Hormone receptor-positive, HER2-negative metastatic breast cancer: redrawing the lines.

Estrogen receptor modulators and estrogen deprivation have become standards of care for hormone receptor-positive metastatic breast cancer. However, after traditional first-line endocrine monotherapy treatment, the disease typically progresses despite the initial high rate of clinical benefit. Multiple studies have aimed at optimizing treatment strategies to improve upon clinical benefit beyond the traditional single-agent endocrine treatment. With the availability of new data and novel therapies, the clinical practice challenge becomes how best to define the optimal treatment sequence to maximize clinical benefit. In this review, we present treatment options clinically relevant to the management of hormone-positive, her2-negative metastatic breast cancer, and we propose a treatment algorithm based on the current literature.


INTRODUCTION
Survival rates for women with metastatic breast cancer (mbca) are improving, especially for those whose tumours express the estrogen or progesterone hormone receptors (hrs) or the human epidermal growth factor receptor 2 (her2) 1,2 . The most common subtype of breast cancer is hr-positive breast cancer, which accounts for approximately 60% -70% of all cases 3 . Oophorectomy was first shown to cause regression of unresectable breast cancer in 1896, and since then, estrogen receptor modulation and estrogen deprivation have become standards of care for hr-positive mbca 4,5 .
Unfortunately, despite the high rate of clinical benefit from initial endocrine treatment, disease progression typically occurs after 1 year of traditional first-line endocrine monotherapy 3 . Multiple studies have aimed at optimizing treatment strategies to further improve on clinical benefit beyond traditional single-agent endocrine therapy. However, with more recent positive data available, the clinical practice challenge has become how to define the optimal treatment sequence to maximize clinical benefit. In this review, we present treatment options clinically relevant to the management of hr-positive, her2-negative mbca, and we propose a treatment algorithm based on the current literature. now available and have also been incorporated into clinical practice 3 . Endocrine therapy might be suitable for patients with hr-positive, her2-negative mbca who have lowburden disease (that is, bone as a single site of metastatic disease, or nonthreatening visceral burden) and for those who have experienced a long disease-free interval (that is, beyond 2 years) to enrich for more endocrine-responsive disease 3,7 . Patients with rapidly progressive visceral disease or with a risk or evidence of end-organ dysfunction or significant disease-related symptoms should be offered chemotherapy 3,7 . The choice of endocrine agent should be based on menopausal status, comorbidities, prior adjuvant therapy, drug availability, patient preference, and drug safety profile 3,7 .

First-Line Treatment Options for HR-Positive, HER2-Negative mBCa
In the past, based on positive results in randomized trials and subsequent meta-analyses, single-agent endocrine therapy was the mainstay of first-line treatment for hr-positive, her2-negative mbca 3 . The two most effective endocrine monotherapy treatment choices in the first-line setting are either aromatase inhibitors (ais) or selective estrogen receptor degraders (fulvestrant). Those endocrine agents have also now been studied in combination with inhibitors of cyclin-dependent kinases 4 and 6 (cdk 4/6) 3 . Table i summarizes the most relevant data in the first-line setting for postmenopausal patients.

AIs:
The ais block the aromatase enzyme, which normally converts naturally occurring androgens into estrogenic compounds, mainly in peripheral tissues. Their use ultimately leads to less available estrogen to stimulate the growth of hr-positive breast cancer cells.
A 2006 meta-analysis 6 of twenty-three randomized trials (8504 patients) showed the efficacy of ais as a first-line treatment for mbca in postmenopausal women ( Table i). Despite in vitro and pharmacodynamic data noting increased potency of aromatase inhibition with letrozole, no clinically meaningful data have demonstrated outcome superiority in comparisons of letrozole with the other ais 20 . For instance, in one trial involving 128 women with advanced breast cancer, the comparison between exemestane and anastrozole resulted in a similar overall response rate (orr) of 15% in both groups and similar overall survival (os) durations of 30.5 months and 33.3 months respectively 21 . Data in the adjuvant setting comparing letrozole with anastrozole also showed similar outcomes with both agents 22 .
Fulvestrant: Fulvestrant is a selective estrogen receptor degrader that blocks estrogen receptor dimerization and dna binding, increases estrogen receptor turnover, and inhibits nuclear uptake of the receptor 23 . Although initially approved as a single-agent monthly intramuscular injection (250 mg per injection), a higher 500 mg dose with a loading dose was proved in the confirm trial to be more effective and is now the preferred dose 23 . Table i describes the phase ii first and the phase iii falcon trials that demonstrated the role of single-agent fulvestrant in the first-line setting for postmenopausal patients [8][9][10][11] The falcon study included only endocrine treatment-naïve patients with mbca, but did allow for one prior line of chemotherapy in the advanced disease setting 11 .

Fulvestrant Plus Anastrozole:
The fact and swog S0226 trials explored fulvestrant-anastrozole as first-line combination therapy, but with conflicting results, as Table i shows 12,13 . The swog S0226 trial enrolled more endocrinenaïve patients and, in addition, used the more effective 500 mg fulvestrant regimen; those differences might explain the difference in outcomes. Additional studies are needed to clarify those discrepancies and to determine whether combination therapy with fulvestrant-anastrozole is truly superior to anastrozole or fulvestrant alone.

CDK 4/6 Inhibitors Plus Endocrine Therapy:
Knowledge of the molecular heterogeneity of breast cancer has led to the identification of the role that cell-cycle signalling plays in breast cancer oncogenesis-in particular, for patients with hr-positive mbca 24 . The cdks drive cell-cycle progression and control transcriptional processes 24,25 . The dysregulation of multiple cdk family members commonly occurs in human cancer. The cyclin D-cdk4/6-retinoblastoma protein-ink4 axis is particularly disrupted, facilitating cancer cell proliferation, thus leading to research targeting cdk 4/6 as a therapeutic approach 24,25 . Palbociclib was the first oral small-molecule cdk 4/6 inhibitor to be developed. It is known to arrest cells in G1 phase by blocking phosphorylation of retinoblastoma protein at cdk 4/6-specific sites [24][25][26] . Preclinical studies suggested growth-inhibitory activity in hr-positive breast cancer cells and potential synergy with endocrine agents 26 .
The paloma-1 trial, an open-label randomized phase ii study of letrozole (2.5 mg daily) with or without palbociclib (125 mg daily on days 1-21 of a 4-week cycle) as first-line therapy, demonstrated activity and clinical benefit for the combination 27 . Postmenopausal women with hr-positive, her2-negative mbca who had received no systemic treatment for their advanced disease were eligible to participate (n = 165). At the time of the final analysis, progression-free survival (pfs) was superior in the palbociclib-letrozole group compared with the letrozole-only group [median: 20.2 months (range: 13.8-27.5 months) vs. 10.2 months (range: 5.7-12.6 months); hazard ratio: 0.49; 95% confidence interval: 0.32 to 0.75; one-sided p = 0.0004]. Grade 3 or 4 neutropenia was reported in 45 of 83 patients (54%) in the palbociclib-letrozole group compared with 1 of 77 patients (1%) in the letrozole group, leucopenia in 16 patients (19%) compared with none, and fatigue in 4 patients (4%) compared with 1 (1%). No cases of febrile neutropenia or neutropenia-related infections were reported during the study 27 . Based on those results, palbociclib-letrozole received U.S. Food and Drug Administration accelerated approval in February 2015 as first-line therapy for hrpositive, her2-negative mbca 28 .
In an underpowered analysis, the os results of the paloma-1 trial were presented at the 2017 American Society of Clinical Oncology meeting, demonstrating a nonsignificant difference in os for palbociclib-letrozole compared  29 . Subsequent phase iii trials in the first-line setting for postmenopausal patients were developed with palbociclib (paloma-2), ribociclib (monaleesa-2), and abemaciclib (monarch 3), as described in Table i [14][15][16][17][18][19] . Those trials excluded patients who had received prior therapy for advanced disease, but did not exclude patients who had been exposed to prior neoadjuvant or adjuvant treatment, including prior endocrine treatment, provided that the disease-free interval after exposure to a nonsteroidal ai was more than 12 months. The results of all studies of cdk 4/6 inhibitors in association with endocrine therapy in the first-line setting were consistent, showing increased rates of orr and pfs, with the os data still being immature.

Second-Line Treatment Options for HR-Positive, HER2-Negative mBCa
For a patient experiencing disease progression after initial endocrine therapy, ongoing endocrine treatment is a reasonable option provided that symptoms from underlying metastatic disease are not present, that the disease continues to be slowly progressive, and that the patient experienced a reasonable response to first-line endocrine therapy. Patients with rapidly progressive or life-threatening metastatic disease should be treated with palliative chemotherapy instead 3 . However, it is important to point out that data published so far with respect to the second-line treatment of hr-positive, her2-negative mbca do not inform the question of how best to sequence therapy after progression on cdk 4/6 inhibitors. Table ii summarizes the most relevant data for postmenopausal patients in the second-line setting. Table ii lists combination trials of cdk 4/6 inhibitors (palbociclib, ribociclib, and abemaciclib) with endocrine therapy in the second-line setting [30][31][32]41 .

CDK 4/6 Inhibitors Plus Endocrine Therapy:
The first reported trial of this combination was palbociclib-fulvestrant in the paloma-3 trial, in which superior orr and pfs rates favoured the combination 30,31 . Eligible patients included those experiencing disease relapse or progression during treatment with prior endocrine therapy for advanced disease or within 12 months of completion of adjuvant therapy. Notably, premenopausal or perimenopausal women (21% of the trial population) were included and received goserelin together with the study treatment.
Interestingly, a description of patterns of disease progression and subsequent therapies and an analysis of the effect of the study treatment on subsequent therapies for participants in the paloma-3 trial were presented at the 2016 San Antonio Breast Cancer Symposium 41 . In both the palbociclib-fulvestrant and the placebo-fulvestrant groups, the most common sites of disease progression were liver [75.3% (n = 149) and 72.3% (n = 94) respectively] and  . The treatment effect for palbociclib-fulvestrant appears to be retained through the immediate next line of treatment after progression. The analysis showed that, for patients with post-study disease progression, the median time until the start of subsequent follow-up treatment was longer in the palbociclib group than in the placebo group. The end of the immediate follow-up therapy was also later in the palbociclib group, regardless of post-treatment modality. Ribociclib is also being evaluated in this setting in the monaleesa-3 trial, a phase iii trial in the first-or secondline setting in which ribociclib is being compared with placebo-fulvestrant, with no reported results to date (see NCT02422615 at http://ClinicalTrials.gov).
Abemaciclib was studied in the phase iii monarch 2 trial, which enrolled women with hr-positive, her2-negative advanced breast cancer who progressed while receiving neoadjuvant or adjuvant endocrine therapy, at 12 months or fewer from end of adjuvant endocrine therapy, or on first-line endocrine therapy for mbca and who had not received chemotherapy for metastatic disease 32 . Patients were stratified by metastatic site (visceral, bone only, or other) and resistance to prior endocrine therapy (primary vs. secondary). Premenopausal and perimenopausal patients received a gonadotropin-releasing hormone agonist. Abemaciclib-fulvestrant was superior to fulvestrant alone for orr and pfs.

Everolimus-Exemestane:
Studies show that the mtor (mechanistic target of rapamycin) inhibitor everolimus is an option, in combination with endocrine therapy, in postmenopausal women for the treatment of ai-resistant hr-positive mbca. The bolero-2 trial described the benefit of everolimus plus the steroidal ai exemestane for orr and pfs, as described in Table ii. The trial enrolled women who had progressed on ais.

Everolimus-Fulvestrant:
The phase ii precog 0102 trial demonstrated a benefit in pfs in favour of combined everolimus-fulvestrant compared with placebo-fulvestrant in postmenopausal women with hr-positive, her2-negative mbca resistant to ai therapy (Table ii) 36 .

AIs:
In the second-line setting, as evidenced in the first line, the nonsteroidal ais show no differences in efficacy 37 . When it comes to exemestane in the second-line setting after progression on a nonsteroidal ai, there is evidence of drug efficacy as described in Table ii 38 .

Fulvestrant As a Single Agent or in Combination with an AI:
Most of the trials evaluating fulvestrant in the second-line setting were designed to use a lower dose of fulvestrant (250 mg monthly) than the dose that the confirm trial proved to be superior 23,42,43 . At the lower dose, no benefit was seen when fulvestrant was compared with ais. Furthermore, the combination of fulvestrant (250 mg monthly) with anastrozole did not show any advantage compared with fulvestrant (250 mg monthly) or an ai alone as second-line treatment, as seen in the phase iii sofea trial described in Table ii 39 .

Tamoxifen:
The available data assessing the benefit of tamoxifen in the second-line setting are limited, but activity has been described for this drug (Table ii) 40 .

Third-or Later-Line Therapy
For women who progress after two lines of endocrine therapy, treatment must be individualized based on prior treatment response, tumour burden, and preferences for treatment. In general, patients who have progressed after multiple lines (>3) of endocrine therapy should likely receive chemotherapy. However, for patients who are asymptomatic with slowly progressive disease, continuation of endocrine therapy is a reasonable strategy 3 . Additionally, new studies of monotherapy with cdk 4/6 inhibitors have shown promising responses in later-line settings.

CDK 4/6 Inhibitors As Single Agents
Recent data from a single-arm phase ii trial in mbca positive for retinoblastoma protein that evaluated palbociclib as a single agent after a median of 2 prior cytotoxic regimens demonstrated activity for that agent. The overall clinical benefit rate was 19%. Grades 3 and 4 toxicities included neutropenia (51%), anemia (5%), and thrombocytopenia (22%). No tumour biomarker identified a sensitive population 44 . Abemaciclib showed activity after a median of 3 lines of prior systemic treatment, as demonstrated in the singlearm phase ii monarch 1 study, which was designed to evaluate the single-agent activity and adverse event profile of that drug. The orr was 19.7%, the clinical benefit rate was 42.4%, the median pfs was 6.0 months, and the median os was 17.7 months. The most common adverse events of any grade were diarrhea, fatigue, and nausea 45 .

Pre-and Perimenopausal Patients
Historically, for pre-and perimenopausal patients with hr-positive, her2-negative mbca, data are available for ovarian suppression alone, for single-agent tamoxifen, for ovarian suppression plus tamoxifen, and for ovarian suppression plus ais 46-50 . However, recent data now show the benefits of the addition of targeted agents to endocrine therapy compared with endocrine monotherapy, as described in Table iii.

Adverse Events With Combination Therapy Using Endocrine and Targeted Agents
The combination of endocrine treatment with targeted agents has shown increased response rates and improved pfs in many trials. However, combination therapy is associated with increased toxicity, which has to be considered when choosing the optimal therapy for each individual patient based on comorbidities, preferences, burden of disease, financial and social supports, and drug availability. Table iv summarizes the adverse events observed with targeted therapy.  The adverse events most commonly seen with cdk 4/6 inhibitors in combination with endocrine therapy in phase iii trials were neutropenia, infections, fatigue, and nausea. Rates of febrile neutropenia were low in all trials. In particular, trials of ribociclib showed increased values in liver function tests and prolongation of QTcF interval, and abemaciclib trials showed higher rates of diarrhea. The use of mtor inhibitors was associated with the risk of stomatitis, rash, fatigue, diarrhea, decreased appetite, and pneumonitis.

DISCUSSION AND SUMMARY
In recent years, a significant evolution has occurred in the management of hr-positive mbca. Given the emerging evidence, it is now essential to optimize therapy and to choose a treatment sequence strategy that considers both patient-and tumour-related factors.
In general, endocrine therapy represents the mainstay for most patients with hr-positive mbca, with palliative chemotherapy being reserved for life-threatening advanced disease or patients with visceral crisis. Endocrine monotherapy is still considered an effective treatment option, especially for patients whose disease course is more indolent (for example, a disease-free interval prolonged beyond 2 years) or for patients presenting with de novo low-burden and non-visceral metastatic disease. Combination therapy with endocrine and targeted agents, including either cdk 4/6 inhibitors or mtor inhibitors, is now considered a treatment option in patients who do not meet the foregoing criteria for chemotherapy or endocrine monotherapy. We propose an algorithm based on the inclusion criteria in the key studies described in our review, on current guidelines, on the efficacy information available to date, and on results from important subgroups evaluated in the relevant studies (Figure 1) 52 .
Although the first-line treatment approach might be more straightforward, many questions remain unanswered, including the ideal treatment sequence that will optimize survival based on tumour biology and de novo or acquired treatment resistance factors. In routine clinical practice, clinicians and patients have to evaluate several factors beyond those that can be considered in our algorithm, including quality of life, patient preference, and access to therapies.
In Canada, based on the Ontario Drug Benefit list price, the approximate costs of a 28-day course of treatment were CA$39 for letrozole, CA$36 for anastrozole, CA$37 for exemestane, and CA$10 for tamoxifen. The addition of palbociclib at the recommended dose of 125 mg once daily for 21 days, followed by 7 days off treatment, adds CA$6250 per 28-day course at the list price 51 and brings a need for monthly monitoring and bloodwork for neutropenia, together with the associated costs. Based on those findings, cost-effectiveness analyses of new targeted agents are needed to implement them in routine mbca care and in various health care systems. For instance, based on the paloma-2 trial, a Swiss cost-effectiveness study evaluated the burden of the addition of palbociclib to letrozole in mbca 53 . The results showed that a considerable price reduction for palbociclib would be needed to make the drug cost-effective, given the estimated additional annual cost of approximately US$22 million to the system. By themselves, some drugs might therefore bring an additional amount to the total treatment cost that might not be affordable for patients, health care systems, and government funding bodies.
It is also important to point out that the treatment of hr-positive, her2-negative mbca is rapidly evolving; results from ongoing clinical trials expected to be published in the next few years will most likely affect our proposed treatment algorithm. Lastly, the hope is that, in the genomic era, novel predictive biomarkers other than hrs and her2 will be available to narrow the population of patients who will ultimately derive the greatest magnitude of benefit with the addition of the new targeted agents delivered on a backbone of hormonal therapy.