Everolimus Plus Endocrine Therapy for Postmenopausal Women With Estrogen Receptor–Positive, Human Epidermal Growth Factor Receptor 2–Negative Advanced Breast Cancer: A Clinical Trial

Key Points

Question  Does first-line everolimus plus endocrine therapy provide a clinical benefit for patients with estrogen receptor–positive, human epidermal growth factor receptor 2–negative advanced breast cancer?

Findings  In this phase 2, single-arm study, 202 patients treated with everolimus plus letrozole in the first-line setting achieved a median progression-free survival of 22.0 months; median overall survival was not reached. For 50 patients whose cancer progressed and who received continued treatment with everolimus plus exemestane, median progression-free survival was 3.7 months.

Meaning  These results suggest a rationale for providing first-line everolimus plus letrozole therapy to patients with estrogen receptor–positive advanced breast cancer.

Importance  Cotargeting the mammalian target of rapamycin pathway and estrogen receptor may prevent or delay endocrine resistance in patients receiving first-line treatment for advanced breast cancer.

Objective  To investigate the combination of everolimus plus endocrine therapy in first-line and second-line treatment settings for postmenopausal women with estrogen receptor–positive, human epidermal growth receptor 2–negative advanced breast cancer.

Design, Setting, and Participants  In the multicenter, open-label, single-arm, phase 2 BOLERO-4 (Breast Cancer Trials of Oral Everolimus) clinical trial, 245 patients were screened for eligibility; 202 were enrolled between March 7, 2013, and December 17, 2014. A median follow-up of 29.5 months had been achieved by the data cutoff date (December 17, 2016).

Interventions  Patients received first-line treatment with everolimus, 10 mg/d, plus letrozole, 2.5 mg/d. Second-line treatment with everolimus, 10 mg/d, plus exemestane, 25 mg/d, was offered at the investigator’s discretion upon initial disease progression.

Main Outcomes and Measures  The primary end point was investigator-assessed progression-free survival in the first-line setting per Response Evaluation Criteria in Solid Tumors, version 1.0. Safety was assessed in patients who received at least 1 dose of study medication and at least 1 postbaseline safety assessment.

Results  A total of 202 women treated in the first-line setting had a median age of 64.0 years (interquartile range, 58.0-70.0 years) with metastatic (194 [96.0%]) or locally advanced (8 [4.0%]) breast cancer. Median progression-free survival was 22.0 months (95% CI, 18.1-25.1 months) with everolimus and letrozole. Median overall survival was not reached; 24-month estimated overall survival rate was 78.7% (95% CI, 72.1%-83.9%). Fifty patients started second-line treatment; median progression-free survival was 3.7 months (95% CI, 1.9-7.4 months). No new safety signals were observed. In the first-line setting, the most common all-grade adverse event was stomatitis (139 [68.8%]); the most common grade 3 to 4 adverse event was anemia (21 [10.4%]). In the second-line setting, the most common adverse events were stomatitis and decreased weight (10 [20.0%] each); the most common grade 3 to 4 adverse event was hypertension (5 [10.0%]). There were 50 (24.8%) deaths overall during the study; 40 were due to study indication (breast cancer).

Conclusions and Relevance  The results of this trial add to the existing body of evidence suggesting that everolimus plus endocrine therapy is a good first-line treatment option for postmenopausal women with estrogen receptor–positive, human epidermal growth factor receptor 2–negative advanced breast cancer.

Trial Registration  clinicaltrials.gov Identifier: NCT01698918

Endocrine-based single-agent or combination therapy is the standard of care for postmenopausal women with hormone receptor–positive (HR+), human epidermal growth factor receptor 2–negative (HER2–) advanced breast cancer.¹ However, most responders to initial endocrine therapy eventually acquire resistance, experience relapse, and develop progressive disease.^2,3 Dual inhibition of the mammalian target of rapamycin (mTOR) pathway and estrogen receptor can restore sensitivity to patients with endocrine therapy–resistant advanced breast cancer.^2,4 Preclinical studies have also shown that early intervention with mTOR inhibitors plus endocrine therapy could delay or prevent endocrine resistance by preventing the emergence of hormone-independent cells.⁵

Everolimus is an oral, potent, selective mTOR inhibitor that acts synergistically with letrozole.^3,6,7 Phase 2 studies combining everolimus with endocrine therapy, such as tamoxifen (Tamoxifen Plus Everolimus [TAMRAD] study), letrozole, or fulvestrant (PrECOG 0102 study) reported improved efficacy vs endocrine therapy alone in postmenopausal women with HR+ breast cancer in the neoadjuvant and advanced settings.^4,6,8 In the phase 3 BOLERO-2 (Breast Cancer Trials of Oral Everolimus) trial, everolimus plus exemestane significantly improved median progression-free survival (PFS) vs placebo plus exemestane in women with HR+, HER2– advanced breast cancer progressing with nonsteroidal aromatase inhibitors (7.8 vs 3.2 months; hazard ratio [HR], 0.45; 95% CI, 0.38-0.54), which led to the approval of everolimus plus exemestane in this setting.^3,9 In an exploratory analysis of BOLERO-2, median PFS was also improved in patients receiving first-line treatment with everolimus plus exemestane following recurrence after adjuvant therapy (HR, 0.39; 95% CI, 0.25-0.62).¹⁰ The consistent efficacy of everolimus in these studies supports the rationale for combining everolimus with endocrine therapy for first-line treatment of HR+ advanced breast cancer.

Herein, we assess first-line treatment with everolimus plus letrozole in postmenopausal women with estrogen receptor–positive (ER+), HER2– advanced breast cancer. We also investigate continuing treatment with everolimus in combination with exemestane in patients whose disease progressed during everolimus plus letrozole therapy.

BOLERO-4 is an open-label, multicenter, single-arm, phase 2 study conducted at 56 centers across 13 countries (eTable 1 in the Supplement). Eligible patients were postmenopausal women with ER+, HER2– advanced breast cancer. Patients were required to have measurable disease per Response Evaluation Criteria in Solid Tumors (RECIST), version 1.0,¹¹ or bone lesions (lytic or mixed). Eastern Cooperative Oncology Group performance status of 0 to 2 and adequate bone marrow, coagulation, liver, and renal function were also required. Prior neoadjuvant endocrine therapy was permitted, but therapy with nonsteroidal aromatase inhibitors must have been completed 1 or more years before enrollment. Patients who had received prior hormonal or other systemic therapy for advanced disease or prior treatment with (or known hypersensitivity to) an mTOR inhibitor were excluded. Further inclusion and exclusion criteria are presented in the eMethods in the Supplement.

All patients provided written informed consent before enrollment. Study conduct adhered to Good Clinical Practice guidelines, local regulations, and the ethical principles of the Declaration of Helsinki,¹² and approval came from institutional review boards, independent ethics committees, and/or research ethics boards at each center (eAppendix in the Supplement). A steering committee supervised the study per the protocol.

Patients received first-line oral treatment with everolimus, 10 mg/d, and letrozole, 2.5 mg/d, until disease progression, unacceptable toxic effects, or withdrawal of consent. Patients who discontinued treatment due to disease progression could then, at the investigator’s discretion, receive second-line oral treatment with everolimus, 10 mg/d, and exemestane, 25 mg/d, until further disease progression, unacceptable toxic effects, or withdrawal of consent. Dose modifications were allowed and are described in the eMethods in the Supplement.

Tumors were assessed by the investigators (M.R., T.B., C.V., M.Ö., S.J.A., F.M.C., R.H., T.T., C.F., J.J., V.S.) and/or designated radiologist per RECIST, version 1.0, using mainly computed tomography or magnetic resonance imaging at screening and every 8 weeks (±1 week) thereafter until the end of first-line treatment; assessments were also performed every 8 weeks in the second-line setting. A confirmation scan or photograph for measurable skin lesions was obtained at least 4 weeks after the initial observation of a response.

An exploratory, randomized substudy that assessed alcohol-free dexamethasone mouthwash vs the local standard of oral care for the treatment of symptomatic stomatitis among patients in the United States has been previously reported (eMethods in the Supplement).¹³

The primary end point was investigator-assessed PFS in the first-line setting per RECIST, version 1.0. Secondary end points included overall survival (OS) in the first-line setting, PFS in the second-line setting per RECIST, version 1.0, and overall response (complete or partial response) and clinical benefit (complete or partial response, or stable disease, each lasting ≥24 weeks) in both settings per RECIST, version 1.0. Definitions of PFS and OS are given in the eMethods in the Supplement. Safety, assessed by the frequency of adverse events graded per National Cancer Institute Common Terminology Criteria for Adverse Events, version 4.0,¹⁴ was determined in both settings.

First-line and second-line efficacy were analyzed in all patients assigned to first-line treatment (full analysis set) and all patients who received at least 1 dose of second-line treatment. Median PFS and median OS were estimated using the Kaplan-Meier method and presented with exact 95% CIs. Interquartile ranges (IQRs) were defined as 25th to 75th percentiles.

If the disease did not progress or the patient did not die by the analysis cutoff date, or if a patient received any further antineoplastic therapy (including the per-protocol second-line treatment for the analysis of the first-line data), PFS was censored at the time of the last tumor assessment. Overall survival was censored at the date of last contact if a death was not observed by the data cutoff. The proportion of patients with an overall response and clinical benefit are presented with exact 95% CIs based on the Clopper-Pearson method. Safety was reported for all patients who received at least 1 dose of study medication and at least 1 postbaseline safety assessment.

Sample size was calculated based on an estimated median PFS with reasonable accuracy (width of 95% CI) for first-line treatment with everolimus plus letrozole. It was estimated that, with 200 patients enrolled, considering a recruitment period of 18 months and 1 year of follow-up, expected 95% CIs for median PFS at 11, 12, 13, and 14 months (with 10% lost to follow-up) were 9.32-12.98, 10.13-14.22, 10.93-15.46, and 11.73-16.71, respectively. SAS, version 9.4 (SAS Institute Inc) was used in data analysis.

Between March 7, 2013 and December 17, 2014, 245 patients were screened for eligibility, 202 of whom were enrolled and provided with first-line treatment with everolimus and letrozole (Figure 1). In the first-line setting, median age was 64.0 years (IQR, 58.0-70.0 years) (eTable 2 in the Supplement). Most of the 202 patients had metastatic (194 [96.0%]) rather than locally advanced (8 [4.0%]) breast cancer, 3 or more metastatic sites (137 [67.8%]), and visceral (excluding central nervous system) metastases (123 [60.9%]). Median time since the most recent relapse or recurrence to starting first-line treatment was 1.2 months (IQR, 0.76-1.84 months); the date of the most recent recurrence could have been the time of first diagnosis of advanced disease. The data cutoff for the present analyses was December 17, 2016, corresponding to 24 months after the last patient’s first visit.

First-line treatment was ongoing in 50 (24.8%) patients, while 152 (75.2%) had discontinued treatment. The most frequent reasons given as the primary cause of first-line treatment discontinuation were disease progression (104 [51.5%]) and adverse events (32 [15.8%]). Median exposure to everolimus and letrozole was 64.2 weeks (IQR, 16.3-111.4 weeks) and 75.1 weeks (IQR, 21.9-112.3 weeks), respectively; median actual dose intensity of everolimus was 8.5 mg/d (IQR, 6.0-9.9 mg/d). Following disease progression in the first-line setting, 50 (24.8%) patients received second-line treatment with everolimus plus exemestane at the investigator’s discretion. In the second-line setting, median patient age was 62.0 years (IQR, 57.0-66.0 years) (eTable 2 in the Supplement). Second-line treatment was ongoing in 16 (32.0%) patients, while 34 (68.0%) had discontinued treatment. The most frequent reason for second-line treatment discontinuation was disease progression (28 [56.0%]). Median treatment exposure to everolimus or exemestane was 12.6 weeks (IQR, 8.0-29.6 weeks); median dose intensity of everolimus was 8.3 mg/d (IQR, 5.0-10.0 mg/d). Treatments received by patients after permanently discontinuing first- and second-line therapy are provided in eTable 3 in the Supplement.

In the first-line setting, 108 PFS events were reported. Median duration of follow-up was 29.5 months (IQR, 26.4-33.6 months) from the start of first-line treatment. Median PFS was 22.0 months (95% CI, 18.1-25.1 months) with everolimus and letrozole in the full analysis set (Table 1; Figure 2). Median PFS determined without censoring patients who initiated new anticancer therapies was 20.9 months (95% CI, 17.9-24.0 months). Overall response and clinical benefit rates were 45.0% (95% CI, 38.1%-52.2%) and 74.3% (95% CI, 67.7%-80.1%), respectively. The majority of patients had a reduction in tumor size from baseline with everolimus plus letrozole (eFigure in the Supplement). Median OS in patients receiving everolimus plus letrozole was not reached (95% CI, 37.0-not estimable) (Table 2; Figure 3). Fifty (24.8%) deaths overall had been reported and data from 152 (75.2%) patients had been censored. The estimated OS rate at 24 months was 78.7% (95% CI, 72.1%-83.9%).

In the second-line setting, 31 PFS events were reported. Median PFS was 3.7 months (95% CI, 1.9-7.4 months) with everolimus and exemestane (Table 1; Figure 2). Overall response and clinical benefit rates were 6.0% (95% CI, 1.3%-16.5%) and 28.0% (95% CI, 16.2%-42.5%), respectively.

In the first-line setting, 202 patients receiving everolimus plus letrozole were assessed for safety. Everolimus dose reductions and interruptions were required in 110 (54.5%) and 143 (70.8%) patients, respectively (further details are provided in eTable 4 in the Supplement). Adverse events (any grade) requiring dose adjustments or interruptions occurred in 146 (72.3%) patients, the most common of which was stomatitis (41 [20.3%]); median time to first stomatitis occurrence was 3.1 weeks. The most common adverse events (any grade) occurring in 40% or more of patients, regardless of causality, were stomatitis (139 [68.8%]), decreased weight (89 [44.1%]), and diarrhea (83 [41.1%]) (eTable 5 in the Supplement). Most of these events were grade 1 to 2 in severity; the most common grade 3 to 4 adverse event was anemia (21 [10.4%]). Stomatitis was the most common all-grade adverse event suspected to be drug related (137 [67.8%] patients). Serious adverse events occurred in 62 (30.7%) patients, were suspected to be drug related in 24 (11.9%) patients, and were mostly grade 3 to 4 in severity (eTable 6 in the Supplement). Pneumonia was the most common (any grade) serious adverse event regardless of causality (9 [4.5%] patients) and was suspected to be drug related in 5 (2.4%) patients. Pneumonia was also the most common grade 3 to 4 serious adverse event regardless of causality (6 [3.0%] patients) and suspected to be drug related in 4 (2.0%) patients. Adverse events leading to discontinuation occurred in 51 (25.2%) patients overall and were suspected to be drug related in 43 (21.3%) patients (eTable 7 in the Supplement); stomatitis was the most common adverse event leading to discontinuation regardless of causality (7 [3.5%] patients).

In the second-line setting, 50 patients receiving everolimus plus exemestane were assessed for safety. Everolimus dose reductions and interruptions were required in 9 (18.0%) and 14 (28.0%) patients, respectively (further details provided in eTable 4 in the Supplement). Adverse events (any grade) requiring dose adjustments or interruptions occurred in 14 (28.0%) patients, the most common of which was pneumonitis (3 [6.0%]). The most common adverse events (any grade) occurring in 20% or more of patients, regardless of causality, were stomatitis and decreased weight (10 [20.0%] each) (eTable 5 in the Supplement). Similar to the first-line setting, most adverse events were grade 1 to 2 in severity; the most common grade 3 to 4 adverse event was hypertension (5 [10.0%] patients). Stomatitis was the most common all-grade adverse event suspected to be drug related (10 [20.0%] patients). Serious adverse events occurred in 8 (16.0%) patients, were suspected to be drug related in 3 (6.0%) patients, and were mostly grade 3 to 4 in severity (eTable 6 in the Supplement). Adverse events leading to discontinuation occurred in 3 (6.0%) patients, and all were suspected to be drug related (eTable 7 in the Supplement).

A total of 50 (24.8%) patients died during the study, most commonly due to study indication (breast cancer, 40 [19.8%]). Seven (3.5%) patients died within 28 days of completing first-line treatment due to study indication (4 [2.0%]) and septic shock, pyrexia, and pneumonia (1 each [0.5%]). Two (4.0%) patients died within 28 days of completing second-line treatment due to study indication and septic shock (1 each [2.0%]). Overall, 9 (4.5%) patients experienced serious adverse events (any grade) suspected to be drug related before their deaths (details provided in eResults in the Supplement).

Current guidelines recommend offering endocrine therapy as a first-line treatment to patients with HR+, HER2– advanced breast cancer, excluding those with immediately life-threatening disease.^1,15,16 In this single-arm, open-label, phase 2 study, median PFS with the combination of everolimus and letrozole in the first-line setting was 22.0 months (95% CI, 18.1-25.1 months) in postmenopausal women with ER+, HER2– advanced breast cancer; the median OS had not been reached after a median follow-up of 29.5 months. In addition to the statistically significant median PFS improvement observed in the BOLERO-2 study following second-line treatment with everolimus plus exemestane vs placebo plus exemestane (investigator review: 7.8 vs 3.2 months; HR, 0.45; 95% CI, 0.38-0.54 months; central review: 11.0 vs 4.1 months; HR, 0.38; 95% CI, 0.31-0.48 months), the findings of BOLERO-4 support the rationale for dual inhibition of the mTOR pathway and estrogen receptor in the first-line setting.^3,9 The median PFS achieved in this study was particularly impressive given the heavy baseline disease burden of the patient population: 96.0% had metastatic disease, 67.8% had 3 or more metastatic sites, and 60.9% had visceral metastases. A sensitivity analysis, performed without censoring patients who started new anticancer therapies, achieved a median PFS that was similar to the primary analysis. The treatment landscape for HR+, HER2– advanced breast cancer has been evolving in recent years with the licensing of cyclin-dependent kinase inhibitors, after clinical studies demonstrated the benefit of adding palbociclib or ribociclib succinate to endocrine therapy in the first-line setting.^17,18 Despite these newer treatment options, previously reported preclinical and clinical data show that combinations of mTOR inhibitors and endocrine therapy remain important and effective options. The combination of everolimus and letrozole has demonstrated synergistic activity in preclinical models and in the neoadjuvant breast cancer setting.^5,6 In the second-line setting, the TAMRAD study also showed that the addition of everolimus to tamoxifen provided improved efficacy compared with tamoxifen alone,³ while the PrECOG 0102 study confirmed the clinical benefit of combining everolimus with fulvestrant.⁸ The study reported herein provides further indication that dual inhibition with the mTOR inhibitor everolimus combined with endocrine therapy offers an important option in the management of HR+, HER2– advanced breast cancer.

A limited number of studies have investigated mTOR inhibitors combined with endocrine therapy in the first-line setting, and in 1 such study, the combination of temsirolimus and letrozole did not improve PFS or OS in the first-line setting vs letrozole alone.¹⁹ However, caution should be used when comparing these data with BOLERO-4 study data due to differences between the study designs and patient populations (BOLERO-4 was a single-arm, open-label phase 2 study of patients with HER2– disease that had progressed during prior aromatase inhibitor therapy, whereas HORIZON was a phase 3, placebo-controlled study of aromatase inhibitor–naive patients with HER2– or HER2-positive disease that was stopped early due to lack of efficacy) and between the modes of action of the study drugs. Data describing continued treatment with an mTOR inhibitor but combined with a different endocrine therapy partner in patients whose disease has progressed with prior mTOR inhibitors are equally scarce. The BOLERO-4 study also explored continued treatment with everolimus plus exemestane in patients whose disease progressed with everolimus plus letrozole in the first-line setting. Other second-line studies in patients whose disease had not progressed during everolimus therapy have shown that everolimus combined with exemestane or fulvestrant offered median PFS improvements of 4.6 to 5.3 months vs endocrine therapy alone in patients with aromatase inhibitor–resistant breast cancer.^8,9 Although in our study second-line treatment with everolimus and exemestane showed limited efficacy (median PFS, 3.7 months; 95% CI, 1.9-7.4 months), the treatment was offered only to a select number of patients. In addition, the results should be interpreted with caution due to the small sample size, pretreatment with or prior exposure to everolimus in the first-line setting, the relatively advanced disease present among the patients, the potential similar mechanisms of letrozole and exemestane, and differences in study design.

No new safety signals were observed in either setting of this study. Overall safety was generally manageable, consistent with the known safety profile of everolimus, and incidences of on-treatment death due to pneumonia or septic shock were consistent with the BOLERO-2 study.³ Everolimus dose reductions and interruptions were relatively high in the first-line setting, and most were due to adverse events. However, most adverse events were grade 1 to 2 in severity, and grade 3 to 4 adverse events were relatively uncommon. Higher rates of stomatitis were noted in the first-line setting compared with the second-line setting, reflective of the rapid onset associated with everolimus-related stomatitis.²⁰ Overall safety was better in the second-line setting and likely reflects the selection of patients for continued treatment based on their tolerability of everolimus in the first-line setting.

Although the results from this study are impressive, cross-trial comparisons should be made with caution due to the single-arm, open-label, phase 2, nonrandomized design of this BOLERO-4 study.

The PFS benefit demonstrated by everolimus plus letrozole was within range of that observed with combinations of aromatase inhibitors and cyclin-dependent kinase 4/6 (CDK4/6) inhibitors in the advanced breast cancer setting, despite the heavy disease burden among the postmenopausal ER+, HER2– advanced breast cancer patient population. While combinations of aromatase inhibitors and CDK4/6 inhibitors are already a preferred first-line option in many countries, these combinations have not yet shown that they can provide an additional OS benefit vs aromatase inhibitors alone and are not universally accessible due to their high cost. The PFS benefit demonstrated by everolimus plus letrozole in this study also exceeded that observed with letrozole alone in previous phase 3 studies of postmenopausal women with HR+ advanced breast cancer (median PFS, 9.0-9.4 months).^19,21 There may therefore still be a place for everolimus in the advanced breast cancer treatment landscape, and our hypothesis that the combination of everolimus and letrozole can enhance the effectiveness of endocrine therapy in the first-line setting and delay or potentially prevent the development of endocrine resistance remains an interesting concept.

Following the arrival of CDK4/6 inhibitors, there is also interest in determining the optimal sequence of therapies. Despite the challenges posed by the patient population, our findings are insightful and provide data describing everolimus in areas of the advanced breast cancer landscape where existing data are scarce. Furthermore, with new and emerging phosphatidylinositol 3-kinase inhibitors receiving significant attention, there is sufficient evidence to suggest that the phosphatidylinositol 3-kinase/mTOR pathway remains important and warrants further exploration in the treatment of HR+, HER2– advanced breast cancer.

Accepted for Publication: January 3, 2018.

Corresponding Author: Melanie Royce, MD, PhD, University of New Mexico Comprehensive Cancer, 7605 Via de Calma NE, Albuquerque, NM 87113 (mroyce@mroycemdphd.com).

Published Online: March 22, 2018. doi:10.1001/jamaoncol.2018.0060

Open Access: This article is published under the JN-OA license and is free to read on the day of publication.

Author Contributions: Drs Royce and Lin had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Royce, Azevedo, Melo Cruz, Arce, Ridolfi, Lin, Cardoso.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Royce, Bachelot, Villanueva, Özgüroğlu, Melo Cruz, Toyama, Gradishar, Ridolfi, Lin, Cardoso.

Critical revision of the manuscript for important intellectual content: Royce, Bachelot, Villanueva, Özgüroğlu, Azevedo, Melo Cruz, Debled, Hegg, Falkson, Jeong, Srimuninnimit, Gradishar, Arce, Ridolfi, Lin, Cardoso.

Statistical analysis: Gradishar, Ridolfi, Lin.

Administrative, technical, or material support: Debled, Toyama, Falkson, Lin.

Study supervision: Royce, Villanueva, Azevedo, Melo Cruz, Hegg, Jeong, Arce, Lin, Cardoso.

Conflict of Interest Disclosures: Dr Royce received research funding from Novartis Pharmaceuticals Corporation, was a consultant for Celltrion and BCI Pharma, and received honoraria from Novartis Pharmaceuticals Corporation and Syndax. Dr Bachelot received research funding from Roche and Novartis Pharmaceuticals Corporation and was a consultant for and received travel expenses from AstraZeneca, Roche, Novartis Pharmaceuticals Corporation, and Pfizer. Dr Villanueva was an advisory board member for Novartis Pharmaceuticals Corporation. Dr Melo Cruz received research funding from Novartis Pharmaceuticals Corporation, Janssen, Roche, and Celgene; and received travel, accommodation, and expenses from Janssen Pharmaceuticals. Dr Toyama received research funding from Eisai, Chugai, Novartis Pharmaceuticals Corporation, Nippon Kayaku, Kyowa Hakko Kirin, Daiichi Sankyo, and Takeda. Dr Falkson received research funding from Novartis Pharmaceuticals Corporation, Cascadian Therapeutics, Genentech, and EMD Serono, and was a consultant for and received honoraria from Biotheranostics. Dr Jeong received research funding from Dong-A, Boryung, LG Life Sciences, and Antigen, and received honoraria from Roche, Alvogen, Novartis Pharmaceuticals Corporation, Pfizer, and Covidien. Ms Arce is a Novartis Pharmaceuticals Corporation employee and reported stock ownership with BD. Ms Ridolfi is a Novartis Pharma S.A.S. employee. Dr Lin is a Novartis Pharmaceuticals Corporation employee and reported stock ownership with Novartis Pharmaceuticals Corporation. Dr Cardoso received research funding for clinical trials by institution and was a consultant for Astellas/Medivation, AstraZeneca, Celgene, Daiichi Sankyo, Eisai, GE Oncology, Genentech, GSK, MacroGenics, Merck, Merus BV, Novartis Pharmaceuticals Corporation, Pfizer, Pierre Fabre, Roche, Sanofi, and Teva. No other disclosures are reported.

Funding/Support: This study was funded by Novartis Pharmaceuticals Corporation.

Role of the Funder/Sponsor: The study was designed, conducted, and analyzed by the funder in conjunction with the investigators and study steering committee. The funder provided the study drugs. The authors confirm adherence to the study protocol and vouch for the accuracy and completeness of the data. Dr Royce also had access to the study data, was involved in their interpretation and analysis, and had final responsibility for the decision to submit for publication.

Additional Contributions: We thank the patients who participated in the BOLERO-4 trial and the investigators, nurses, and clinical research associates from the participating centers for their support. Novartis Pharmaceuticals Corporation provided financial support for medical editorial assistance performed by Matthew Young, DPhil, and Sara Shaw, PhD.