Enhancing Endocrine Therapy Combination Strategies for the Treatment of Postmenopausal HR+/HER2– Advanced Breast Cancer

Evidence supports use of first‐line dual endocrine therapy (ET) regimens, particularly in ET‐naïve patients, or palbociclib plus letrozole, as well as everolimus plus exemestane or palbociclib plus fulvestrant as second‐line therapy for HR‐positive, HER2‐negative advanced breast cancer. Some combinations are associated with increased risk of class‐specific toxicities.

Abstract

Breast cancer (BC) is the most common malignancy in women worldwide, with approximately two‐thirds having hormone receptor‐positive (HR+) tumors. New endocrine therapy (ET) strategies include combining ET agents as well as adding inhibitors targeting growth factors, angiogenesis, the mechanistic target of rapamycin, phosphoinositide 3‐kinase (PI3K), or cyclin‐dependent kinase 4/6 to ET. Level 1 evidence supports use of fulvestrant plus anastrozole or palbociclib plus letrozole as first‐line therapy for HR+/HER− advanced BC with special consideration for the former in ET‐naïve patients, as well as everolimus plus exemestane or palbociclib plus fulvestrant as second‐line therapy with special consideration in select first‐line patients. Although the safety profiles of these combinations are generally predictable and manageable, both everolimus and palbociclib are associated with an increased risk of potentially serious or early‐onset toxicities requiring individualized a priori adverse event risk stratification, earlier and more rigorous agent‐specific monitoring, and patient education. Although each of these combinations improves progression‐free survival, none with the exception of anastrazole plus fulvestrant have demonstrated improved overall survival. PI3K catalytic‐α mutations assessed from circulating tumor DNA represent the first potentially viable serum biomarker for the selection of ET combinations, and new data demonstrate the feasibility of this minimally invasive technique as an alternative to traditional tissue analysis. Therapeutic ratios of select ET combinations support their use in first‐ and second‐line settings, but optimal sequencing has yet to be determined.

Implications for Practice.

Emerging data show that new endocrine therapy (ET) combinations can improve progression‐free and overall survival outcomes in patients with hormone receptor‐positive, HER2‐negative (HR+/HER−) advanced breast cancer. Level 1 evidence supports consideration of dual ET regimens, particularly in ET‐naïve patients, or palbociclib plus letrozole as first‐line therapy, as well as the addition of mTOR or CDK4/6 inhibitors to established ET in the second‐line setting and in select first‐line patients. Some combinations are associated with increased risk of class‐specific toxicities that will require individualized risk stratification, earlier and more rigorous agent‐specific monitoring, and patient education. Recent data on a noninvasive biomarker assay that predicts response to a phosphoinositide 3‐kinase inhibitor demonstrates the feasibility of this minimally invasive technique as an alternative to traditional tissue analysis.

Introduction

Breast cancer (BC) accounts for one in four new cancer diagnoses among women, resulting in an estimated 521,900 cancer deaths worldwide in 2012, second only to lung cancer in female cancer mortality in developed countries [1], [2]. Approximately two‐thirds (70%–75%) have hormone receptor (HR)‐positive disease (estrogen receptor [ER]‐positive, progesterone receptor [PR]‐positive, or both) [3], [4]. Twenty to thirty percent of patients with early BC will experience disease recurrence [5], and as many as 15%–25% may present with advanced disease, even in developed countries [1]. Luminal A tumors (ER‐ or PR‐positive, HER2‐negative, Ki‐67 < 14%) have the lowest rate of relapse in comparison with HER2‐enriched and basal subtypes [6].

Therapeutic strategies targeting estrogen modulation have evolved during the last few decades. Predictive biomarkers such as ER and human epidermal growth factor receptors (EGFR/HER2) have allowed for individualized treatment decisions and optimization of clinical benefit within defined BC subsets. Endocrine therapies (ET) that lower estrogen levels or inhibit the stimulatory activity of estrogen on ER‐positive BC cells are often preferred for postmenopausal women with advanced HR‐positive/HER2‐negative BC because of their highly favorable therapeutic index [3], [7]. Tamoxifen—as well as the aromatase inhibitors (AIs) anastrozole, letrozole, or exemestane—is used in both adjuvant and advanced settings [8], with recent evidence also supporting consideration of the selective estrogen receptor downregulator fulvestrant [9]. Despite the success of these approaches, there is an ongoing need for new strategies to delay acquired resistance and improve survival while maintaining the favorable therapeutic index of ET.

Combination strategies to improve the therapeutic ratio of ET in advanced disease, either through synergistic activity or by overcoming ET resistance, are areas of active research. Resistance to ET can arise via a number of mechanisms—including upregulated signaling through alternate pathways such as EGFR/HER2, phosphoinositide 3‐kinase (PI3K), protein kinase B, mammalian target of rapamycin (mTOR), and cyclin‐dependent kinase 4/6 pathways [10], [11]—which increase cancer cell proliferation, growth, cell cycle progression, and survival (Fig. 1).

Figure 1.

Signal pathways targeted to improve response to endocrine therapy (ET) or to reverse ET resistance. A number of biological pathways have been implicated in the development of ET resistance, including neoangiogenesis through the overexpression of vascular endothelial growth factor (VEGF) and VEGF receptor, in addition to upregulated signaling through epidermal growth factor receptor/ HER2, phosphoinositide 3‐kinase, protein kinase B, mammalian target of rapamycin, and cyclin‐dependent kinase 4/6 pathways, which can also increase cancer cell proliferation, growth, and survival or promote cell cycle progression. Inhibitors of these pathways are actively being developed in combination with established ET to improve response or reverse resistance to ET agents.

Abbreviations: Akt, protein kinase B; CDK4/6, cyclin‐dependent kinase 4/6; E2F, eukaryotic transcription factor; EGF(R), epidermal growth factor (receptor); ER, estrogen receptor; HER‐1/2, human epidermal growth factor receptor‐1/2; mTOR(C1/2), mechanistic target of rapamycin (complex 1/2); P, phosphate; PI3K, phosphoinositide 3‐kinase; PIP2/3, phosphatidylinositol (3,4,5)‐bi/trisphosphate; RAS, rat sarcoma viral oncogene homolog; Rb, retinoblastoma protein; Rheb, ras homolog‐enriched in brain; SRC, proto‐oncogene tyrosine‐protein kinase; STAT3, signal transducer and activator of transcription 3; TSC1/2, tuberous sclerosis complex 1/2; VEGF(R), vascular endothelial growth factor (receptor).

This article reviews phase III clinical trials evaluating ET combination therapy in postmenopausal patients with HR‐positive HER2‐negative advanced BC (ABC), with the purpose of providing practical clinical guidance.

Methods

Recent studies investigating ET for HR‐positive locally advanced or metastatic BC were identified and reviewed. We searched PubMed (January 1, 2010 to December 15, 2015), the Proceedings of American Society of Clinical Oncology (2014–2015), the Annual Congress of the European Society for Medical Oncology (2014–2015), and the San Antonio Breast Cancer Symposium (2014–2015) meetings using the key search terms “hormone receptor‐positive” AND “endocrine therapy” AND “breast” AND “advanced” AND “RCTs” (OR respective aliases). Findings were supplemented with bibliographic and directed searches as needed. Records were vetted at abstract level and confirmed at full text. Primary English language study reports of phase III trials investigating ET combinations as induction in postmenopausal HR‐positive, HER2‐negative ABC with outcomes were identified (Fig. 2). Studies with mixed populations based on HER2 or menopausal status were included if subgroup analyses describing outcomes for the population of interest were available, or at least 75% of the population represented the population of interest. Eligible trials were classified by line of treatment, with those having a majority (>75%) of patients receiving no prior ET for ABC‐designated first‐line and all others given a second‐line (or beyond) designation. Adverse events of interest (differential AEs) for each combination were identified by ranking total any grade AEs that were >10% more frequent for the combination in comparison with controls.

Figure 2.

PRISMA diagram. a, JCO database; b, European Cancer Congress 18/European Society of Medical Oncology (ESMO) 2015: European Journal of Cancer database, ESMO 2014: Annals of Oncology database; c, San Antonio Breast Cancer Symposium (SABCS) 2014: Cancer Research (journal) database, SABCS 2015: SABCS database; d, Or respective aliases; e, Primary reports of eligible studies that were not identified through database search; primary reports were defined as the most detailed and current report of the primary endpoint analysis; f, Trials in endocrine therapies pretreated populations may include both first‐line and second‐line or beyond patients.

Abbreviations: ASCO, American Society of Clinical Oncology; BC, breast cancer; CT, clinical trial; ECCO, European Cancer Congress; ESMO, European Society of Medical Oncology; ET, endocrine therapy; HR, hormone receptor; RCT, randomized controlled trial; SABCS, San Antonio Breast Cancer Symposium.

Results

The literature search produced 334 such records, representing 12 reports (Fig. 2).

ET Combinations First‐Line

A total of seven trials investigating dual ET combinations first‐line were identified (Table 1).

Table 1. Phase III efficacy outcomes of endocrine therapy combination strategies for first‐linea treatment of postmenopausal, hormone receptor‐positive, HER2‐negative advanced breast cancer.

^a”First‐line” defined as the majority of patients receiving endocrine therapy as their first treatment for advanced disease.

^bAlso premenopausal women receiving a gonadotropin‐releasing hormone agonist (2.9%).

^cIndicates primary endpoint.

^dTime to progression, time from randomization until objective progression or death owing to any cause in the absence of progression.

^eProtocol amendment, prior adjuvant aromatase inhibitor (AI) allowed ( > 12 months prior), but essentially none had received an Al as adjuvant therapy.

^fTwenty percent of patients had received an AI as their most recent therapy prior to enrollment.

^gBased on local testing of the most recently analyzed tissue.

^hPopulation assessed for efficacy.

ⁱLetrozole plus temsirolimus compared with letrozole plus placebo on the basis of Cox proportional hazards model stratified by prior bone disease status and geographic region.

^jMedian follow‐up for progression‐free survival.

^kObjective response from 197 patients with measurable disease and evaluable for response.

Abbreviations: (A)BC, (advanced) breast cancer; Al, aromatase inhibitor; CI, confidence interval; ER, estrogen receptor; ET, endocrine therapy; HER2, human epidermal growth factor receptor‐2; HR, hazard ratio; HR+, hormone receptor‐positive; MBC, metastatic breast cancer; n, number; NE, not estimable; neo(adj), neo(adjuvant); NR, not reported; ORR, overall response rate; OS, overall survival; PFS, progression‐free survival; q2w, every 2 weeks; q3w, every 3 weeks; q4w, every 4 weeks; VEGF, vascular endothelial growth factor.

ET Combinations With an Epidermal Growth Factor Receptor Inhibitor.

The randomized, double‐blind, placebo‐controlled, EGF30008 study (n = 1,286) evaluated the addition of an orally active dual EGFR/HER2 tyrosine kinase inhibitor lapatinib (1,500 mg daily) to letrozole in patients with predominantly HER2‐negative (83% HER2‐negative or unknown) ABC [12]. At a median follow‐up of 24 months, small but significant improvements in median progression‐free survival (PFS) were observed with the addition of lapatinib in the intent‐to‐treat (ITT) population (11.9 vs. 10.8 months; hazard ratio [HR]: 0.86; 95% confidence interval [CI]: 0.76, 0.98; p = .026), but not in the 952 centrally confirmed HER2‐negative BC patients (HR: 0.90; 95% CI: 0.77, 1.05; p = .188) (Table 1). Serious AEs occurred in 8% of patients receiving the combination versus 4% on letrozole plus placebo.

Dual ET Combinations.

Two trials investigated the addition of 250 mg of fulvestrant to anastrozole. The open‐label, randomized FACT trial (n = 514) enrolled mostly pretreated patients (67.7% had received prior adjuvant ET) [13]. No significant improvement in the primary endpoint, time to progression (median TTP: 10.8 vs. 10.2 months; HR: 0.99; 95% CI: 0.81, 1.20; p = .91), or overall survival (median OS: 37.8 vs. 38.2 months; HR: 1.00; 95% CI: 0.76, 1.32; p = 1.00) was observed (Table 1) [13]. Treatment discontinuations due to AEs were more common in the combination arm (6.3% vs. 3.1%; Table 2). The Southwest Oncology Group [SWOG] S0226 study (n = 694) was an open‐label, randomized trial evaluating the same ET combination, with approximately 60% of patients ET naïve [14]. The combination significantly improved median PFS (15.0 vs. 13.5 months; HR: 0.80; 95% CI: 0.68, 0.94; p = .007) and median OS (47.7 vs. 41.3 months; HR: 0.81; 95% CI: 0.65, 1.00; p = .049) in the ITT population (Table 1), and median PFS (17.0 vs. 12.6 months; HR: 0.74; 95% CI: 0.59, 0.92; p = .006) and OS (HR: 0.74; 95% CI: 0.56, 0.98; p = .04) in the 414 patients who had not received prior adjuvant tamoxifen. Treatment discontinuations were more frequent with the combination (3.2% vs. 1.2%; Table 2), although deaths possibly related to treatment were low in both arms (0.9% vs. 0). The most common grade 3/4 differential AEs with the combination were flu‐like symptoms (2.9% vs. 1.8%), endocrine (2.0% vs. 0.3%), and dermatologic/skin (0.3% vs. 0.3%) [14] (Table 2).

Table 2. Select phase III safety outcomes of endocrine therapy combination strategies for the treatment of postmenopausal hormone receptor‐positive, HER2‐negative advanced breast cancer.

^aDifferential adverse events (AEs) were calculated by noting and ranking any grade combination AEs that were >10% more frequent than single‐agent AEs.

^bSignificant improvements in progression‐free survival (PFS) and overall survival for the experimental arm.

^cSignificant improvements in PFS for the experimental arm.

^dOnly hematologic AEs reported.

^eIncludes clustered Medical Dictionary for Regulatory Activity (MedDRA) preferred terms.

^fFrom overall survival update [22].

Abbreviations: AE, adverse event; ALT, alanine aminotransferase; AST, aspartate aminotransferase; Gr, grade; n, number; NR, not reported; q2w, every 2 weeks; q4w, every 4 weeks.

ET Combinations With a Cyclin‐Dependent Kinase 4/6 Inhibitor.

Addition of the cyclin‐dependent kinase 4/6 (CDK4/6) inhibitor palbociclib (125 mg daily for 3 weeks, followed by 1 week off) to first‐line letrozole was initially assessed in the phase II PALOMA‐1 study [15]. This regimen was also evaluated in the randomized, double‐blind, placebo‐controlled phase III PALOMA‐2 trial in patients not deemed AI resistant (n = 666), the majority of whom (56%) had prior (neo)adjuvant ET [16]. A preliminary analysis showed a net 10.3‐month median investigator‐assessed PFS gain with palbociclib (24.8 vs. 14.5 months; HR 0.58; 95% CI: 0.46, 0.72; p < .000001; Table 1), with significant benefit across prespecified subgroups. Discontinuations due to AEs occurred in 9.7% of patients receiving palbociclib in comparison with 5.9% in the placebo arm (Table 2). Deaths due to AEs were reported in 2.3% and 1.8% of patients in the palbociclib and placebo arms, respectively. Reported grade 3/4 differential AEs with the palbociclib combination were hematological, including neutropenia (grade 3/4, 56/10% vs. 1/<1%), leukopenia (grade 3/4, 24/1% vs. 0/0), anemia (grade 3/4, 5/<1% vs. 2/0), and thrombocytopenia (grade 3/4, 1/<1% vs. 0/0) (Table 2), although rates of febrile neutropenia were low in both arms (1.6% vs. 0). OS analysis has yet to be reported.

ET Combinations With an mTOR‐PI3K Inhibitor.

The randomized, double‐blind, placebo‐controlled HORIZON trial (n = 1,112) evaluated the addition of the mTOR inhibitor temsirolimus 30 mg on an intermittent schedule (daily for 5 days every 2 weeks) to first‐line letrozole [17]. No significant improvement in investigator‐assessed median PFS was observed with temsirolimus in the overall population (8.9 vs. 9.0 months; HR 0.90; 95% CI: 0.76, 1.07; p = .25; Table 1). More patients receiving temsirolimus had a permanent dose reduction because of AEs (4% vs. 1%).

ET Combinations With Antiangiogenic Agents.

Two trials evaluated the addition of 15 mg/kg of bevacizumab every 3 weeks, a monoclonal antibody against vascular endothelial growth factor‐A, to ET. The open‐label, randomized, LEA study (n = 374) reported no significant difference in median PFS (19.3 vs. 14.4 months; HR 0.83; 95% CI: 0.65, 1.06; p = .126) or OS (52.1 vs. 51.8 months; HR 0.87; 95% CI: 0.58, 1.32; p = .518) with the addition of bevacizumab to ET (letrozole or fulvestrant, 250 mg, every 4 weeks; Table 1) [18]. The randomized CALGB 40503 trial (n = 343), however, observed a significant improvement in median PFS with bevacizumab plus letrozole in comparison with letrozole (20.2 vs. 15.6 months; HR 0.75; 95% CI: 0.59, 0.96; p = .016), although a significant improvement in median OS was not observed (47.2 vs. 43.9 months; HR 0.87; 95% CI: 0.65, 1.18; p = .188; Table 1) [19]. Discontinuations due to AEs were much more common with combination therapy than with ET in both trials (14% and 20.5% vs. 1% and 0%), as were deaths due to AEs in the LEA study (4.2% vs. 0).

ET Combinations Second‐Line and Beyond

A total of five combination trials were identified that evaluated either dual ET targeting or combinations of ET with mTOR‐PI3K inhibitors, growth factor inhibitors, or CDK4/6 inhibitors in the second‐line setting (or beyond; Table 2).

Dual ET Combinations.

SoFEA (n = 723), a multicenter, randomized, placebo‐controlled study evaluated in part the addition of anastrozole or placebo to fulvestrant 250 mg (500‐mg loading dose) in predominantly HER2‐negative or unknown (93.4%) patients [20]. The addition of fulvestrant to anastrozole did not improve median PFS (4.4 vs. 4.8 months; HR 1.00; 95% CI: 0.83, 1.21; p = .98) or median OS (20.2 vs. 19.4 months; HR 0.95; 95% CI: 0.76, 1.17; p = .61; Table 3). The lack of PFS benefit was consistent across subgroups. Discontinuations due to AEs were similar across all treatment arms (2.9%–3.6%). There were no deaths attributed to AEs or differential grade 3/4 AEs reported [20].

Table 3. Phase III efficacy outcomes of endocrine therapy combination strategies for second‐linea treatment of postmenopausal, hormone receptor‐positive, HER2‐negative advanced breast cancer.

^a“Second‐line” defined as the majority of patients receiving endocrine therapy following another line of therapy in the advanced disease setting.

^bLoading dose 500 mg on day 1, followed by 250‐mg injections on days 15 and 29. Thereafter, 250‐mg intramuscular injections were done every 28 days.

^clnterquartile range.

^dIndicates primary endpoint.

^eHazard ratio for fulvestrant plus anastrozole versus fulvestrant plus placebo.

^fHazard ratio for fulvestrant plus placebo versus exemestane.

^gAII patients had HER2‐negative tumors (by protein or gene analysis), except two patients for whom the result was missing.

^hMedian follow‐up for primary endpoint,

ⁱOS data immature at time of PFS analysis, with a trend in favor of the experimental arm,

^jOriginally, HER2 status eligibility was as follows: HER2 1+, 2+, or 3 + by immunohistochemistry; fluorescent in situ hybridization positive; or serum HER2 extracellular domain > 15 ng. The protocol was subsequently amended to include tumors regardless of HER2 status.

Abbreviations: (A)BC, (advanced) breast cancer; adj, adjuvant; adv, advanced; AI, aromatase inhibitor; CI, confidence interval; ER, estrogen receptor; ET, endocrine therapy; HER2, human epidermal growth factor receptor‐2; HR, hazard ratio; HR+, hormone receptor‐positive; MBC, metastatic breast cancer; mt, mutated; n, number; NR, not reported; (NS)AI, (nonsteroidal) aromatase inhibitor; ORR, overall response rate; OS, overall survival; PFS, progression‐free survival; PIK3CA, phosphatidylinositol‐4,5‐bisphosphate 3‐kinase, catalytic subunit α; q2w, every 2 weeks; q4w, every 4 weeks.

ET Combinations With an mTOR‐PI3K Inhibitor.

Two trials evaluated mTOR/PI3K pathway inhibitors in combination with ET in predominantly second‐line and beyond populations. The double‐blind, placebo‐controlled BOLERO‐2 study (n = 724) assessed the mTOR inhibitor everolimus, 10 mg, or placebo plus exemestane [21]. Patients enrolled in this trial had progressed on a prior AI and were heavily pretreated (81% had received prior therapy for advanced disease, including chemotherapy [26%], and more than half had received ≥3 prior therapies). The majority had visceral disease (56%) [21], [22], [23]. At a median follow‐up of 18 months, everolimus plus exemestane resulted in a statistically significant 4.6‐month improvement in median PFS (7.8 vs. 3.2 months; HR 0.45; 95% CI: 0.38, 0.54; p < .0001; Table 3) [23]. PFS benefits were comparable across all subgroups [21], [23], [24], [25], including the 19% of the study population receiving the combination as first‐line therapy (11.5 vs. 4.1 months; HR 0.39; 95% CI: 0.25, 0.62) [26]. A nonsignificant 4.4‐month net gain in median OS was also observed in the overall population (31.0 vs. 26.6 months; HR 0.89; 95% CI: 0.73, 1.10; p = .14) [22]. The everolimus combination did not adversely affect patient‐reported health‐related quality of life (HRQoL) [27], and improvements in median time to definitive deterioration (TDD) in HRQoL were reported, favoring the addition of everolimus (8.3 vs. 5.8 months; HR 0.74; 95% CI: 0.58, 0.95; p = .0084) [28]. Treatment discontinuation due to AEs was higher in patients receiving the combination than in those receiving placebo at a median follow‐up of 18 and 39.3 months (26.3% and 29% vs. 5% for placebo at both follow‐ups; Table 2) [22], [23]. Deaths on treatment at 39.3 months occurred in 4.6% of patients on everolimus in comparison with 1.7% on placebo, with similar rates of death due to AEs (1.7% vs. 0.4%) [22]. The most common grade 3/4 differential AEs associated with the everolimus combination at 18 months were stomatitis (8/0% vs. < 1/0%), anemia (7/<1% vs. < 1/<1%), dyspnea (5/<1 vs. < 1/<1), hyperglycemia (5/<1% vs. < 1/0%), and pneumonitis (3/0% vs. 0/0) (Table 2) [23].

The recently presented, randomized, placebo‐controlled BELLE‐2 trial (n = 1,147) evaluated fulvestrant, 500 mg, plus the pan PI3K inhibitor buparlisib, 100 mg daily, or placebo [29]. A modest but significant 1.9‐month net improvement in median PFS was seen in the ITT population with the addition of buparlisib in comparison with placebo (6.9 vs. 5.0 months; HR 0.78; 95% CI: 0.67, 0.89; p < .001; Table 3). OS outcomes were premature. Discontinuations due to AEs occurred more frequently in the buparlisib arm (13.2% vs. 1.8%; Table 2). The most common differential grade 3/4 AEs included transaminitis (alanine transaminase increase, 18.7/6.8% vs. 1.1/0% and aspartate aminotransferase increase, 15.0/3.0% vs. 2.8/0%), hyperglycemia (15.2/0.2% vs. 0.2/0%), rash (7.7/0.2% vs. 0/0), and mood disorders (anxiety, 5.2/ 0.2% vs. 0.9/0%, and depression, 3.7/0.7% vs. 0.4/0%) [29] (Table 2).

ET Combinations With a Growth Factor Inhibitor.

The randomized, double‐blind CALGB 40302 trial (n = 295) assessed 250 mg of fulvestrant plus 1,500 mg of lapatinib daily or placebo in predominantly HER2‐negative patients (81%) [30]. No significant improvements in outcomes were seen for the addition of lapatinib (median PFS, 4.7 vs. 3.8 months; HR 1.04; 95% CI: 0.82, 1.33; p = .37; median OS, 30.0 vs. 26.4 months; HR 0.91; 95% CI: 0.68, 1.21; p = .25; Table 2). Treatment discontinuations due to AEs were significantly more frequent in the lapatinib arm (12% vs. 2%; p = .001) [30].

ET Combinations With a CDK4/6 Inhibitor.

The randomized, double‐blind, placebo‐controlled PALOMA‐3 trial (n = 521) evaluated the CDK4/6 inhibitor palbociclib (125 mg per day orally for 3 weeks, followed by 1 week off) plus 500 mg of fulvestrant. Patients in this trial had progressed on ET therapy. The majority of enrolled patients were postmenopausal (79%), pretreated (86% of patients had received prior AI therapy), or had extensive disease (60% of patients had visceral involvement) [31]. At a median follow‐up of 8.9 months, a net 4.9‐month median PFS gain was observed with the addition of palbociclib (9.5 vs. 4.6 months; HR 0.46; 95% CI: 0.36, 0.59; p < .0001; Table 3), with comparable PFS benefits in pre/peri‐ and postmenopausal patients (HR for progression or death, 0.50 and 0.45, respectively), and a significant 4.1‐month net median PFS improvement for palbociclib plus ET in the 22% of the study population receiving first‐line therapy (9.5 vs. 5.4 months; HR 0.55; p = .02). Significantly improved estimated overall global QoL (66.1; 95% CI: 64.5, 67.7 vs. 63.0; 95% CI: 60.6, 65.3; p = .0313) and improvement from baseline in pain (−3.3; 95% CI: −5.1 to −1.5 vs. 2.0; 95% CI: −0.6 to 4.6; p = .0011) were observed for palbociclib plus fulvestrant, in addition to delayed deterioration in global QoL (p < .025) and pain (p < .001) in comparison with fulvestrant alone [32]. Discontinuations due to AEs (4% vs. 2%; Table 2) and deaths on treatment (1.4% vs. 0) were similar for the palbociclib and placebo arms, and no deaths due to AEs were reported [31]. The most common grade 3/4 differential AEs with the palbociclib combination were hematological, including neutropenia (64.6% vs. 0.6%) and leucopenia (27.5% vs. 1.2%). Rates of febrile neutropenia were low and comparable across arms (0.9% vs. 0.6%) [31]. Other common grade 3/4 differential AEs included any grade infection (41.7% vs. 30.2%) and fatigue (39.1% vs. 28.5%) (Table 2).

Discussion

ET Combinations First‐Line

Among the four ET combinations investigated as first‐line strategies (dual ET targeting, CDK4/6 inhibitor, mTOR inhibitor, and antiangiogenic combinations), fulvestrant plus anastrozole and palbociclib plus letrozole showed significant benefit. In the SWOG 0226 study, fulvestrant plus anastrozole resulted in a small but statistically significant 1.5‐month improvement in median PFS in comparison with anastrozole (HR 0.80; p = .007) and a marginally significant 6.4‐month median OS gain (HR 0.81; p = .049) [14]. Similar improvements, however, were not seen in the FACT and SoFEA studies [13], [20]. Observed differences in survival benefit may have been influenced by treatment bias related to the open‐label design of the SWOG study, the possible underpowering of FACT and SoFEA, or differences in postprogression treatment or bisphosphonate use. The SWOG study also included a large proportion of ET naïve patients (60% vs. 32% and 0 in the FACT and SoFEA trials, respectively) [13], [14], [20], and a post hoc subgroup analysis observed significant benefits in the combination subset of the ET‐naïve cohort (PFS, HR 0.74, p = .006; OS, HR 0.74, p = .04), although the interaction between treatment and prior tamoxifen use was not significant (p = .22) [14]. Given the improved survival and favorable safety profile, fulvestrant (500 mg on day 1 and 250 mg on days 14 and 28 and monthly thereafter) plus anastrozole should be considered for use as a first‐line therapy, with preferential consideration in ET‐naïve patients. Although fulvestrant injections are well tolerated [33], [34], some patients may not favor monthly injections [35].

Recent results from the PALOMA‐2 trial have shown a greater than 10‐month net PFS gain for the addition of palbociclib to letrozole [16], with significant benefit across all prespecified subgroups. The regimen was generally well tolerated in spite of high hematologic toxicity rates; the incidence of neutropenic fever was low. Early results from this study confirm benefits observed in the phase II PALOMA‐1 trial and represent the longest improvement in median PFS to date in this setting [16]. Further follow‐up will determine whether these benefits translate into improved OS. Given the substantial PFS benefits and manageable safety profile, palbociclib plus letrozole should be considered for use in the first‐line setting, although benefit in AI‐resistant patients is unclear.

ET Combinations Second‐Line and Beyond

From the five ET combinations evaluated as second‐line therapy or beyond [20], [21], [29], [30], [31], adding an mTOR or CDK4/6 inhibitor to ET significantly improved outcomes in ITT populations, although with increased risk of class‐specific toxicities [23], [31]. Both ET combinations resulted in a greater than twofold improvement in median PFS in comparison with single‐agent ET (everolimus, 3.2 to 7.8 months, HR 0.45, p < .0001; and palbociclib, 4.6 to 9.5 months, HR 0.46, p < .0001) [23], [31] and improved QoL (everolimus TDD in HRQoL, p = .0084 and palbociclib, overall global QoL, p = .0313, and TDD in global QoL, p < .025) [28], [32].

OS was not statistically significantly improved by the addition of everolimus to exemestane in BOLERO‐2 [22], and further follow‐up is required to assess OS in the PALOMA‐3 trial. Although significance was not reached, it is important to note that neither trial was powered to detect OS and that poststudy treatment leading to prolonged postprogression survival may confound the detection of OS benefits in this setting [22]. Approximately 20% of patients in BOLERO‐2 and PALOMA‐3 were treated with first‐line ET combinations. Subgroup analyses from both trials observed substantial median PFS benefit with combination therapy that mirrored ITT effects (palbociclib, 9.5 vs. 5.4 months, HR 0.55, p = .0214; and everolimus, 11.5 vs. 4.1 months, HR 0.39, 95% CI: 0.25, 0.62) [26], [31].

ET combinations involving everolimus and palbocilib are associated with distinct AE profiles and with an increased risk of severe (everolimus, pneumonitis) or early‐onset (everolimus; stomatitis and palbociclib; hematological) toxicities. AEs associated with palbociclib are primarily laboratory based, and although overall grade 3/4 toxicities were more frequent in patients on the palbociclib combination (73%) than those on everolimus (41%), palbociclib was associated with a much lower rate of treatment discontinuation due to AEs (4% vs. 26.3%). Differential grade 3/4 toxicities associated with palbociclib included neutropenia (65%), leukopenia (28%), and anemia (3%) [31]. Those associated with everolimus included stomatitis (8/0%), anemia (7/< 1%), hyperglycemia (5/< 1%), and pneumonitis (3/0%) (Table 2) [23], although a recent phase II study showed that a 0.5 mg per 5 mL dexamethasone mouthwash eliminated grade 3/4 stomatitis and substantially reduced grade 2 stomatitis (2.4% vs. 25%) in comparison with BOLERO‐2 [36]. The substantial improvements in ITT PFS, increased QoL, and manageable safety profiles observed with everolimus or palbociclib combinations support their role in second‐line therapy, although their use in ET‐naïve patients is unclear. There are also data to support their use first‐line although evidence in this setting is less robust.

Enhanced Multidisciplinary Management Strategies

Traditionally, single‐agent ET therapy has been associated with a highly favorable risk‐benefit profile, allowing for treatment with minimal need for close clinical supervision or toxicity monitoring. The anastrazole plus fulvestrant combination is well tolerated and appears to improve PFS and OS. Novel ET combination strategies can improve PFS but are associated with an increased risk of potentially serious and early‐onset class‐specific toxicities [31], [37], [38]. These AEs, however, are generally manageable through early detection, dose adjustments, and interruptions [37], [38], [39]. Off‐study clinical adoption of these novel combinations will require coordinated efforts between oncology clinicians, nurses, and pharmacists, and should be accompanied by individualized, a priori AE risk stratification based on comorbidities, anticipated tolerance, and available support systems. Additional strategies should include enhanced education and earlier, as well as more frequent, monitoring, including regimen‐specific call‐back schedules to facilitate the identification and resolution of treatment‐emergent toxicities. With proper support, ET combinations have the potential to safely extend the benefits of ET therapy while delaying the need for more aggressive interventions, such as chemotherapy.

Biomarkers and Patient Selection

Given the increased risk of toxicity and additional costs, the ability to identify patients who might preferentially benefit from ET combinations is of paramount interest. Many predictive biomarkers have been evaluated for ET combinations, including acquired ESR1 mutations and increased cyclin D1 or AP‐2γ for AIs or fulvestrant [40], [41], [42], [43], genetic alterations in phosphatidylinositol‐4,5‐bisphosphate 3‐kinase catalytic subunit α (PIK3CA), CCND1, or fibroblast growth factor receptors 1 and 2 for everolimus [44], and increased Rb expression, cyclin D1 amplification, or p16 loss for sensitivity to CDK4/6 inhibitors [45]. These putative biomarkers, however, have not yet demonstrated sufficient predictive ability or reproducibility to be clinically useful.

Tumors with constitutively active PIK3CA because of exons 9 and 20 mutations have proposed sensitivity to agents targeting this pathway [46]. An exploratory biomarker analysis from BOLERO‐2 suggested greater PFS benefit from everolimus in tumors with lower chromosomal instability or those with mutations in the exon 9 functional domain as opposed to the exon 20 functional domain of PIK3CA [47]. Although the statistically significant 1.9‐month net median PFS improvement (HR 0.78; p < .001) observed with the addition of buparlisib to ET in BELLE‐2 was modest and may be of limited clinical relevance, an exploratory subgroup analysis of patients with PIK3CA mutations in circulating tumor DNA (ctDNA) revealed a statistically significant 3.8‐month gain in median PFS (HR 0.56; p < .001) in comparison with those with detectable ctDNA who did not harbor a PIK3CA mutation [29]. PI3K catalytic α (PI3KCA) mutation status was also assessed in CDK4/6 inhibitors, although it did not significantly influence the treatment effect of palbociclib in PALOMA‐3 (interaction p = .83) [31]. Although PI3KCA mutations do not appear predictive for CDK4/6 inhibitors, they may be relevant for PI3K inhibitors by using a minimally invasive analytic technique that could serve as a convenient alternative to tissue biopsy [40], [48], [49], [50]. Prospective validation will inform the ultimate clinical utility and value of this biomarker and help define the role of ET combinations incorporating an mTOR or PIK3CA inhibitor for treating advanced disease.

ET Therapy Optimal Sequencing

There is a lack of evidence informing optimal sequencing of available therapies in the treatment of advanced HR‐positive/ HER2‐negative BC. Although there is interest in tailoring treatment to specific patient subsets on the basis of underpowered subgroup analyses in the pivotal clinical trials, caution must be applied when extrapolating these data to the real‐world clinical population. Individual patient characteristics such as age, preference, treatment history, and disease status—in addition to efficacy, safety, convenience, cost, and availability—will continue to influence treatment decisions, pending new evidence informing patient selection and optimal sequencing. There is a need for ongoing research in this area, including noninterventional, prospective, real‐world observational studies such as the German BRAWO [51] and the Canadian Treat ER+ight [52] studies. To inform standards of care, correlative research aimed at identifying biomarkers predictive of preferential benefit, optimal sequencing, or both is an important ongoing avenue of research.

There is a lack of evidence informing optimal sequencing of available therapies in the treatment of advanced HR‐positive/HER2‐negative BC. Although there is interest in tailoring treatment to specific patient subsets on the basis of underpowered subgroup analyses in the pivotal clinical trials, caution must be applied when extrapolating these data to the real‐world clinical population.

Conclusion

Level 1 evidence supports use of fulvestrant plus anastrazole as well as palbociclib plus letrozole as first‐line therapy for HR‐positive and HER2− advanced BC patients with special consideration for the former in ET‐naïve patients, as well as everolimus plus exemestane and palbociclib plus fulvestrant as second‐line therapy, with special consideration in select first‐line patients. Although the safety profiles of these combinations, particularly of fulvestrant plus anastrazole, are generally predictable and manageable, the palbociclib and everolimus combinations are associated with an increased risk of class‐specific toxicities requiring earlier and more regular monitoring, as well as enhanced patient education. Optimal sequencing for ET combination therapy has yet to be determined.

Author Contributions

Conception/Design: Kathleen I. Pritchard, Christine Simmons, Daniel Rayson, Louise Provencher, Alexander Paterson, Deanna McLeod, Stephen K. Chia

Collection and/or assembly of data: Kathleen I. Pritchard, Deanna McLeod Data analysis and interpretation: Kathleen I. Pritchard, Christine Simmons, Daniel Rayson, Louise Provencher, Alexander Paterson, Deanna McLeod, Stephen K. Chia

Manuscript writing: Kathleen I. Pritchard, Christine Simmons, Daniel Rayson, Louise Provencher, Alexander Paterson, Deanna McLeod, Stephen K.Chia

Final approval of manuscript: Kathleen I. Pritchard, Christine Simmons, Daniel Rayson, Louise Provencher, Alexander Paterson, Deanna McLeod, Stephen K. Chia

Disclosures

Kathleen I. Pritchard: AstraZeneca, Pfizer, Roche, Amgen, Novartis, GlaxoSmithKline, Eisai (C/A, H, Other); Christine Simmons: Eisai, Amgen, AstraZeneca (C/A), Amgen, Roche (RF); Daniel Rayson: Pfizer Canada, Novartis Canada, and Roche Canada (RF); Louise Provencher: Roche, Pfizer, Novartis, Amgen (C/A), Roche (RF); Alexander Paterson: Pfizer, Novartis, Roche, Roche Diagnostics, and Nanostring (C/A); Deanna McLeod: Novartis Pharmaceuticals Canada, Inc. (RF); Stephen K. Chia: Novartis, Pfizer (C/A). OI [spouse]); Yousuf Houck: Genentech (C/A).

(C/A) Consulting/advisory relationship; (RF) Research funding; (E) Employment; (ET) Expert testimony; (H) Honoraria received; (OI) Ownership interests; (IP) Intellectual property rights/ inventor/patent holder; (SAB) Scientific advisory board