER-breast cancer who were ineligible for breast-conserving surgery that randomized patients to receive neoadjuvant letrozole or tamoxifen for 4 months.¹⁰⁵ In this trial, letrozole had a significantly higher overall RR (ORR) (55% vs. 36%), and more patients on the letrozole arm were able to undergo breast-conserving surgery (45% vs. 35%).¹⁰⁵ Tumors from the P024 trial were later analyzed for posttreatment ER status, Ki-67 proliferation index, histologic grade, pathologic tumor size, node status, and treatment response. After identification of the factors associated with relapse-free survival, the preoperative endocrine prognostic index (PEPI) score was developed and subsequently validated in the IMPACT trial.^{106, 107} The IMPACT trial (Immediate Preoperative Anastrozole, Tamoxifen, or Combined With Tamoxifen) compared 3 months of neoadjuvant tamoxifen, anastrozole, or the combination in postmenopausal women with early stage, ER-positive/HER2-negative breast cancer. In this study, there was no difference in the RR between anastrozole, tamoxifen, or the combination (37%, 36%, and 39%, respectively); however, more patients on the anastrozole arm were able to undergo breast-conserving surgery (44% vs. 31%).¹⁰⁷ Similarly, the PROACT trial (ClinicalTrials.gov identifier NCT00291525) compared 3 months of neoadjuvant tamoxifen versus anastrozole, demonstrating that anastrozole was superior to tamoxifen with regard to the objective response measured by ultrasound evaluation (36.6% vs. 24.4%).¹⁰⁸ Subsequently, the ACOSOG Z1031 trial (ClinicalTrials.gov identifier NCT00265759) compared 4 months of NET versus exemestane, letrozole, or anastrozole and found no difference in the RR between neoadjuvant AIs.^{109, 110} The ALTERNATE trial (ClinicalTrials.gov identifier NCT01953588) was another study that evaluated whether the endocrine-sensitive disease rate improved with the addition of fulvestrant. Fulvestrant alone or in combination with an AI did not significantly improve the endocrine-sensitive disease rate compared with AI alone (22.7%, 20.5%, and 18.6%, respectively).¹¹¹ In addition, the second primary end point for the ALTERNATE trial is RFS for those with modified PEPI scores of 0 or a pCR who did not receive chemotherapy in the anastrozole arm. This end point was included to evaluate whether achieving a modified PEPI score of 0 or a pCR could be used to help select patients who do not need adjuvant chemotherapy; results are pending. Currently, the use of NET is an evolving area of practice, and more data are needed to define which patients are appropriate candidates for this approach. Little data exist for NET in premenopausal women, but a drop in the Ki-67 index after 4 weeks of combined OFS and ET has been associated with improved outcomes.¹¹² Guidelines suggest that NET can be considered for patients for whom NACT is not deemed to be appropriate and who have a delay to the time of surgery, a desire to try to downstage their tumor before surgery, and/or if the response to NET will help influence decisions about adjuvant chemotherapy, with more data for use in postmenopausal women at this time.¹¹³

Fertility considerations

Premenopausal patients with early stage breast cancer who are interested in future fertility should be referred to a reproductive endocrinologist to discuss oocyte preservation, particularly before receiving chemotherapy or if they will need to be on prolonged ET. At the 2022 SABCS, data from the POSITIVE trial (ClinicalTrials.gov identifier NCT02308085) were presented, which evaluated premenopausal women wishing to become pregnant who had been on 18–30 months of ET for stage I-III breast cancer and were interested in taking a break from ET to try to get pregnant. Of the women in this study, 368 of 497 (74%) had one or more pregnancies with relatively few complications (11%). There was no difference in breast cancer-free interval at the time of data cutoff; 76% of patients resumed ET, but 15% had not yet resumed it.¹¹⁴ These data are encouraging because taking a break from ET for pregnancy appears to be safe, although the caveat is that follow-up for this population is still short, and relatively few patients had stage III disease, so discussions with each patient depending on their individual risk are key.

THERAPY FOR LOCALLY ADVANCED, UNRESECTABLE/METASTATIC, HR-POSITIVE/HER2-NEGATIVE BREAST CANCER

Advanced unresectable and metastatic, HR-positive/HER2-negative breast cancer includes disease that is not surgically resectable or that has spread beyond the axilla to other organs. Metastatic HR-positive/HER2-negative breast cancer is not considered curable but is treatable, and patients can often live for many years on therapy. ET plus a CDK4/6 inhibitor is the current preferred first-line therapy in HR-positive/HER2-negative MBC, followed by other ET and targeted agents. Chemotherapy and ADCs are used once ET and targeted therapy options are exhausted. In this section, we discuss systemic therapy options for locally advanced and metastatic, HR-positive/HER2-negative breast cancer, including promising areas of research. Tables 3–6 ^115-158, summarize key completed and ongoing trials in HR-positive/HER2-negative MBC, and Figure 3 shows a general treatment approach to HR-positive/HER2-negative MBC based on current available data.

TABLE 3. Phase 2/3 trials of endocrine therapy for hormone receptor-positive/human epidermal growth factor receptor 2-negative locally advanced or metastatic breast cancer.

Class	Trial	Agent	Phase	No. of patients	Randomized	Line of therapy	Arms	PFS (p)	OS (p)	References
AI	TARGET study	Anastrozole	3	668	Yes	First	Anastrozole vs. tamoxifen	8.2 vs. 8.3 months (p = NS)	NR	Bonneterre 2000^115
	Nabholtz et al. 2000	Anastrozole	3	353	Yes	First	Anastrozole vs. tamoxifen	11.1 vs. 5.6 months (p = .005)	NR	Nabholtz 2000^116
	Mouridsen et al. 2001, 2003	Letrozole	3	916	Yes	First	Letrozole vs. tamoxifen	9.4 vs. 6.0 months (p < .001)	34 vs. 30 months (p = NS)	Mouridsen 2001, 2003^117, 118
	Meta-analysis of 23 randomized trials	AI	NA	8504	No	First	AI vs. tamoxifen; AI vs. other ET	NR	Superior with AIs vs. tamoxifen [HR, 0.89] and other ET [HR, 0.86]	Mauri 2010^119
SERD	CONFIRM	Fulvestrant	3	736	Yes	Second	Fulvestrant 250 mg vs. 500 mg	6.5 vs. 5.5 months; HR, 0.80 (p = .006)	26.4 vs. 22.3 months; HR, 0.81; 95% CI, 0.69–0.96	Di Leo 2024^120
	FIRST	Fulvestrant	2	205	Yes	First	Fulvestrant vs. anastrozole	23.4 vs. 13.1 months; HR, 0.66; 95% CI, 0.47–0.92	54.1 vs. 48.4 months; HR, 0.70; 95% CI, 0.50–0.98	Robertson 2009, 2012,^121Ellis 2014^123
	FALCON	Fulvestrant	3	462	Yes	First	Fulvestrant vs. anastrozole	16.6 vs. 13.8 months; HR, 0.797; 95% CI, 0.637–0.999 (p = .0486)	NR	Robertson 2016^124
Combination AI + SERD	FACT	Anastrozole ± fulvestrant	3	514	Yes	First	Anastrozole ± fulvestrant	10.2 vs. 10.8 months; HR, 0.99; 95% CI, 0.81–1.20	38.2 vs. 37.8 months; HR, 1.0; 95% CI, 0.76–1.32 (p = 1.00)	Bergh 2012^125
	SWOG SO226	Anastrozole ± fulvestrant	3	707	Yes	First	Anastrozole ± fulvestrant	15 vs. 13.5 months; HR, 0.80; 95% CI, 0.68–0.94	49.8 vs. 42.0 months; HR, 0.82; 95% CI, 0.69–0.98 (p = .03)	Mehta 2019^126

• Abbreviations: ±, with or without; AI, aromatase inhibitor; CI, confidence interval; ET, endocrine therapy; HR, hazard ratio; NA, not available; NR, not reported; NS, not significant; OR, objective response; OS, overall survival; PFS, progression-free survival; SERD, selective estrogen receptor down-regulator; SWOG, Southwest Oncology Group.

TABLE 4. Phase 2/3 trials of cyclin-dependent kinase 4/6 inhibitors for hormone receptor-positive/human epidermal growth factor receptor-negative locally advanced or metastatic breast cancer.

CDK4/6 inhibitor	Trial	Randomized	Phase	Line of therapy	Arms	No. of patients	PFS (p)	OS (p)	References
Palbociclib	PALOMA-1	Yes 1:1	2	First	Palbo/LET vs. LET	165	20.2 vs. 10.2 months; HR, 0.488 (p = .004)	37.5 vs. 34.5 months; HR, 0.89 (p = .281)	Finn 2015^127
	PALOMA-2	Yes 2:1	3	First	Palbo/LET vs. PBO/LET	666	24.8 vs. 14.5 months; HR, 0.58 (p < .0001)	53.9 vs. 51.2 months; HR, 0.956 (p = .3378)	Finn 2016, 2022^128, 129
	PALOMA-3	Yes 2:1	3	Prior ET; one prior chemo with progression^a	Palbo/FUL vs. PBO/FUL	521	9.5 vs. 4.6 months; HR, 0.46 (p < .0001)	34.8 vs. 28.0 months; HR, 0.814 (p = .0221)	Cristofanilli 2016^130
								No prior chemo; 39.3 vs. 29.7 months (p = .008)
Ribociclib	MONALEESA-2	Yes 1:1	3	First	Ribo/LET vs. PBO/LET	668	25.3 vs. 16.0 months; HR, 0.568 (p = 9.63 × 10^−8)	63.9 vs. 51.4 months; HR, 0.76 (p = .004)	Hortobagyi 2016, 2018, 2021^131-133
	MONALEESA-3	Yes 2:1	3	First or second line; no prior chemo	Ribo/FUL vs. PBO/FUL	726	20.5 vs. 12.8 months; HR, 0.59 (p ≤ .001)	53.7 vs. 41.5 months; HR, 0.73 (p = NA)	Slamon 2020, 2018^134, 135
								First line; NR vs. 51.8 months; HR, 0.64
								Second line; 39.7 vs. 33.7 months; HR, 0.78
	MONALEESA-7	Yes 1:1	3	First line; one prior chemo as induction^a	Ribo/ET vs. PBO/ET^b	672	23.8 vs. 13.0 months; HR, 0.55 (p < .0001)	58.7 vs. 48.0 months; HR, 0.76 (p = NA)	Tripathy 2020^136
Abema	MONARCH-1	No	2	Prior ET; one or two lines of chemo	Abema 400 mg daily	132	6.0	22.3	Dickler 2017^137
	MONARCH-2	Yes 2:1	3	Prior ET; no prior chemo	Abema 300 mg daily/FUL vs. PBO/FUL	669	16.4 vs. 9.3 months; HR, 0.553 (p < .001)	46.7 vs. 37.3 months; HR, 0.757 (p = .01)	Sledge 2020^138
	MONARCH-3	Yes 2:1	3	First	Abema 300 mg daily/ANA or LET vs. PBO/ANA or LET	493	28.1 vs. 14.7 months; HR, 0.54 (p = .000002)	NA	Goetz 2017^139

• Abbreviations: Abema, abemaciclib; ANA, anastrozole; CDK4/6, cyclin-dependent kinase 4/6; chemo, chemotherapy; ET, endocrine therapy; EXE, exemestane; FUL, fulvestrant; HR, hazard ratio; LET, letrozole; NA, not available; NR, not reported; OS, overall-survival; Palbo, palbociclib; PBO, placebo; PFS, progression-free survival; Ribo, ribociclib; TAM: tamoxifen.
• ^a One prior line of chemotherapy in the advanced setting allowed.
• ^b Plus gonadotropin-releasing hormone therapy for chemical ovarian suppression.

TABLE 5. Phase 2/3 clinical trials of targeted agents for patients with hormone receptor-positive/human epidermal growth factor receptor 2-negative locally advanced or metastatic breast cancer.

Target	Trial	Agent	Phase	No. of patients	Randomized	Line of therapy	Arms	PFS (p)	OS (p)	References
PI3Ki	SOLAR-1	Alpelisib	3	341	Yes 1:1	Prior ET	ALP/FUL vs. PBO/FUL	11.0 vs. 5.7 months; HR, 0.65 (p < .001)	39.3 vs. 31.4 months; HR, 0.86 (p = .15)	Andre 2019^140
	BYLieve	Alpelisib	2	121	No	Prior ET + CDK4/6i	ALP/FUL, single arm	7.3 months	17.3 months	Rugo 2021^141
AKTi	FAKTION	Capivasertib	3	140	Yes 1:1	Prior ET	FUL/CAP vs. PBO/CAP	10.3 vs. 4.8 months; HR, 0.58 (p = .0044)	26.0 vs. 20.0 months; HR, 0.59 (p = .071)	Jones 2020^142
	IPATunity130, cohort B	Ipatasertib	3	222	Yes 2:1	No prior chemo for ABC or relapse >1 year of NAC	IPAT/PAC vs. PBO/PAC	9.3 vs. 9.3 months; HR, 1.00	NA	Turner 2020^143
mTORi	BOLERO-2	Everolimus	3	724	Yes 2:1	Prior ET	EVE/EXE vs. PBO/EXE	7.8 vs. 3.2 months; HR, 0.45 (p < .0001)	31.0 vs. 26.6 months; HR, 0.89 (p = .14)	Yardley 2013,^144 Piccart 2014^145
	TAMRAD	Everolimus	2	111	Yes 1:1	Prior ET	EVE/TAM vs. TAM	8.6 vs. 4.5 months; HR, 0.54 (p = NA)	NA	Bachelot 2012^146
	PrE0102	Everolimus	2	131	Yes 1:1	Prior ET	FUL/EVE vs. FUL/PBO	10.3 vs. 5.1 months; HR, 0.61 (p = .02)	NA	Kornblum 2018^147
PARPi	OlympiAD	Olaparib	3	302	Yes 2:1	Two or less lines of CT for MBC	Olaparib vs. TPC^a	7.0 vs. 4.2 months; HR, 0.58 (p < .001)	19.3 vs. 17.1 months; HR, 0.90 (p = .513)	Robson 2017, 2019^148, 149
				152, HR+
	EMBRACA	Talazoparib	3	431	Yes 2:1	Less than three lines of CT for MBC	Talazoparib vs. TPC^b	8.6 vs. 5.6 months; HR, 0.54 (p < .001)	19.3 vs. 19.5 months; HR, 0.848 (p = .17)	Litton 2020^150
				241, HR+
	BROCADE3	Veliparib	3	509	Yes 2:1	Two or less lines of CT for MBC	Carbo/PAC/veliparib vs. Carbo/PAC/PBO	14.5 vs. 12.6 months; HR, 0.71 (p = .0016)	33.5 vs. 28.2 months; HR, 0.9, (p = 0.67)	Dieras 2020^151
Anti-VEGF	LEA	Bevacizumab	3	374	Yes 1:1	First	BEV/LET or FUL vs. LET or FUL	19.3 vs. 14.4 months; HR, 0.83 (p = .126)	52.1 vs. 51.8 months; HR, 0.87 (p = 0.518)	Martin 2015^152
	CALGB 40503	Bevacizumab	3	343	Yes 1:1	First; one prior chemo^c	BEV/LET vs. LET	20.2 vs. 15.6 months; HR, 0.75 (p = .016)	47.2 vs. 43.9 months; HR, 0.87 (p = .188)	Dickler 2016^153
HDACi	ENCORE 301	Entinostat	2	130	Yes 1:1	Prior ET	ENT/EXE vs. PBO/EXE	4.3 vs. 2.3 months; HR, 0.73 (p = .055)	28.1 vs. 19.8 months; HR, 0.59 (p = .036)	Yardley 2013^154
	E2112	Entinostat	3	608	Yes 1:1	Prior ET	ENT/EXE vs. PBO/EXE	3.3 vs. 3.1 months; HR, 0.87 (p = .30)	23.4 vs. 21.7 months; HR, 0.99 (p = .94)	Connolly 2021^155
	NCT03538171	Entinostat	3	354	Yes 2:1	Prior ET	ENT/EXE vs. PBO/EXE	6.3 vs. 3.7 months; HR, 0.74 (p < .001)	NA	Xu 2021^156
Anti-TROP2	TROPICS-02	Sacituzumab govitecan	3	543	Yes 1:1	Two to four prior chemos for MBC	Sacituzumab govitecan vs. TPC^d	5.5 vs. 4.0 months; HR, 0.66 (p = .0003)	13.9 vs. 12.3 months; HR, 0.84 (p = .143)	Rugo 2022^157
Anti-HER2 IgG1 MoAb	DESTINY-BREAST 04	Trastuzumab deruxtecan	3	557	Yes, 2:1	One or two lines of prior chemo for MBC; before ET if HR+	Trastuzumab deruxtecan vs. TBC^e	ITT: 9.9 vs. 5.1 months; HR, 0.50 (p < .0001); HR+: 10.1 vs. 5.4 months; HR, 0.51 (p < .0001)	ITT: 23.4 vs. 16.8 months; HR, 0.64 (p = .0010); HR+: 23.9 vs. 17.5 months; HR, 0.64 (p = .0028)	Modi 2022^158

• Abbreviations: ABC, advanced breast cancer; ALP, alpelisib; anti-VEGF, antivascular endothelial growth factor; BEV, bevacizumab; CALGB, Cancer and Leukemia Group B; CAP, capivasertib; Carbo, carboplatin; CDK4/6i, cyclin-dependent kinase 4/6 inhibitor; chemo, chemotherapy; ENT, entinostat; ET, endocrine therapy; EXE, exemestane; EVE, everolimus; FUL, fulvestrant; HDACi, histone deacetylase inhibitor; HER2, human epidermal growth factor receptor 2; HR, hazard ratio; HR+, hormone receptor-positive; IPAT, ipatasertib; ITT, intention to treat; LET, letrozole; MBC, metastatic breast cancer; MoAb, monoclonal antibody; mTORi, mammalian target of rapamycin inhibitor (member of the phosphatidylinositol-3 kinase–related kinase family of protein kinases); NA, not available; NAC, neoadjuvant chemotherapy; NCT, ClinicalTrials.gov identifier; OS: overall survival; PAC, paclitaxel; PARPi, poly(ADP-ribose) polymerase inhibitor; PBO, placebo; PFS, progression-free survival; TAM, tamoxifen; TPC, treatment of physician’s choice.
• ^a TPC included capecitabine, vinorelbine, or eribulin.
• ^b TPC included capecitabine, eribulin, gemcitabine, or vinorelbine.
• ^c One prior line of chemotherapy in the advanced setting allowed; plus gonadotropin-releasing hormone therapy for chemical ovarian suppression.
• ^d TPC included capecitabine, eribulin, gemcitabine, paclitaxel, or nanoparticle albumin-bound-paclitaxel.
• ^e TPC included capecitabine, eribulin, gemcitabine, paclitaxel, or nanoparticle albumin-bound paclitaxel.

TABLE 6. Key ongoing phase 3 clinical trials for hormone receptor-positive/human epidermal growth factor receptor 2-negative metastatic breast cancer.

Trial name (NCT identifier)	Eligibility	Agent	Control arm	Phase	No. planned/enrolled n	Primary end point
Endocrine therapies
EMBER-3 (NCT04975308)	Progression on prior AI/CDK4/6i and eligible for ET	Imlunestrant vs. imlunestrant plus abemaciclib	ET	3	860	PFS
SERENA-4 (NCT04711252)	Previously untreated HR+/HER2− MBC	Camizestrant plus palbociclib	Anastrozole plus palbociclib	3	1342	PFS
SERENA-6 (NCT04964934)	Patients on AI plus CDK4/6i with detection of ESR1 mutation	Camizestrant plus palbociclib or abemaciclib	AI plus CDK4/6i	3	302	PFS
PersevERA (NCT04546009)	Previously untreated HR+/HER2− MBC	Giredestrant plus palbociclib	Letrozole plus palbociclib	3	978	PFS
Post-MONARCH (NCT05169567)	HR+/HER2− MBC after progression on ET plus CDK4/6i	Abemaciclib plus fulvestrant	Placebo plus fulvestrant	3	350	PFS
Targeted therapies
Capitello-292 (NCT04862663)	Must have progressed on prior tamoxifen or AI; less than one prior chemo	Capivasertib plus palbociclib plus fulvestrant	Placebo plus palbociclib plus fulvestrant	1b/3	700	DLT, safety, PFS
ARTEST (NCT04869943)	AR staining ≥40% as assessed by central laboratory	Enobosarm	Exemestane, exemestane plus everolimus, or SERM	3	210	PFS
Antibody-drug conjugates
Destiny-Breast06 (NCT04494425)	ET resistance (first-line chemo)	T-DXd	IC chemo	3	850	PFS
TROPION-Breast01 (NCT05104866)	ET resistance and one or two prior lines of chemo	Dato-DXd	IC chemo	3	700	PFS
Immunotherapy
Keynote B49 (NCT04895358)	Progression on prior ET and now a chemo candidate	Pembrolizumab plus chemo	Placebo plus chemo	3I	800	PFS

• Abbreviations: −, negative; +, positive; AI, aromatase inhibitor; AR, androgen receptor; CDK4/6i, cyclin-dependent 4/6 inhibitor; chemo, chemotherapy; Dato-DXd, datopotamab deruxtecan; DLT, dose-limiting toxicity; ET, endocrine therapy; HR, hormone receptor; IC, investigator's choice; MBC, metastatic breast cancer; NCT identifier, ClinicalTrials.gov identification number; PFS, progression-free survival; SERM, selective estrogen receptor modulator; T-DXd, trastuzumab deruxtecan.

FIGURE 3

HR-positive/HER2-negative metastatic breast cancer treatment paradigm. Shown is a treatment paradigm for HR-positive/HER2-negative metastatic breast cancer based on current data. *If rapidly progressive disease and/or visceral crisis, may also consider chemotherapy as first-line metastatic therapy: consider endocrine therapy plus a targeted therapy as maintenance. BC indicates breast cancer; gBRCA, genetic breast cancer gene mutation; CDK4/6i, cyclin-dependent kinase 4/6 inhibitor; chemo, chemotherapy; ET, endocrine therapy; HER2, human epidermal growth factor receptor 2; HR, hormone receptor; mTORi, mammalian target of rapamycin inhibitor (member of the phosphatidylinositol 3-kinase-related kinase family of protein kinases); PARPi, poly(ADP-ribose) polymerase inhibitor; PIKCA, phosphatidylinositol 3-kinase catalytic subunit α.

Endocrine therapy

Aromatase inhibitors

AIs in combination with CDK4/6 inhibitors are the standard-of-care first-line treatment for patients with HR-positive/HER2-negative MBC. Several historic studies compared the use of AIs versus tamoxifen for HR-positive/HER2-negative MBC. First, in the European Tamoxifen and Arimidex Randomized Group Efficacy and Tolerability (TARGET) trial, the time to progression (TTP) was equivalent between the tamoxifen and anastrozole groups, although only 45% of patients had confirmed HR-positive tumors.¹¹⁵ In an identically designed North American trial that included 89% of patients with confirmed HR-positive tumors, the TTP and the clinical benefit rate (CBR) were significantly better for the anastrozole group compared with the tamoxifen group.¹¹⁶ Another phase 3 study compared letrozole versus tamoxifen and observed improved efficacy with letrozole in terms of TTP (41 vs. 26 weeks), objective response rate (ORR) (30% vs. 20%; p = .006), and CBR (49% vs. 38%; p = .001).¹¹⁷ Letrozole also improved PFS (9.4 vs. 6.0 months; p < .001), and the median OS was slightly prolonged for the letrozole arm (34 vs. 30 months), but this was not statistically significant.¹¹⁸ In addition, in a meta-analysis of 23 randomized trials evaluating AIs versus tamoxifen, AIs were superior to tamoxifen (hazard ratio, 0.89; 95% CI, 0.80–0.99).¹¹⁹ Collectively, these studies demonstrated that AIs are superior to tamoxifen for the treatment of HR-positive/HER2-negative MBC. In small studies of head-to-head comparisons of AIs, the data suggest equivalent efficacy,¹⁵⁹ so any AI is appropriate to use.

Selective estrogen receptor degraders

Fulvestrant is a selective ER degrader (SERD) that blocks ER dimerization and DNA binding, increases ER turnover, and inhibits nuclear uptake of the receptor. It was initially approved as a single-agent monthly injection (250 mg per injection), but the CONFIRM trial (ClinicalTrials.gov identifier NCT00099437) demonstrated that a loading dose plus 500 mg was a more effective dose and was associated with a 19% reduction in the risk of death and a 4.1-month difference in median OS (26.4 vs. 22.3 months; hazard ratio, 0.81; p = .02).¹²⁰ The FIRST trial (Fulvestrant First-Line Study Comparing Endocrine Treatments) and the FALCON trial (ClinicalTrials.gov identifier NCT01602380) demonstrated the efficacy of single-agent fulvestrant in the first-line setting for postmenopausal patients with HR-positive/HER2-negative MBC. The FIRST trial was a phase 2, randomized, open-label, multicenter study of high-dose fulvestrant (500 mg/mg plus 500 mg on day 14 of month 1) versus anastrozole (1 mg daily). The CBR was similar between the fulvestrant and anastrozole arms (72.5% vs. 67.0%, respectively), and the ORR was also similar (36.0% vs. 35.5%, respectively), with no significant difference in the incidence of prespecified AEs.^{121, 122} The median OS was longer in the fulvestrant arm than in the anastrozole arm (54.1 vs. 48.4 months; hazard ratio, 0.70; 95% CI, 0.50–0.98; p = .04).¹²³ In the phase 3 FALCON study, 462 patients with HR-positive MBC who were ET-naive were randomly assigned to receive fulvestrant (500 mg intramuscularly on days 0, 14, and 28 and every 28 days thereafter) or anastrozole (1 mg daily).¹²⁴ The median PFS was significantly longer in the fulvestrant arm (16.6 vs. 13.8 months; hazard ratio, 0.797; p = .0486). The most common AEs were arthralgias (17% fulvestrant vs. 10% anastrozole).

In addition to the evaluation of single-agent fulvestrant, there have also been studies evaluating the efficacy of fulvestrant in combination with AIs. The FACT trial (ClinicalTrials.gov identifier NCT00256698) and Southwest Oncology Group (SWOG) S0226 trials (ClinicalTrials.gov identifier NCT00075764) studied the combination of fulvestrant and anastrozole as first-line treatment for HR-positive/HER2-negative MBC with conflicting results. In the phase 3 FACT study, patients with HR-positive MBC with first relapse on ET were randomized to receive fulvestrant (250 mg plus loading dose regimen) plus anastrozole versus anastrozole alone.¹²⁵ The median TTP was 10.8 versus 10.2 months in the combination arm versus the control arm (hazard ratio, 0.99; 95% CI, 0.81–1.20; p = .91), and the median OS was 37.8 and 38.2 months, respectively (hazard ratio, 1.0; 95% CI, 0.76–1.32;, p = 1.00). In the SWOG S0226 study, patients with HR-positive/HER2-negative MBC were randomized to receive fulvestrant (500 mg plus loading dose) and anastrozole versus anastrozole alone. The median PFS was longer in the combination arm than in the control arm (15.0 vs. 13.5 months; hazard ratio, 0.80; 95% CI, 0.68–0.94; p = .007). The OS was higher in the combination group compared with the anastrozole-only group (49.8 vs. 42.0 months; hazard ratio, 0.82; 95% CI, 0.69–0.98; p = .03).¹²⁶ Of note, the SWOG S0226 study enrolled more endocrine-naive patients and used a higher dose of fulvestrant, which may explain the differences in outcomes. Generally, fulvestrant-AI combination therapy is not currently recommended based on these data.

Novel endocrine therapy agents

Tamoxifen, AIs, and fulvestrant are the only approved endocrine agents for MBC at this time, but several novel ET agents are currently under investigation with the goal of increasing efficacy and limiting toxicity. These agents include oral SERDs, novel SERMs, SERM/SERD hybrids, complete ER antagonist (CERANs), selective ER covalent antagonists (SERCAs), and proteolysis targeted chimera (PROTAC) agents.

Currently, there are at least 10 oral SERDs in development, with additional details provided in other reviews.¹⁶⁰ Several trials have shown encouraging results with oral SERDs to date. In the phase 3 EMERALD trial (ClinicalTrials.gov identifier NCT03778931), the oral SERD elacestrant was administered as monotherapy compared with standard-of-care ET in patients with HR-positive/HER2-negative MBC who had progressed or relapsed on or after one or two lines of ET for advanced disease, one of which was given in combination with a CDK4/6 inhibitor. The trial met both primary end points, demonstrating statistically significant and clinically meaningful extension of PFS versus standard-of-care ET in the overall population (2.79 vs. 1.89 months; hazard ratio,0.70) with an even more clinically meaningful increase in PFS in the ESR1-mutant population (3.78 vs. 1.87 months; hazard ratio, 0.55).¹⁶¹ In the phase 2 SERENA-2 trial (ClinicalTrials.gov identifier NCT04214288), the oral SERD camizestrant (at various doses) was compared with fulvestrant in patients who had ER-positive/HER2-negative MBC and had received from zero to two prior lines of therapy. According to results presented at the 2022 SABCS, at a median follow-up of 16.6 months, this trial also met its primary end point of improvement in PFS (3.7 vs. 7.2 months [hazard ratio, 0.58] at the 75-mg dose and 3.7 vs. 7.7 months [hazard ratio, 0.67] at the 150-mg dose).¹⁶² However, there have also been several trials with oral SERDs that have not met their primary end points. In the phase 2 AMEERA-3 study (ClinicalTrials.gov identifier NCT04059484), the oral SERD amcenestrant was compared with physician’s choice of ET (intramuscular fulvestrant, AI, or tamoxifen) in patients with MBC, but it did not meet its primary end point of improvement in PFS.¹⁶³ The phase 3 trial AMEERA-5 (ClinicalTrials.gov identifier NCT04478266) evaluated amcenestrant plus palbociclib versus letrozole plus palbociclib in patients with HR-positive MBC. In a prespecified interim analysis of this trial, amcenestrant plus palbociclib did not meet the prespecified boundary for continuation compared with the control group, so the Independent Data Monitoring Committee recommended stopping the trial. No new safety signals were observed. Based on these two negative studies, Sanofi withdrew amcenestrant from further development.¹⁶⁴ In the phase 2 study acelERA of the oral SERD giredestrant versus treatment of physician’s choice (TPC) ET monotherapy (ClinicalTrials.gov identifier NCT04576455), there was also no improvement in PFS.¹⁶⁵ It is unclear whether some oral SERD studies have been positive and others negative because of differences in study design versus the differential efficacy of these agents. Adverse side effects of oral SERDs include gastrointestinal side effects (nausea, vomiting, and diarrhea), bradycardia, visual disturbances, hot flashes, arthralgias, and fatigue.

Additional studies with oral SERDs are ongoing. In the metastatic setting, the phase 3 EMBER-3 study (ClinicalTrials.gov identifier NCT04975308), which randomizes patients with HR-positive/HER2-negative MBC after progression on an AI plus a CDK4/6 inhibitor to receive imlunestrant versus ET versus imlunestrant plus abemaciclib. The phase 3 SERENA-4 trial (ClinicalTrials.gov identifier NCT04711252) evaluates patients with a new diagnosis of HR-positive MBC to receive either camizestrant plus palbociclib or anastrozole plus palbociclib in the first line, and the phase 3 persevERA trial (ClinicalTrials.gov identifier NCT04546009) evaluates giredestrant plus palbociclib versus letrozole plus palbociclib in the first line for HR-positive/HER2-negative MBC. The phase 3 SERENA-6 trial (ClinicalTrials.gov identifier NCT04964934) enrolls patients with HR-positive MBC on an AI plus a CDK4/6 inhibitor who have detection of an ESR1 mutation without evidence of disease progression and randomizes them to switch to camizestrant in combination with a CDK4/6 inhibitor versus continuing the AI in combination with a CDK4/6 inhibitor.

Novel oral SERMs and SERCAs are also being evaluated in multiple studies. SERMS do not inactive ER but, instead, modulate ER. Lasofoxifene is a next-generation, nonsteroidal SERM that was originally developed to treat vulvovaginal atrophy and osteoporosis. It is currently being evaluated compared with fulvestrant in patients with HR-positive/HER2-negative MBC who have ESR1 mutations in the phase 2 ELAINE trial (ClinicalTrials.gov identifier NCT03781063) and in combination with abemaciclib in the phase 2 ELAINE-2 study (ClinicalTrials.gov identifier NCT04432454). Preliminary results from ELAINE-2 demonstrated a CBR of 69% at 24 weeks (ORR, 50%) with a PFS of 13 months.¹⁶⁶ There are also mixed SERM/SERD hybrid agents, such as baxedoxifene. This agent is approved in Europe for osteoporosis and showed activity in tamoxifen-resistant xenografts. It is currently being evaluated in combination with palbociclib in HR-positive/HER2-negative MBC (ClinicalTrials.gov identifier NCT02448771). SERCAs target a unique cysteine (C530) in wild-type and mutant ER that are not conserved in other nuclear hormone receptors. H3B-6545 is a first-in-class SERCA that binds ERα irreversibly and enforces a novel antagonist confirmation without degrading ERα. It appears to have increased efficacy in combination with palbociclib. In a phase 1 study, H3B-6545 monotherapy demonstrated an ORR of 17% and a CBR of 29% in a heavily pretreated HR-positive MBC population.¹⁶⁷ Collectively, these data suggest that novel SERMs and SERCAs have a favorable safety profile and promising activity in HR-positive MBC. Larger phase 3 studies are needed to better understand the efficacy of these drugs as single agents and in combination with CDK4/6 inhibitors and other therapies.

CERANs and PROTACs are other novel ETs. These agents do not modulate ER but instead antagonize it; PROTACs also degrade ER. OP-1250 is a first-in-class CERAN that is currently being studied in phase 1/2 studies. Preliminary data for the single-agent use of this agent was presented at the 2021 SABCS, with an ORR of 18% and a CBR 38% in the anticipated recommended phase 2 dose range (60–120 mg).¹⁶⁸ The safety profile included grade 1/2 nausea, fatigue, bradycardia, and grade 3/4 neutropenia. The first-in-class molecule PROTAC ARV-471 (vepdegestrant) was studied in a phase 1 study that was presented at the American Association for Cancer Research meeting in 2021. In a heavily pretreated patient population, including prior CDK4/6 inhibitors and investigational oral SERDS, the CBR was 42%.¹⁶⁹ In the VERITAC phase 2 expansion trial (ClinicalTrials.gov identifier NCT04072952) presented at the 2022 SABCS, among patients at the 200-mg daily dose of ARV-471 (n = 35), the CBR was 37.1%, with a median PFS of 3.5 months in all patients (n = 35) and 5.5 months in patients who had with an ESR1 mutation (n = 19).¹⁷⁰ It was relatively well tolerated with most common AEs including grade 1 and 2 fatigue and nausea.

Targeted therapies

CDK4/6 inhibitors

The development and approval of CDK4/6 inhibitors has transformed the treatment landscape of HR-positive/HER2-negative MBC. Currently, three selective CDK4/6 inhibitor agents, including palbociclib, ribociclib, abemaciclib, are FDA-approved for the treatment of HR-positive MBC.¹²⁷ These agents are given in combination with ET and have demonstrated improvements in PFS and OS.

Multiple preclinical studies have investigated the in vitro sensitivity of CDK4/6 inhibitors in human breast cancer cell lines.¹⁷¹ Promising preclinical results lead to early phase clinical trials of these agents. In the randomized phase 2 PALOMA-1 trial (ClinicalTrials.gov identifier NCT02917005), patients with HR-positive/HER2-negative MBC were randomized to receive letrozole plus palbociclib versus letrozole alone as first-line therapy. In this trial, there was marked improvement in PFS with the addition of palbociclib (20.2 vs. 10.2 months; hazard ratio, 0.488; p < .001),¹²⁷ which lead to the FDA approval of palbociclib in 2015.

Subsequently, phase 3 studies evaluated all three CDK4/6 inhibitors in combination with ET in the first-line metastatic setting, and all three demonstrated improvements in PFS with similar hazard ratios. In the PALOMA-2 phase 3 study (ClinicalTrials.gov identifier NCT01740427), palbociclib was studied in combination with letrozole as first-line therapy, demonstrating an improvement in PFS (24.8 vs. 14.5 months; hazard ratio, 0.58; p < .001) and the ORR (55.3% vs. 44.4%) in the palbociclib arm compared with the letrozole-only arm.¹²⁸ Final OS data were presented at the 2022 ASCO meeting: OS was numerically longer in the palbociclib arm, but the results were not statistically significant (53.9 vs. 51.2 months; hazard ratio, 0.956; p = .3378).¹²⁹ In the MONALEESA-2 trial (ClinicalTrials.gov identifier NCT01958021), ribociclib was evaluated in combination with letrozole as first-line therapy and resulted in an improvement in PFS (25.3 vs. 16.0 months; hazard ratio, 0.56) and the ORR (43% vs. 29%) in the ribociclib arm compared with the letrozole-only arm.^{131, 132} Based on MONALEESA-2, in March 2017, the FDA approved ribociclib in combination with an AI as initial endocrine-based therapy for the treatment of postmenopausal women with HR-positive/HER2-negative advanced breast cancer or MBC. OS data from this trial were presented at the ESMO meeting in 2021, and ribociclib plus letrozole demonstrated a statistically significant OS benefit compared with placebo plus letrozole (63.9 vs. 51.4 months; hazard ratio, 0.76; p = .004),¹³³ and this is the only combination that has shown an OS benefit to date. In the phase 3 MONARCH-3 trial (ClinicalTrials.gov identifier NCT02246621), abemaciclib was studied in combination with an AI (letrozole or anastrozole) as first-line therapy with an improvement in PFS (28.1 vs. 14.7 months; hazard ratio, 0.54) and ORR (59% vs. 44%; p = .004).¹³⁹ The MONALEESA-7 trial (ClinicalTrials.gov identifier NCT02278120) evaluated ribociclib with ET (goserelin with tamoxifen, letrozole, or anastrozole) compared with ET alone in 672 premenopausal or perimenopausal women, which was the only trial that focused on premenopausal women receiving chemical OFS. In this trial, both median PFS (23.8 vs. 13.0 months; hazard ratio, 0.55; p < .0001) and OS (58.7 vs. 48.0 months; hazard ratio for death, 0.76) were improved with the addition of ribociclib.¹³⁶ Based on these findings, the combination of an AI plus a CDK4/6 inhibitor is now the standard of care for first-line, metastatic, HR-positive/HER2-negative breast cancer.

Based on the efficacy of AIs in combination with CDK4/6 inhibitors, it was hypothesized that fulvestrant plus CDK4/6 inhibitors would also be effective in either the first-line or second-line setting. In the phase 3 PALOMA-3 study (ClinicalTrials.gov identifier NCT01942135), patients with HR-positive/HER2-negative MBC who had progressed on ET were randomized to receive fulvestrant plus palbociclib versus fulvestrant plus placebo. Patients who received palbociclib demonstrated an improvement in median PFS of almost 6 months (9.5 vs. 4.6 months; hazard ratio, 0.46; p < .00001).¹³⁰ Of note, the PFS for the fulvestrant control arm in PALOMA-3 was shorter than that for the fulvestrant control arm in MONALEESA-3 (ClinicalTrials.gov identifier NCT02422615) and MONARCH-2 (ClinicalTrials.gov identifier NCT02107703). In the phase 3 MONALEESA-3 study, patients were randomly assigned to receive either ribociclib or placebo in addition to fulvestrant as first-line or second-line treatment. The addition of ribociclib improved both PFS and OS, with an estimated OS at 42 months of 57.8% in the ribociclib group and 45.9% in the placebo group.^{134, 135} In the phase 3 MONARCH-2 trial, patients with HR-positive/HER2-negative MBC who had progressed on ET were randomized to receive fulvestrant with or without abemaciclib. The addition of abemaciclib to fulvestrant significantly improved median PFS (16.4 vs. 9.3 months; hazard ratio, 0.55; p < .0001) and OS (46.7 vs. 37.3 months; hazard ratio, 0.757; p = .01).¹³⁸ The findings from this study led to FDA approval of abemaciclib with fulvestrant in women with HR-positive /HER2-negative advanced breast cancer and progression on prior ET regardless of menopausal status in September 2017.

Of note, CDK4/6 inhibitors appear to be more effective in earlier lines of therapy, particularly when given before chemotherapy. In the phase 3 trials of ribociclib and abemaciclib that did not allow prior chemotherapy, there was an improvement in OS with ribociclib and abemaciclib.^{138, 172} However, OS was improved only in the endocrine-sensitive subset of PALOMA-2 with palbociclib.^{173, 174} Further studies are needed to better understand why this is the case.

In addition to studying the combination of ET and CDK4/6 inhibitors, abemaciclib has also been evaluated as a single agent at a higher dose (400 mg daily). In the single-arm phase 2 MONARCH-1 study (ClinicalTrials.gov identifier NCT02102490), 132 patients with metastatic HR-positive/HER2-negative breast cancer were randomized to receive 200 mg abemaciclib orally twice daily on a continuous schedule until they developed progressive disease or unmanageable toxicity. In this study, 19.7% of patients achieved an objective response, the median PFS was 6 months, and the median OS was 22.3 months.¹³⁷ Based on these results, abemaciclib was also approved by the FDA for use as a monotherapy for patients with refractory, heavily pretreated, HR-positive/HER2-negative MBC who have progressed on prior chemotherapy regimens for MBC.

The efficacy of palbociclib plus capecitabine was evaluated in patients with HR-positive/HER2-negative MBC resistant to AIs in a randomized phase 3 trial.¹⁷⁵ Patients were randomized 1:1 to receive either palbociclib plus exemestane or capecitabine in cohort 1. After discovering resistance to AIs in patients with ESR1 mutations, the trial was modified to add a second cohort. In cohort 2, patients were randomized to palbociclib plus fulvestrant or capecitabine. The trial did not show PFS superiority of palbociclib plus fulvestrant or palbociclib plus ET in patients without ESR1 mutations compared with capecitabine in patients with AI-resistant MBC. The efficacy of the addition of capecitabine to palbociclib plus ET was similar. Toxicity with capecitabine was higher than with ET. Consequently, ET combined with CDK4/6 inhibitors could be the best choice for these patients.

In the phase 3 PADA-1 study (ClinicalTrials.gov identifier NCT03079011), 1017 patients were treated with an AI plus palbociclib in the first-line metastatic setting with evaluation for ESR1 mutations at baseline, 1 month, and every 2 months thereafter. After a median 15.6 months of follow-up, 172 participants developed ESR1 mutations without evidence of disease progression. These patients were randomized either to continue with standard AI plus palbociclib therapy (n = 84) or to switch to fulvestrant plus palbociclib (n = 88).¹⁷⁶ The risk of disease progression or death was 37% lower for the patients who switched to fulvestrant. The median PFS was 11.9 months in the fulvestrant arm and 5.7 months in the standard therapy arm. Further work is needed to evaluate this strategy.

CDK4/6 inhibitors are generally well tolerated, and AEs can be managed with dose modification and supportive care. Hematologic toxicities, particularly neutropenia (approximately 20%–80%), are common with all three CDK4/6 inhibitors, especially palbociclib and ribociclib.¹⁷⁷ Neutropenia can be managed with dose interruption or reduction, and serious events such as febrile neutropenia are rare (<2%). Palbociclib and ribociclib are administered 21 days on followed by 7 days off, which allows for recovery of the hematologic precursors, and myeloid growth factors (granulocyte-colony–stimulating factor) are not recommended. Abemaciclib, which is more selective to CDK4 and CDK6, is administered continuously because of the lower risk of hematologic toxicities.¹⁷⁸ Other than hematologic toxicities, each agent has unique toxicities that may require special monitoring. For example, abemaciclib often causes diarrhea (approximately 80%), requiring the use of antidiarrheal medications in many patients. Ribociclib may cause hepatotoxicity, so liver function tests should be monitored at baseline and throughout therapy. Ribociclib has also been associated with reversible prolongation of the QT interval (2%–7%),^{131, 135, 179} so patients should undergo baseline electrocardiogram and electrolyte monitoring.¹³² CDK4/6 inhibitors cause toxicities, including fatigue, nausea, vomiting, stomatitis, alopecia, rash, diarrhea, and decreased appetite, which are generally mild.

Several studies have evaluated the efficacy of continuing CDK4/6 inhibitors after progression on prior CDK4/6 inhibitors. A multicenter retrospective analysis showed a PFS of 5.8 months for patients who received abemaciclib plus ET after progression on palbociclib or ribociclib, which was suggestive of potential benefit.¹⁸⁰ The phase 2 trial MAINTAIN (ClinicalTrials.gov identifier NCT02632045) randomized 120 patients with HR-positive/HER2-negative MBC whose disease had progressed on ET plus any CDK4/6 inhibitor to receive fulvestrant or exemestane with or without ribociclib. Data presented at the ASCO 2022 meeting at a median follow-up of 18.2 months showed that the median PFS for patients in the ribociclib group was 5.29 versus 2.76 months for patients in the placebo group, suggesting that switching CDK4/6 inhibitors after progression on prior CDK4/6 inhibitor improved outcomes.¹⁸¹ The phase 2 BioPER study (ClinicalTrials.gov identifier NCT03184090) evaluated the efficacy of continuing palbociclib plus ET beyond progression on prior palbociclib-based regimens for HR-positive/HER2-negative MBC. Among 33 patients enrolled, the CBR was 34.4%, and 13.0% of tumors showed loss of retinoblastoma protein expression, meeting both primary end points.¹⁸² The phase 2 PACE trial (ClinicalTrials.gov identifier NCT03147287) evaluated fulvestrant alone, fulvestrant plus palbociclib, or fulvestrant, palbociclib, and avelumab in patients with HR-positive/HER2-negative MBC whose disease progressed on prior ET plus a CDK4/6 inhibitor (91% had received prior palbociclib). Data presented at the 2022 SABCS demonstrated that the median PFS was not improved by continuing palbociclib (4.8 months of fulvestrant monotherapy vs. 4.6 months of fulvestrant plus palbociclib)¹⁸³; it is unclear whether the change in CDK4/6 inhibitor is important, so we await further data from phase 3 studies. Interestingly, in PACE, there was a longer PFS with fulvestrant plus palbociclib plus avelumab (8.1 months) without significant increased immune toxicity in this study, so this finding warrants further investigation.¹⁸³ In addition, the ongoing phase 1/2 MORPHEUS trial (ClinicalTrials.gov identifier NCT03280563) is investigating the safety and efficacy of atezolizumab and fulvestrant combined with multiple agents, including CDK4/6 inhibitors, after progression on a prior CDK4/6 inhibitor. Of note, the studies described here were not powered to assess either the benefit of continuing CDK4/6 inhibitor after progressive disease or the value of switching to another CDK4/6 inhibitor, so, at this point, they are hypothesis generating only, and further work is needed to define the optimal clinical approach. Ongoing studies aim to address this question, such as the phase 3 postMONARCH study (ClinicalTrials.gov identifier NCT05169567).

Phosphoinositide 3-kinase inhibitors

The PI3K/AKT/mammalian target of rapamycin (mTOR) pathway regulates important cellular functions, such as growth, proliferation, differentiation, translational regulation of protein synthesis, glucose metabolism, cell migration, angiogenesis, and survival, and it is a key mechanism of oncogenesis.¹⁸⁴ Activation of this signaling pathway was reported to play a role in endocrine resistance, leading to the investigation of this pathway as a potential therapeutic target.^{185, 186}

Studies evaluating the activity of pan-class I PI3K inhibitors including, buparlisib and pictilisib, resulted in disappointing efficacy and severe toxicity.^187-189 In the phase 3 SANDPIPER trial (ClinicalTrials.gov identifier NCT02340221), postmenopausal patients with disease recurrence or progression after AI treatment were randomized to receive the β-sparing PIK3 inhibitor taselisib versus placebo combined with fulvestrant.^{189, 190} The median PFS was 7.4 months in the taselisib arm versus 5.4 months in the placebo arm (hazard ratio, 0.70; p = .004). The taselisib combination arm had a significantly higher ORR (28% vs. 11.9%) in patients with PIK3CA mutations. However, many patients experienced diarrhea and hyperglycemia, leading to drug discontinuation.

Alpelisib is the first selective PI3Kα inhibitor studied as a single agent that showed activity against tumors with PIK3CA mutations. In a phase 1a trial, alpelisib and ET demonstrated clinical benefit in HR-positive MBC with acquired resistance to ET, particularly in PIK3CA-mutated cancers.¹⁹¹ Subsequently, the phase 3 SOLAR-1 trial (ClinicalTrials.gov identifier NCT02437318) evaluated fulvestrant plus alpelisib (300 mg once daily) versus placebo in 572 postmenopausal women and men with HR-positive/HER2-negative advanced breast cancer that progressed on AI treatment. The median PFS of patients with a PIK3CA mutation was significantly better in the alpelisib plus fulvestrant arm compared with the placebo plus fulvestrant arm (11 vs. 5.7 months; hazard ratio, 0.65; p < .001).¹⁴⁰ The ORR was also higher in the alpelisib arm than in the control arm (26.6% vs. 12.8%). However, there was no benefit of adding alpelisib in patients without a PIK3CA mutation (7.4 vs. 5.6 months). The OS result in the PIK3CA-mutant cohort did not meet the prespecified criteria for statistical significance, although the absolute difference was 8 months (39.3 vs. 31.4 months; p = .15). AEs caused by alpelisib were associated with specific p110α inhibition and included hyperglycemia (63.7% vs. 9.8%), diarrhea (57.7% vs. 15.7%), and rash (35.6% vs. 5.9%). Fasting plasma glucose should be assessed weekly for at least 2 weeks and every 4 weeks thereafter, in addition to periodic hemoglobin A1c monitoring. Early interventions, such as metformin and diet regulation, can help patients to continue alpelisib. Prophylactic antihistamines are recommended to reduce the frequency and severity of skin rash. Based on the results of the SOLAR-1 trial, the FDA approved the combination of alpelisib plus fulvestrant for patients with HR-positive/HER2-negative PIK3CA-mutated advanced breast cancer and MBC in 2019.

When the SOLAR-1 trial was designed, CDK4/6 inhibitors were not yet approved, and only 20 patients had previously been treated with a CDK4/6 inhibitor. Therefore, it was unknown whether alpelisib with ET is effective and safe after progression on prior CDK4/6 inhibitors. Therefore, the phase 2 BYLieve trial (ClinicalTrials.gov identifier NCT03056755) was designed as a single-arm study to address this question. Among patients who had progression on a prior CDK4/6 inhibitor, alpelisib was still effective, with a median PFS of 7.3 months, which was consistent with the SOLAR-1 subgroup analysis.¹⁴¹ Fewer treatment discontinuations were reported in BYLieve (2%) than in SOLAR-1 (25%).^{140, 141}

Inavolisib is an investigational small molecule that selectively inhibits mutant PI3Kα. It has a unique mechanism of action leading the degradation of mutant p110α protein.¹⁹² Early phase studies with inavolisib demonstrated antitumor activity and a manageable safety profile in patients with PIK3CA-mutant breast cancer.^{193, 194} The safety and efficacy of inavolisib plus palbociclib and fulvestrant versus placebo plus palbociclib and fulvestrant is being evaluated in patients with PIK3CA-mutant, HR-positive/HER2-negative, locally advanced breast cancer and MBC in an ongoing phase 3 study (ClinicalTrials.gov identifier NCT04191499).

AKT inhibitors

AKT pathway activation occurs in many HR-positive/HER2-negative breast cancers through alterations in PIK3CA, AKT1, and PTEN; and AKT signaling has been implicated in the development of ET resistance. Therefore, it was hypothesized that AKT inhibitors may be effective for HR-positive/HER2-negative MBC. In the phase 2 FAKTION study (ClinicalTrials.gov identifier NCT01992952), patients with ET-resistant, HR-positive/HER2-negative MBC were randomized to receive fulvestrant plus the AKT1–AKT3 isoform inhibitor capivasertib versus fulvestrant plus placebo. The addition of capivasertib improved both PFS and OS, with the most benefit seen in the pathway-altered population.¹⁴² Based on this study, the phase 3 CAPItello-291 trial (ClinicalTrials.gov identifier NCT04305496) evaluated the combination fulvestrant plus capivasertib versus fulvestrant plus placebo in patients who had recurrence or progression while on an AI for MBC or were <12 months from the end of adjuvant AI. According to data presented at the SABCS in 2022, the addition of capivasertib improved PFS in the overall population (7.2 vs. 3.6 months; hazard ratio, 0.60; p < .001) and in the AKT pathway–altered population (7.3 vs. 3.1 months; hazard ratio, 0.50; p < .001).¹⁹⁵ The trial was not powered to look at the subgroup of patients with nonaltered tumors, and some patients had an unknown mutation profile, so further work is needed to understand the efficacy in this group. Benefit was seen across subgroups, including those who received a prior CDK4/6 inhibitor and had visceral metastases. OS data are immature at this time. The most common grade 3–4 AEs were diarrhea (9.3%) and rash (12%). CAPItello-292 (ClinicalTrials.gov identifier NCT04862663) is currently underway evaluating fulvestrant/palbociclib with or without capivasertib.

In addition to capivasertib, the highly selective oral ATP-competitive AKT inhibitor, ipatasertib, is also under evaluation. In the phase 2 LOTUS study (ClinicalTrials.gov identifier NCT04157270), ipatasertib plus paclitaxel was compared with placebo plus paclitaxel as first-line therapy for metastatic TNBC, and an improved PFS was observed (6.2 vs. 4.9 months; hazard ratio, 0.60; p = .037).¹⁹⁶ A similar combination was investigated in the IPATunity130 trial (ClinicalTrials.gov identifier NCT03337724) to evaluate the activity of ipataserib plus paclitaxel for PIK3CA/AKT1/PTEN–altered advanced breast cancer in patients with both TNBC and HR-positive/HER2-negative disease. Recently, the results from IPATunity130 cohort B, enrolling HR-positive/HER2-negative patients who were not eligible for ET, was reported: the median PFS was identical in the two arms (9.3 months; hazard ratio, 1.00), and OS data were immature.¹⁴³ Ipatasertib is currently being evaluated in combination with ET and palbociclib in patients with HR-positive/HER2-negative MBC in the phase 1 TAKTIC trial (ClinicalTrials.gov identifier NCT03959891). The phase 2 FINER trial (ClinicalTrials.gov identifier NCT04650581) is also ongoing, which is evaluating the safety and efficacy of ipatasertib plus fulvestrant in patients with HR-positive/HER2-negative MBC who progressed after first-line therapy with an AI and a CDK4/6 inhibitor. Finally, the phase 1b/2 IPATunity 150 trial (ClinicalTrials.gov identifier NCT04060862) is evaluating the safety of ipatasertib in combination with palbociclib and fulvestrant in patients with HR-positive/HER2-negative MBC who relapsed during adjuvant ET or progressed during the initial 12 months of first-line ET in the metastatic setting.

mTOR inhibitors

mTOR belongs to the PI3K-related kinase family and is involved in the regulation of cell growth, aging, and metabolism. Everolimus is an oral mTOR inhibitor that has been approved in combination with exemestane in patients with HR-positive MBC based on findings in the BOLERO-2 trial (ClinicalTrials.gov identifier NCT00863655). In this study, the addition of everolimus to exemestane in patients who progressed on nonsteroidal AI improved PFS to 7.8 months versus 3.2 months with exemestane alone (hazard ratio, 0.45; p < .0001) and numerically prolonged OS by 4.4 months, but without reaching statistical significance (hazard ratio, 0.89; p = .14).^{144, 145} The most common toxicity was stomatitis, with grade 3 stomatitis seen in 8% of patients. A subsequent study demonstrated that preventive steroid mouthwash significantly decreased the incidence of stomatitis (2% with steroid mouthwash vs. 33% without steroid mouthwash).¹⁹⁷ These results led to the FDA approval of everolimus with exemestane in 2012. In the phase 2 TAMRAD study (ClinicalTrials.gov identifier NCT01298713), everolimus plus tamoxifen improved the CBR and PFS compared with tamoxifen alone (61.1% vs. 42.1% and 8.6 vs. 4.5 months, respectively).¹⁴⁶ In addition, the PRE0102 phase 2 study (ClinicalTrials.gov identifier NCT01797120) evaluated the combination of fulvestrant plus everolimus versus fulvestrant plus placebo and demonstrated an improvement in PFS with the addition of everolimus (10.3 vs. 5.1 months; hazard ratio, 0.61; p = .02).¹⁴⁷ Finally, preclinical work demonstrated that targeting PI3K in combination with CDK4/6 inhibitors and ET can overcome acquired resistance to CDK4/6 inhibitors.¹⁹⁸ Based on this work, the phase 1/2 TRINITI-1 study (ClinicalTrials.gov identifier NCT02732119) evaluated triplet therapy with ribociclib, exemestane, and low-dose everolimus in patients with ET-refractory, HR-positive/HER2-negative MBC who had progressed on a CDK4/6 inhibitor. In this single-arm study, the CBR was 41.1% at week 24,¹⁹⁹ meeting its primary end point, but further studies are needed to evaluate its efficacy compared with standard regimens.

PARP inhibitors

Approximately 5% of patients with breast cancer carry a germline deleterious mutation in BRCA1 or BRCA2.²⁰⁰ Two PARP inhibitors, olaparib and talazoparib, are FDA-approved for the treatment of patients with deleterious or suspected deleterious, germline BRCA-mutated, HER2-negative MBC. The phase 3 OlympiAD study (ClinicalTrials.gov identifier NCT02000622) evaluated olaparib versus TPC (capecitabine, vinorelbine, or eribulin) in patients with germline BRCA1 or BRCA2 mutations and HER2-negative MBC. The median PFS was significantly improved with olaparib versus TPC (7.0 vs. 4.2 months; hazard ratio, 0.58; p < .001).¹⁴⁸ For the HR-positive cohort, the ORR was 65.4% compared with 36.4% in the control arm, demonstrating the significant activity of olaparib. There was no significant difference in median OS with olaparib versus TPC (19.3 vs. 17.1 months; hazard ratio, 0.90; p = .513).¹⁴⁹ In an exploratory subgroup analysis, olaparib showed a greater OS benefit than TPC in the first-line setting for metastatic disease (22.6 vs. 14.7 months; hazard ratio, 0.51). However, the OS benefit was similar in the second-line or third-line setting for metastatic disease.¹⁴⁹ This could be explained by the development of resistance to the medication, which increases by two-fold to three-fold for the subsequent line of therapy. The olaparib arm had fewer grade ≥3 AEs and also a lower rate of treatment discontinuation compared with the TPC arm. The most common grade 1/2 AEs with olaparib were nausea (58%) and anemia (40%) and the grade 3 AEs were anemia (16.1%) and neutropenia (9.3%). Myelodysplastic syndrome/acute myeloid leukemia were rarely seen in long-term follow-up (<1.5% and <1% respectively).

Subsequently, the phase 3 EMBRCA trial (ClinicalTrials.gov identifier NCT01945775) evaluated talazoparib in the same setting. In this trial, 431 patients with advanced breast cancer were randomized 2:1 to receive talazoparib (n = 287) or TPC (capecitabine, eribulin, and gemcitabine or vinorelbine; n = 144). PFS was significantly improved with talazoparib compared with single-agent chemotherapy (8.6 vs. 5.6 months; hazard ratio, 0.54; p < .001).²⁰¹ The efficacy of talazoparib was identical for HR-positive and HR-negative patients, with a hazard ratio for PFS of 0.47 and an ORR of 63.2% in HR-positive disease. The median OS was 19.3 months in the talazoparib group and 19.5 months in the standard therapy group (p = .17). The most common AEs were anemia, fatigue, and nausea.

PARP inhibitors are also being investigated in combination with other therapies in the metastatic setting. In the phase 3 BROCADE3 study (ClinicalTrials.gov identifier NCT02163694) in patients with BRCA-mutated, HER2-negative advanced breast cancer, the addition of veliparib to carboplatin and paclitaxel improved median PFS compared with placebo (14.5 vs. 12.6 months; hazard ratio, 0.71; p = .0016).¹⁵¹ These findings suggest that veliparib monotherapy may be beneficial after a discontinuation of combination therapy with veliparib plus carboplatin and paclitaxel. Ongoing studies are evaluating the role of PARP inhibitors in patients with nongermline BRCA homologous recombination repair mutations, homologous recombination deficiency, or in those without documented germline BRCA mutations.

Other emerging targeted therapies

Angiogenesis and the vascular endothelial growth factor signaling pathway may also have a role in resistance to ET,²⁰² so there has also been interest in targeting this pathway. In the phase 2 LEA trial (EudraCT number GEICAM/2006-11_GBG_51) and the Cancer and Leukemia Group B 40503 study (ClinicalTrials.gov identifier NCT00601900), the vascular endothelial growth factor inhibitor bevacizumab plus ET was compared with ET alone. In the LEA trial, the addition of bevacizumab numerically improved PFS, but this difference was not statistically significant.¹⁵² In the Cancer and Leukemia Group B 40503 trial, the addition of bevacizumab improved PFS in HR-positive/HER2-negative MBC.¹⁵³ A pooled analysis of the two trials demonstrated an improvement in PFS (19.0 vs. 14.3 months; hazard ratio, 0.77), in the ORR (61% vs. 40%) and in the CBR (77% vs. 64%) with ET plus bevacizumab compared with ET. Grade 3–4 hypertension, proteinuria, cardiovascular toxicity, and liver toxicity were significantly higher in the bevacizumab arm.²⁰³ Based on these findings, the role of bevacizumab for the treatment of HR-positive MBC remains unclear.

Another interesting class of medications are histone deacetylase (HDAC) inhibitors, which are thought to modulate epigenetic modifications that lead to ET resistance. In vitro studies reported that HDAC inhibitors enhance the activity of and restore the sensitivity to ET in ER-positive cell lines.²⁰⁴ Based on these results, the ENCORE301 trial (ClinicalTrials.gov identifier NCT00676663) was designed to evaluate the activity of the HDAC inhibitor entinostat plus exemestane versus placebo plus exemestane in patients with MBC who had progressed on a prior AI. The addition of entinostat to exemestane did not improve PFS (4.3 vs. 2.3 months; hazard ratio, 0.73) but significantly improved OS (28.1 vs. 19.8 months; hazard ratio, 0.59), which was an exploratory secondary end point.¹⁵⁴ The most common AEs with entinostat were neutropenia and fatigue. In the phase 3 E2112 trial (ClinicalTrials.gov identifier NCT02115282), 608 patients with HR-positive MBC who had disease progression after nonsteroidal AI were randomly assigned to receive entinostat or placebo with exemestane. The combination of exemestane and entinostat did not improve PFS (3.3 vs. 3.1 months), ORR (5.8% vs. 5.6%), or OS (23.4 vs. 21.7 months).¹⁵⁵ In contrast, in a phase 3 study conducted in China presented at the 2021 SABCS, the addition of entinostat to exemestane improved PFS compared with placebo plus exemestane (6.3 vs. 3.7 months; hazard ratio, 0.74; 95% CI, 0.58–0.96; p = .021).¹⁵⁶ However, most patients in this study did not receive a prior CDK4/6 inhibitor, so the clinical implications of this study in the post-CDK4/6 era are unclear.

Fibroblast growth factor receptor (FGFR) inhibitors emerged as a possible new therapeutic option based on several studies reporting that FGFR gene amplification was associated with ET resistance to CDK4/6 inhibitors.²⁰⁵ In the PALOMA-3 study, baseline FGFR1 gain in circulating tumor DNA (ctDNA) was associated with worse PFS in both the palbociclib/fulvestrant and placebo/fulvestrant arms. In the MONALEESA-2 study, FGFR1 amplification detected in ctDNA was associated with a shorter PFS with ribociclib. However, in MONALEESA-3 and MONALEESA-2, ribociclib had activity regardless of baseline ctDNA FGFR1 alterations. Several phase 1 trials evaluating pan–FGF-R kinase inhibitors were recently completed (ClinicalTrials.gov identifiers NCT01004224, NCT01928459), and other phase 1/2 trials are ongoing.

Another mechanism of resistance to ET is the acquisition of HER2 mutations, which can occur in approximately 2% of patients with MBC overall and more frequently in those with lobular cancer.²⁰⁶ In a small study, HER2-activating mutations were identified in eight of 168 patients with resistance to ET and were mutually exclusive with ESR1 mutations.²⁰⁷ In a single-arm phase 2 trial, patients with HER2-mutated, nonamplified MBC were treated with neratinib: the CBR was 31%, and the median PFS was 16 weeks.²⁰⁸ In the SUMMIT basket trial (ClinicalTrials.gov identifier NCT01953926) evaluating neratinib in HER2-mutated and HER3-mutated cancers, neratinib had the greatest activity in HER2-mutant breast cancer, with an 8-week ORR of 32%. In another study, the combination of neratinib and fulvestrant was evaluated in 47 patients with HR-positive/HER2-negative breast cancer: the ORR was 29.8%, and the CBR was 36.4%.²⁰⁹ In the phase 2 plasmaMATCH trial (ClinicalTrials.gov identifier NCT03182634), the combination of neratinib with fulvestrant showed clinical activity in patients with HER2-mutated, HR-positive advanced breast cancer.²¹⁰

In normal breast tissue, androgens inhibit cellular proliferation. About 70%–95% of breast cancers are androgen-receptor (AR)-positive, and AR agonists have been shown to inhibit tumor growth. Enobosarm is a nonsteroidal, selective AR agonist. In the phase 2 ARTEST trial (ClinicalTrials.gov identifier NCT02463032), 136 postmenopausal patients with heavily treated, AR-positive, ER-positive/HER2-negative MBC were treated with enobosarm 9 mg and enobosarm 18 mg. In this study, the 9-mg dose was more tolerable and had similar efficacy.²¹¹ Among 50 patients in the 9-mg cohort, the CBR at 24 weeks was 32%, and, in the 18-mg arm, it was 29%. Given these promising results, enobosarm was granted fast-track designation by the FDA for the treatment of patients with AR-positive/HR-positive/HER2-negative MBC. The phase 3 ARTEST trial is underway to evaluate the efficacy and tolerability of enobosarm (ClinicalTrials.gov identifier NCT04869943).

Chemotherapy

Historically, chemotherapy was sometimes administered before ET and CDK4/6 inhibitors in patients with visceral metastases. However, the phase 2 RIGHT Choice trial (ClinicalTrials.gov identifier NCT03839823) was the first prospective study to compare these options directly. This study enrolled patients with HR-positive/HER2-negative MBC who had symptomatic visceral metastases, marked symptomatic nonvisceral metastases, and/or rapid disease progression or impending visceral compromise and who were treatment-naive and randomized them to receive ribociclib plus letrozole or anastrozole (with or without goserelin) versus investigator's choice of chemotherapy. The results were presented at the 2022 SABCS and demonstrated that the ORR and CBR were similar between groups, there was no difference in the time to response, and there was less toxicity in the ribociclib plus ET arm.²¹² These results suggest that ribociclib plus ET should be used instead of chemotherapy for most patients with HR-positive/HER2-negative disease, even for those who have visceral disease.

Therefore, chemotherapy is typically used once a patient has exhausted all ET and targeted therapy options. Combination chemotherapy regimens may have a higher RR but also often have increased toxicity, so sequential monotherapy single-agent regimens are generally preferred. Various chemotherapy agents were used to treat MBC, including anthracyclines, taxanes, capecitabine, vinorelbine, gemcitabine, eribulin, ixabepilone, and platinum agents for patients with germline BRCA1 or BRCA2 mutations. In clinical practice, optimal sequencing of these agents is not clear. Toxicity profiles and patient preferences are important in the decision of chemotherapy agents. Chemotherapy agents are given until patients develop disease progression or unacceptable toxicity.

There are many options for single chemotherapy agents that have been shown to be efficacious in patients with HR-positive/HER2-negative MBC. Taxanes, including paclitaxel, nanoparticle albumin-bound paclitaxel, and docetaxel, are frequently used in the metastatic setting. Weekly dosing regimens have favorable toxicity profile compared with every-3-week doses. Nanoparticle albumin-bound-paclitaxel has some advantages because it does not require steroid premedication and has a lower risk of hypersensitivity reactions. Doxorubicin and liposomal doxorubicin have been shown to have similar efficacy for patients with MBC. Liposomal doxorubicin has a less frequent dosing schedule and decreased risk of cardiotoxicity, decreased rate of nausea and vomiting, and lower rates of alopecia and neutropenia. However, it can cause higher rates of palmar–plantar erythrodysesthesia, stomatitis, and mucositis compared with doxorubicin.²¹³ Capecitabine, a prodrug of fluorouracil, is administered orally, has the potential for blood–brain barrier penetration, and is not associated with alopecia or neuropathy. The benefit of capecitabine for patients with MBC has been demonstrated in multiple phase 2 trials, with ORRs of 28%–30%.²¹⁴ Eribulin is a nontaxane inhibitor of microtubule polymerization used for the treatment of patients with MBC who have previously received at least two prior lines of chemotherapy. In the EMBRACE phase 3 trial (ClinicalTrials.gov identifier NCT00388726), eribulin demonstrated improved PFS and OS compared with TPC in heavily pretreated patients. The median OS was longer in patients assigned to the eribulin arm compared with those receiving other chemotherapy agents (13.1 vs. 10.6 months; hazard ratio, 0.81; p = .041).²¹⁵ In another phase 3 trial, eribulin was compared with capecitabine in patients who had progressed on prior chemotherapy, and there were no differences in PFS (4.1 vs. 4.2 months) or OS (15.9 vs. 14.5 months).²¹⁶ Gemcitabine and vinorelbine can also be used as single-agent chemotherapy in heavily pretreated patients with MBC.^{217, 218}

Combination chemotherapy regimens are useful in patients with HR-positive/HER2-negative MBC who have rapid clinical progression or the need for rapid symptom and/or disease control. The NCCN guidelines recommend the following combination chemotherapy regimens as options: AC; epirubicin and cyclophosphamide; docetaxel and capecitabine; gemcitabine and paclitaxel; gemcitabine and carboplatin; cyclophosphamide, methotrexate, and fluorouracil; and carboplatin with paclitaxel or nanoparticle albumin-bound paclitaxel.⁹⁷

Antibody-drug conjugates

Another treatment strategy for HR-positive/HER2-negative MBC is the use of ADCs, which are a rapidly evolving area of HR-positive/HER2-negative breast cancer research and clinical practice. ADCs are composed of a monoclonal antibody linked to a cytotoxic small-molecule payload, engineered to deliver the small molecule preferentially to malignant cells, combining the specificity of the small molecule with the potency of the cytotoxic agent.²¹⁹

Trastuzumab deruxtecan (T-DXd; formerly DS-8201A) is a humanized antibody against HER2 linked to a topoisomerase I inhibitor, extecan derivative (DX-8951 derivative; DXd). Whereas HER2-positive breast cancer comprises only approximately 20% of new breast cancers, a greater proportion of patients (approximately 40%–50%) have breast cancer categorized as HER2-low, defined as having a HER2 IHC score of 1+ or 2+ but in situ hybridization-negative.²²⁰ The HER2-low group is a heterogenous population that includes luminal-type HR-positive and TNBC. The phase 3 DESTINY-Breast01 study (ClinicalTrials.gov identifier NCT03248492) enrolled 184 patients with HER2-positive unresectable breast cancer and/or MBC who had received previous treatment with trastuzumab emtansine and demonstrated an ORR of 60.3%.²²¹ It was hypothesized that this agent would also be effective in HER2-low disease. To address this question, the phase 3 DESTINY-Breast04 study (ClinicalTrials.gov identifier NCT03734019) evaluated the safety and efficacy of trastuzumab deruxtecan versus TPC (capecitabine, eribulin, gemcitabine, paclitaxel, nab-paclitaxel) in patients with HER2-low disease who had progressed on ET (HR-positive patients) received one or two prior lines of chemotherapy in the metastatic setting; 89% of patients were HR-positive, and 11% of were HR-negative in this study.¹⁵⁸ In the intention-to-treat (ITT) group, there was a remarkable improvement in both PFS (9.9 vs. 5.1 months; hazard ratio, 0.50; p < .0001) and OS (23.4 vs. 16.8 months; hazard ratio, 0.64; p = .0010). In the HR-positive cohort, there was also improvement in both PFS (10.1 vs. 5.4 months; hazard ratio, 0.51; p < .0001) and OS (23.9 vs. 17.5 months; hazard ratio, 0.64; p = .0028). The most common grade ≥3 AEs were neutropenia, thrombocytopenia, leukopenia, anemia, hypokalemia, elevated aspartate aminotransferase level, decreased appetite, and diarrhea. Drug-related interstitial lung disease (ILD) occurred in 12.1% of patients on T-DXd, including three drug-related ILD deaths; further work is needed to better predict which patients are at highest risk for ILD, and it is critical to monitor for pulmonary findings with regular computed tomography scans (every 9–12 weeks), stopping treatment, and starting steroids if indicated. Therefore, this study establishes T-DXd as a new standard of care for HER2-low MBC given the remarkable and statistically significant improvements in PFS and OS. Based on this study, T-DXd was approved by the FDA in August 2022 for the treatment of patients with unresectable or metastatic, HER2-low breast cancer.

Sacituzumab govitecan-hziy is another ADC that combines a humanized monoclonal antibody to the human trophoblast cell-surface antigen 2 (Trop-2) conjugated by a cleavable linker to SN-38, the active metabolite of irinotecan. SN-38 binds to the topoisomerase1 cleavage complex on DNA, ultimately causing S-phase–specific cell death. In patients with metastatic TNBC, the phase 3 ASCENT trial (ClinicalTrials.gov identifier NCT02574455) demonstrated that sacituzumab govitecan improved median PFS (5.6 vs. 1.7 months; p < .0001) and OS (12.1 vs. 6.7 months; p < .0001) compared with standard chemotherapy.²²² More recently, the phase 3 TROPiCS-02 study (ClinicalTrials.gov identifier NCT03901339) evaluated the safety and efficacy of sacituzumab govitecan versus TPC chemotherapy in patients with locally advanced, inoperable or metastatic HR-positive/HER2-negative MBC after progression on ET and two to four prior lines of chemotherapy for metastatic disease. In data initially presented at the 2022 ASCO meeting and updated at the 2022 ESMO and SABCS meetings, there was a statistically significant increase in PFS (5.5 vs. 4.0 months; hazard ratio, 0.66; p = .0003) and OS (14.4 vs. 11.2 months; hazard ratio, 0.79; p = .020) in this heavily pretreated population (median, three lines of prior chemotherapy; 95% had visceral disease).^{157, 223, 224} According to data presented at the SABCS, efficacy is independent of TROP2 expression levels.²²⁴ Overall, 74% of patients in the sacituzumab govitecan arm versus 60% of patients in the TPC arm experienced grade ≥3 treatment-emergent AEs, the most common of which were neutropenia (51% vs. 38%, respectively) and diarrhea (9% vs. 1%, respectively). Another trial, ASCENT-07, is planned to evaluate the safety and efficacy of sacituzumab govitecan versus TPC chemotherapy in patients with HR-positive/HER2-negative MBC who are eligible for first-line chemotherapy.

There are also other ADCs under evaluation in early phase clinical trials that have exciting potential. For example, the phase 1 TROPION-PanTumor-01 study (ClinicalTrials.gov identifier NCT03401385) evaluated the safety and efficacy of datopotamab deruxtecan in patients with HR-positive/HER2-negative MBC who had progression after one or two lines of chemotherapy and who had progressed on or were not suitable for ET. In preliminary data from 41 patients presented at the 2022 SABCS, the ORR was 27% (all partial responses), the CBR was 44%, and the median PFS was 8.3 months.²²⁵ The most common AEs of any grade were stomatitis (83%; 10% grade 3–4), nausea (56%; 0% grade 3–4), and alopecia (37%); Two patients experienced pneumonitis (one grade 2, one grade 3).²²⁵ Future work is needed to determine the role and sequencing of these ADCs for patients with HR-positive/HER2-negative MBC.

Immunotherapy

Immune checkpoint inhibitors (ICIs) have revolutionized cancer treatments for many cancer types. In breast cancer, ICIs have been most effective in TNBC to date. In HR-positive/HER2-negative MBC, the first trials evaluating ICI as monotherapy demonstrated modest RRs. In the phase 1b trial KEYNOTE-028 (ClinicalTrials.gov identifier NCT02054806), pembrolizumab monotherapy was evaluated in heavily pretreated patients with HR-positive/HER2-negative MBC who had a tumor combined positive score ≥1. In this study, the ORR was 12%, but the median duration of response was 12 months.²²⁶ In the phase 1 JAVELIN Solid Tumor trial (ClinicalTrials.gov identifier NCT01772004), 168 patients with pretreated MBC of all subtypes (regardless of PD-L1 status) received avelumab monotherapy, including 72 patients (42.9%) with HR-positive/HER2-negative MBC. The ORR for the entire cohort was only 3.0% (five patients), of whom three had TNBC and two had HR-positive/HER2-negative disease.²²⁷ More recently, a phase 2 trial evaluated the addition of pembrolizumab to eribulin in HR-positive/HER2-negative MBC. The addition of pembrolizumab did not improve PFS, ORR, or OS compared with eribulin alone in both the ITT and the PD-L1–positive populations.²²⁸ In another single-arm phase 2 trial, 44 patients with HR-positive/HER2-negative MBC were treated with pembrolizumab plus eribulin. In this study, an objective response was observed in 18 patients (40.9%), with a median PFS of 6 months and a 1-year OS of 59.1%.²²⁹

Given the low RR of ICI monotherapy in HR-positive MBC, more recent studies have evaluated the combination of ICI with other treatment modalities, including PARP inhibitors, CDK4/6 inhibitors, and radiation therapy (RT). PARP inhibitors have antitumor activity through activation of the stimulator of interferon genes (STING) pathway, which has been shown to increase PD-L1 expression in cancer cell lines.²³⁰ In the phase 2 MEDIOLA trial (ClinicalTrials.gov identifier NCT02734004), the activity of durvalumab in combination with olaparib was evaluated in 34 patients with MBC who had germline BRCA1 and BRCA2 mutations, including 13 patients with HR-positive disease. The disease control rate was 85% with a median PFS of 8.2 months. The median OS was 22.4 months in patients with HR-positive disease.²³¹ Patients who had received zero or one prior lines of chemotherapy had better outcomes compared with patients who had received two prior lines of chemotherapy. The combination of ICI and CDK4/6 inhibitors has also been evaluated. CDK4/6 inhibitors may enhance immunogenicity within the tumor microenvironment by increasing CD8-positive T-cell infiltration, decreasing regulatory T-cell infiltration and proliferation and PD-L1 expression.²³² A multicohort phase 1b study evaluated the safety and efficacy of pembrolizumab and abemaciclib in patients with HR-positive/HER2-negative MBC who had progressed on ET. Early data from 28 patients in the pembrolizumab and abemaciclib arm demonstrated an ORR of 29%, with a partial response in eight patients. The median PFS and OS were 8.9 and 26.3 months, respectively.²³³ Finally, ICI plus RT has also been studied because RT can stimulate a systemic, immune-mediated, antitumor response, known as the abscopal effect. In a small phase 2 trial, pembrolizumab plus palliative RT was assessed in eight heavily pretreated patients who had HR-positive MBC, and no objective responses were seen.²³⁴ Further studies are needed to better understand the role of local RT in combination with ICI in patients with MBC.

To date, there are currently no FDA-approved ICI agents for the treatment of HR-positive/HER2-negative MBC. The lack of activity of ICI in HR-positive disease could be because most studies have evaluated ICI in heavily pretreated patients and/or that there is a relatively poor immune response in this subtype. Ongoing trials are assessing the combination of ICI with other agents, and there are also efforts to identify better biomarkers to predict ICI response.

Factors to consider in sequencing therapy for HR-positive/HER2-negative MBC

The current paradigm for sequencing treatment in HR-positive/HER2-negative MBC is shown in Figure 3. ET plus CDK4/6 inhibitors are the current preferred therapy in first-line, HR-positive/HER2-negative MBC, unless there is evidence of rapid disease progression or visceral crisis and chemotherapy is preferred. Generally, it is recommended that ET options are used as second-line and third-line options as well, because ETs are generally better tolerated than chemotherapy options. The combination of alpelisib and fulvestrant is a preferred second-line treatment option for patients with PIK3CA mutations after progression on ET. In taxane-pretreated and anthracycline-pretreated patients who have germline BRCA1/2 mutations, olaparib and talazoparib are preferred treatment options after progression on CDK4/6 inhibitors. After exhausting ET and targeted therapy options, single agent chemotherapy is pursued, followed by ADCs, and then other single-agent and combination chemotherapy options. Patient preference, response to prior therapy, tumor biology and mutations, and toxicity profiles should all be considered to help determine the ideal sequence of therapies for each individual patient.

CONCLUSION

In summary, HR-positive/HER2-negative breast cancer is the most common subtype of breast cancer. Despite the efficacy of current treatment strategies, treatment resistance is still a challenge, and metastatic HR-positive/HER2-negative breast cancer is not curable. One promising treatment strategy is the use of targeted therapies in combination with ET, which may increase the efficacy and delay endocrine resistance. More recently, ADCs have shown promising efficacy in heavily pretreated, HR-positive/HER2-negative MBC. Further studies are needed to clarify which treatment strategies are most effective in the metastatic setting, the most effective sequencing of these therapies, mechanisms of treatment resistance, and whether combination therapies may increase efficacy to improve outcomes for patients with HR-positive/HER2-negative MBC.

CONFLICT OF INTEREST STATEMENT

Hope S. Rugo reports grants/contracts from Astellas Pharma, AstraZeneca, Boehringer Ingelheim, Daiichi-Sankyo Company, Eli Lilly and Company, Genentech USA Inc., Gilead Sciences Inc., GlaxoSmithKline, Hoffman-LaRoche, Macrogenetics, Novartis, OBI Pharma, Pfizer Inc., Pionyr Immunotherapeutics, Seattle Genetics, Sermonix, Taiho Oncology Inc., and Vertu Inc.; personal fees from Blueprint Medicines Corporation, NAPO Pharmaceuticals Inc., Puma, and Scorpion; and travel support from Astellas Pharma, AstraZeneca, Gilead Sciences Inc., and Merck outside the submitted work. Laura A. Huppert, Ozge Gumusay, and Dame Idossa disclosed no conflicts of interest.