Eribulin Plus Pembrolizumab in Patients with Metastatic Triple-Negative Breast Cancer (ENHANCE 1): A Phase Ib/II Study

Triple-negative breast cancer (TNBC) is characterized by aggressive clinicopathologic features (e.g., larger tumor size, higher grade tumors; ref. 1), a poorer prognosis, earlier recurrence, and high death rates (1, 2). While some combination therapies have recently shown promise in the first-line (1L) setting for metastatic TNBC (mTNBC), limited data are available for second-line and later (2L+) therapeutics. The median overall survival (OS) for patients with mTNBC treated in the 2L+ is 8 to 15 months (2). There is, therefore, an unmet clinical need to identify new and more effective treatments for improving the prognosis of patients with mTNBC, especially in the later-line setting.

Approximately 40% of TNBC tumors express programmed death-ligand 1 (PD-L1; ref. 3). Binding of PD-L1 to programmed death receptor-1 (PD-1) may facilitate tumor immunity by suppressing cytotoxic T-cell responses (4). Pembrolizumab, a monoclonal anti-PD-1 antibody, is approved for the treatment of several advanced cancers (5). In a randomized, phase III study in patients with PD-L1–positive (PD-L1⁺) TNBC (KEYNOTE-355), patients with a combined positive score (CPS) ≥10, treated with 1L pembrolizumab plus chemotherapy had improved progression-free survival (PFS) compared with chemotherapy alone [9.7 vs 5.6 months; HR = 0.65, 95% confidence interval (CI): 0.49–0.86; refs. 6, 7]. Patients with mTNBC who received pembrolizumab monotherapy as 2L+ (KEYNOTE-086 cohort A) achieved an objective response rate (ORR) of 5%, regardless of PD-L1 status (8). A randomized, phase III clinical trial comparing nab-paclitaxel with or without atezolizumab (an anti-PD-L1 antibody) in patients with previously untreated advanced TNBC revealed that the addition of atezolizumab significantly improved median PFS compared with chemotherapy alone in both the intent-to-treat population (7.2 vs 5.5 months, respectively, P = 0.002) and in the PD-L1⁺ subgroup of patients (7.5 vs 5.0 months, P < 0.001; ref. 3), and was associated with a clinically relevant improvement in median OS in the PD-L1⁺ subgroup (21 vs 19 months; ref. 9).

Eribulin mesylate (eribulin) is a synthetic analog of halichondrin B that inhibits microtubule dynamics (10–12). Preclinical studies suggest that eribulin reverses epithelial-to-mesenchymal transition, induces vascular remodeling, and suppresses cancer cell migration and invasion (12). In the United States, eribulin is indicated for patients with metastatic breast cancer (MBC) following ≥2 prior lines of chemotherapy for metastatic disease (13). Prior therapy should include an anthracycline and a taxane in the adjuvant or metastatic setting (13). Eribulin monotherapy has shown promise for patients with mTNBC. In a retrospective subgroup analysis of a phase III clinical trial (Study 301), eribulin prolonged median OS versus capecitabine as first-to-third-line (1–3L) treatment for mTNBC (14.4 vs 9.4 months; 95% CI: 0.54–0.91; P = 0.01; ref. 14). In a pooled analysis from two large-scale, phase III MBC trials (Study 301 and EMBRACE), eribulin prolonged median OS versus capecitabine or treatment of physician's choice (12.9 vs 8.2 months; HR: 0.74; P = 0.006) in the mTNBC subgroup (15).

Both eribulin and pembrolizumab monotherapies have shown promising antitumor activity in patients with mTNBC (8, 14–17). This open-label, single-arm, phase Ib/II study evaluated the safety and efficacy of eribulin plus pembrolizumab combination as 1–3L treatment in patients with mTNBC.

This analysis (data cut-off date: July 31, 2019) of an ongoing, open-label, single-arm, phase Ib/II study evaluated the safety and efficacy of eribulin plus pembrolizumab in patients with mTNBC (ClinicalTrials.gov identifier: NCT02513472; ref. 18). The study was conducted across 18 sites in the United States, in accordance with the World Medical Association Declaration of Helsinki, International Conference on Harmonization Good Clinical Practice guidelines, and the U.S. code of federal regulations. The protocol was approved by each institutional review board. All participants provided written informed consent prior to study enrollment.

In the initial, safety run-in cohort (phase Ib), patients received intravenous eribulin 1.4 mg/m² on day 1 and day 8 of each 21-day cycle, plus intravenous pembrolizumab 200 mg on day 1 of each 21-day cycle. This dose became the recommended phase II dose if ≤1 patient had a dose-limiting toxicity (DLT) in the first cycle. If ≥2 patients experienced a DLT at this first dose level, then the dose of eribulin would be lowered to 1.1 mg/m². Patients in the phase II part of the study were enrolled into two strata based on the number of prior systemic therapies in the metastatic setting (stratum 1, no prior treatment or stratum 2, 1–2 prior treatments).

Eligible adults ≥18 years old had mTNBC, as defined by the American Society of Clinical Oncology/College of American Pathologists (19, 20), and measurable disease according to RECIST version 1.1 (21), as assessed locally by investigators. Additional inclusion criteria were an Eastern Cooperative Oncology Group performance status of ≤1, life expectancy of ≥3 months, ≤2 prior systemic anticancer therapies in the metastatic setting, and adequate bone marrow, renal, and hepatic function. Patients were excluded if they had received prior treatment with eribulin or an agent targeting the PD-1–signaling pathway, or adjuvant therapy in the last 6 months. Additional exclusion criteria are detailed in the Online Resource Materials (Supplementary Materials and Methods). Patients with previously treated stable brain metastasis were allowed (exclusions apply, see Supplementary Materials and Methods).

The primary objectives of the phase Ib portion of the study were to evaluate the safety and tolerability of eribulin plus pembrolizumab. All adverse events (AE) were recorded and monitored. Treatment-emergent AEs (TEAE) were graded using Common Terminology Criteria for Adverse Events (version 4.03) and characterized using the Medical Dictionary for Regulatory Activities (version 22.0). AEs were evaluated as related or unrelated to the study drugs by the investigator. DLTs were assessed in the first dosing cycle (further details in Supplementary Materials and Methods). Treatment-related AEs that resulted in study-drug discontinuations, or study-drug delays of ≥2 weeks, were also considered a DLT.

The primary efficacy endpoint of the phase II portion of the study was the ORR, defined as the percentage of patients with best overall response of complete response (CR) or partial response (PR) as assessed by independent imaging review. Secondary efficacy endpoints were PFS, OS, duration of response (DOR), and clinical benefit rate [CBR; the percentage of patients with CRs, PRs, and durable (≥24 weeks) stable disease (SD)]. Subgroup analyses of efficacy by PD-L1–expression status (i.e., positive or negative) were also performed.

Tumor assessments were performed at baseline, then every 9 weeks after the first dose, using CT or MRI scans. CRs or PRs were confirmed by a second examination performed ≥4 weeks after the first observation of response. Best overall response of SD required ≥1 posttreatment assessment that met the SD criteria >8 weeks after the start of treatment. For patients with a history of stable brain metastases, brain scans were repeated if clinically indicated at tumor assessment timepoints. PD-L1–expression status was determined using tissue samples collected at screening. PD-L1⁺ was defined as a CPS of ≥1 using an investigational version of the PD-L1 assay [IHC 22C3 pharmDx (Agilent); ref. 22].

The study planned an initial run-in cohort of ≥6 patients in phase Ib and then a total sample size of approximately 170 patients, including 100 in stratum 2. With 100 patients, the 95% CI for ORR would be 16.9%–34.7% for an observed ORR of 25% based on the binomial distribution. Analyses were performed after all ongoing patients had completed ≥24 weeks of treatment or discontinued from treatment.

DLTs were assessed in the DLT evaluable set (patients who completed the first treatment cycle and had a safety evaluation in phase Ib). Safety, PFS, and OS were assessed in the full analysis set (all patients who received study drugs). Efficacy analyses on ORR and CBR were performed in the evaluable analysis set (all patients, including those in phase Ib who received the phase II dose, with evaluable tumor assessments at baseline and postbaseline, unless discontinued early or died).

The Clopper–Pearson method was used to calculate 95% CIs for ORR and CBR. Median PFS, OS, and DOR were estimated using Kaplan–Meier product limit method, and their corresponding 95% CIs were calculated using a generalized Brookmeyer and Crowley method. Statistical analyses were performed using SAS software version 9.4 (SAS Institute).

All experiments comply with the current laws of the country in which they were performed.

Informed consent was obtained from all individual participants included in the study.

All procedures performed in studies involving human participants were in accordance with the ethical standards of the Institutional Review Board (IRB; Dana-Farber Cancer Institute IRB, Organization number: IORG0000035IRB; federal-wide assurance number: FWA00001121) and/or National Research Committee, and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.

This article does not contain any studies with animals performed by any of the authors.

The data will not be available for sharing at this time as the data are commercially confidential. However, Eisai will consider written requests to share the data on a case-by-case basis.

Between September 21, 2015 and July 31, 2019 (data cut-off date), 167 patients were enrolled and treated (phase Ib, n = 7; phase II, n = 160). At the data cut-off date, 158 patients had discontinued and 9 remained on treatment. As noted in Table 1, 44% of patients had PD-L1⁺ tumors, 45% had PD-L1–negative (PD-L1⁻) tumors, and 11% had unknown tumor PD-L1 status. Overall, 66 (40%) patients were enrolled into stratum 1 (no prior systemic anticancer therapies for metastatic disease) and 101 (60%) patients were enrolled into stratum 2 (1–2 prior systemic anticancer therapies for metastatic disease; Table 1).

None of the 7 patients in the phase Ib part of the study experienced a DLT. As such, the recommended phase II dose was eribulin 1.4 mg/m² on days 1 and 8 plus pembrolizumab 200 mg on day 1 of each 21-day cycle.

The five most common TEAEs (all grades) were fatigue (66%), nausea (58%), peripheral sensory neuropathy (41%), alopecia (40%), and constipation (37%; Table 2). The only grade 3 or higher TEAE occurring in >10% of patients was neutropenia (26%; Table 2). Overall, 21% of patients in this study received growth factor support (filgrastim and/or pegfilgrastim). TEAEs led to study-drug discontinuation of eribulin, pembrolizumab, or both in 6%, 9%, and 11% of patients, respectively. Life-threatening treatment-related TEAEs were experienced by 2 patients (hemophagocytic lymphohistiocytosis and pyrexia, n = 1; neutropenia, n = 1). Overall, 15 patients experienced a fatal TEAE but none of these deaths were considered treatment related (Supplementary Table S1).

TEAEs of clinical interest for pembrolizumab occurred in 43% of patients (Supplementary Table S2). The most commonly reported among these were hypothyroidism (18%), pneumonitis (11%), hyperthyroidism (8%), and infusion-related reaction (3%; Supplementary Table S2). Immune-related TEAEs were of grade 3–4 severity in a minority of patients (12%; Supplementary Table S2).

Treatment with eribulin plus pembrolizumab resulted in an ORR of 23.4% (95% CI: 17.2–30.5; Table 3). CR and PR were observed in 8 (4.8%) and 31 (18.6%) patients, respectively. In addition, 53 patients (31.7%) had SD lasting ≥8 weeks (Table 3). The median DOR (in patients with CR or PR) was 8.3 months (95% CI: 6.5–22.2; Table 3). Subgroup analyses by prior lines of therapy indicated a numerically higher ORR among patients in stratum 1 who had no prior systemic therapy compared with those patients in stratum 2 who had 1–2 prior lines [25.8% (95% CI: 15.8–38.0) vs 21.8% (95% CI: 14.2–31.1), respectively; Table 3]. The overall median PFS was 4.1 months (95% CI: 3.5–4.2), which was similar across strata (Table 3). The overall median OS was 16.1 months (95% CI: 13.3–18.5), with a median OS of 17.4 months (95% CI: 13.2–21.0) for stratum 1 versus 15.5 months (95% CI: 12.5–18.7) for stratum 2 (Table 3).

Subgroup analysis by PD-L1 tumor expression status indicated better efficacy outcomes in patients with PD-L1⁺ tumors versus those with PD-L1⁻ tumors [ORR, 28.4% (95% CI: 18.5–40.1) vs 17.3% (95% CI: 9.6–27.8); PFS, 4.2 months (95% CI: 3.5–6.1) vs 3.9 months (95% CI: 2.3–4.4); and OS, 16.3 months (95% CI: 12.7–24.2) vs 15.2 months (95% CI: 13.3–18.0), respectively]. Most patients showed a decrease in target lesion size regardless of PD-L1–expression status (Fig. 1A and B).

Efficacy outcomes stratified by both stratum and PD-L1–expression status are shown in Table 4. Among patients in stratum 1, the median PFS was higher for patients who had PD-L1⁺ tumors compared with those who had PD-L1⁻ tumors [6.1 months (95% CI: 4.1–10.2) versus 3.5 months (95% CI: 2.0–4.2), respectively; Fig. 2A]. In contrast, patients in stratum 2 had similar median PFS results, regardless of PD-L1–expression status (Fig. 2B). Patients with PD-L1⁺ tumors in stratum 1 had a median OS that was numerically longer than the median OS among those with PD-L1⁻ tumors [21.0 months (95% CI: 8.3–29.0) vs 15.2 months (95% CI: 12.8–19.4), respectively; Fig. 2C]; however, a trend toward longer OS was not observed in patients with PD-L1⁺ tumors in stratum 2 (Fig. 2D).

To further assess PD-L1 tumor expression status as an indicator of greater anticancer activity of eribulin when combined with pembrolizumab, we compared efficacy outcomes for patients with PD-L1⁺ tumors based on an IHC-staining CPS cut-off of either 1 (CPS ≥ 1) or 10 (CPS ≥ 10). In the 1L setting (stratum 1), the ORR for patients with PD-L1⁺ tumors was higher than for those with PD-L1⁻ tumors, regardless of whether the CPS cut-off was 1 or 10 (CPS ≥ 1, ORRs 34.5% vs 16.1%; CPS ≥ 10, ORRs 30.8% vs 23.4%, respectively; Supplementary Table S3A). In the 2–3L setting (stratum 2), the ORR for patients with PD-L1⁺ tumors was also numerically higher versus those with PD-L1⁻ tumors by both CPS cut-offs (CPS ≥ 1, ORR 24.4% vs 18.2%; CPS ≥ 10, ORR 23.8% vs 20.6%, respectively; Supplementary Table S3B).

In this analysis of an ongoing open-label, single-arm, phase Ib/II study, the combination of eribulin plus pembrolizumab was generally well tolerated and showed encouraging antitumor activity as 1–3L therapy for patients with mTNBC, resulting in an ORR of 23.4%. The safety profile of eribulin plus pembrolizumab was similar to what was seen in previous monotherapy studies (16, 23, 24). Neutropenia was the only TEAE of grade ≥3 occurring in >10% of patients. The rate of neutropenia observed in this study (all grades, 36%; grade 3/4, 26%) compared favorably to that observed in the pivotal EMBRACE study (all grades, 52%; grade 3/4, 45%; ref. 23).

Acknowledging the limitations of cross-study comparisons, in this study, patients with mTNBC who received eribulin plus pembrolizumab in the 1L setting (stratum 1) achieved an ORR of 26% (95% CI: 16–38) versus an ORR of 21% (95% CI: 14–31) for patients treated with pembrolizumab monotherapy (KEYNOTE-086 cohort B; ref. 17) and an ORR of 10% for patients receiving eribulin monotherapy (2). Moreover, patients treated in the 2–3L setting (stratum 2) in this study had an ORR of 22% (95% CI: 14–31), which compares favorably with the ORRs of 5% to 12% observed in the 2L+ for pembrolizumab monotherapy [(KEYNOTE-086 cohort A; ref. 8); (KEYNOTE-119; ref. 25)]. Although no ORR for eribulin monotherapy in patients with mTNBC has been reported for the 2L+ setting, the median PFS and median OS observed in this study for patients with mTNBC who were treated with eribulin plus pembrolizumab compared favorably with historical reports for eribulin monotherapy [1–2L+, PFS: 4.1 months in this study vs 2.8 months (15); OS: 16.1 months in this study vs 12.9 months (15)]. Similarly, higher median PFS and OS were observed in patients with PD-L1⁺ mTNBC in this study versus pembrolizumab monotherapy [KEYNOTE-012; 1–2L+, PFS: 4.2 vs 1.9 months, OS: 16.3 vs 11.2 months, respectively (16)]. These results suggest that eribulin plus pembrolizumab has promising antitumor activity in patients with mTNBC and that this activity appears greater than in historical reports of either agent alone.

Limitations of this study are typical of an early-phase study: this is a nonrandomized, single-arm clinical trial, and the patient population in this study was rather heterogeneous with regard to the number of prior systemic therapies (0–2) and tumor PD-L1–expression status. The lack of a direct comparator arm with immunotherapy or eribulin alone limits interpretation, especially as it has been suggested that immunotherapy may stabilize disease but later impact OS. In addition, this study was not powered to assess efficacy based on PD-L1 status. Further study is warranted to delineate which patients with mTNBC might derive the most benefit from this treatment combination.

A subgroup analysis of patients with HER2-negative MBC treated with eribulin, showed that receiving eribulin in the earlier setting (1L) resulted in higher ORRs compared with later-line (2–3L) therapy (26% vs 8%, respectively; ref. 26). Interestingly, while the ORR in this study was higher in the earlier-line setting (stratum 1, 26%), much of the antitumor benefit conferred by eribulin plus pembrolizumab was still retained in the later-line setting (stratum 2, 22%). Moreover, previous studies indicated that patients who had PD-L1⁺ mTNBC and who were treated with pembrolizumab monotherapy had an ORR of 21% for 1L versus 5% in 2L+ use (8, 17). In this study, patients with PD-L1⁺ mTNBC achieved an ORR of 34.5% in 1L (stratum 1) compared with 24.4% in the 2–3L setting (stratum 2), which further demonstrates that this combination therapy of eribulin plus pembrolizumab may have promising antitumor efficacy even in the later-line settings.

Consistent with results from a previous study (9), patients with mTNBC, no prior systemic anticancer therapy, and PD-L1⁺ tumors in this study had the most robust response to the combination of immunotherapy and chemotherapy (median OS, 21 months). On the basis of both CPS cut-offs (≥1 and ≥10), the ORR was higher in patients with PD-L1⁺ tumors versus those with PD-L1⁻ tumors in both stratum 1 and stratum 2; the benefit appears to be more pronounced for patients with PD-L1⁺ tumors in the 1L setting (stratum 1). Interestingly, among patients with evaluable PD-L1 status, based on a cut-off CPS of ≥1, all 6 patients who had a CR had PD-L1⁺ tumors, whereas only 2 of 6 patients with a CR had PD-L1⁺ tumors based on a cut-off CPS of ≥10 (Supplementary Table S3). While ORR was higher among the PD-L1⁺ cohort, how we optimally select patients who are likely to benefit from eribulin plus pembrolizumab therapy requires further investigation.

Atezolizumab in combination with paclitaxel protein bound was approved by the FDA for patients with mTNBC and PD-L1⁺ tumors (defined as PD-L1–stained tumor-infiltrating immune cells of any intensity covering ≥1% of the tumor area; ref. 27). In addition, the FDA recently approved pembrolizumab in combination with chemotherapy (i.e., paclitaxel, paclitaxel protein bound, or gemcitabine plus carboplatin) for the treatment of patients with mTNBC whose tumors express PD-L1 (CPS ≥ 10; ref. 5). Given these recent FDA approvals, many patients with mTNBC and PD-L1⁺ may be treated with immunotherapy plus chemotherapy in the 1L setting. It should be noted that those patients with PD-L1⁺ tumors in this study derived a clinical benefit in the 2L+ setting (ORR, 24.4%). Importantly, those patients with PD-L1⁻ tumors in this study also derived benefit from the combination of eribulin plus pembrolizumab in both the 1L and 2L+ settings (ORRs, 16.1%–18.2%). Thus, while the immediate clinical utility of the study-drug combination requires further investigation, the combination of eribulin plus pembrolizumab may be a potential treatment option for those patients with mTNBC and PD-L1⁺ tumors who are either treatment naïve or have experienced disease progression on a previous line of therapy, and/or those patients with mTNBC who have PD-L1⁻ tumors.

Eribulin has well-documented efficacy in the 2L+ setting for patients with MBC, including mTNBC (13–15). However, a recent clinical study of patients with pretreated (2–3L) mTNBC (KEYNOTE-119) failed to prove superiority of pembrolizumab versus standard chemotherapy regimens (including eribulin; ref. 25). In the current study (ENHANCE 1), eribulin plus pembrolizumab shows promising antitumor activity for patients with mTNBC in the 2–3L setting. However, caution is warranted in interpreting these data as it remains to be determined whether the addition of immunotherapy to standard chemotherapy in the 2–3L setting confers a greater benefit for patients with mTNBC than chemotherapy alone. Nonetheless, the median OS of 15.5 months and median PFS of 4.1 months observed in this study for patients with mTNBC and treated with eribulin plus pembrolizumab in the 2–3L setting is encouraging and warrants further investigation.

Eribulin plus pembrolizumab was generally well tolerated with a predictable safety profile in patients with mTNBC. The study-drug combination showed promising antitumor activity in the 1–3L setting. While the activity appeared enhanced in patients with PD-L1⁺ tumors, the study-drug combination also demonstrated promising antitumor activity in patients with PD-L1⁻ tumors. These results support further clinical development of eribulin plus pembrolizumab as a potential antitumor strategy for patients with mTNBC.

S.M. Tolaney reports grants from Eisai during the conduct of the study. S.M. Tolaney also reports grants and personal fees from AstraZeneca, Eli Lilly, Merck, Novartis, Nektar, Pfizer, Genentech/Roche, Immunomedics/Gilead, BMS, Eisai, Nanostring, Sanofi, Odonate, and Seattle Genetics; grants from Exelixis and Cyclacel; and personal fees from Puma Biotechnology, Daiichi Sankyo, Silverback Therapeutics, G1 Therapeutics, Athenex, OncoPep, Kyowa Kirin Pharmaceuticals, Samsung Bioepsis Inc., CytomX, Mersana Therapeutics, and Certara outside the submitted work. K. Kalinsky reports personal fees from Eisai during the conduct of the study. K. Kalinsky also reports personal fees from Eli Lilly, Pfizer, Novartis, Eisai, AstraZeneca, Immunomedics, Merck, Seattle Genetics, and Cyclocel; other from Grail, Array BioPharma, and Pfizer; and personal fees from Eli Lilly outside the submitted work. V.G. Kaklamani reports grants from Eisai, as well as personal fees from Novartis, Pfizer, Seagen, Daiichi Sankyo, Immunomedics, and AstraZeneca outside the submitted work. D.R. D'Adamo reports other from Eisaai during the conduct of the study. G. Aktan reports personal fees and other from Merck (employee) during the conduct of the study. G. Aktan also reports personal fees and other from Merck outside the submitted work, and employment of his son, Burak Aktan, with Merck. R.M. O'Regan reports grants from University of Wisconsin during the conduct of the study, as well as personal fees from Puma Biotechnology, Novartis, Eli Lilly, Biotheranostics, Macrogenics, Pfizer, and Seagen outside the submitted work. P.A. Kaufman reports grants from Eisai and Merck during the conduct of the study. P.A. Kaufman also reports personal fees from Eisai and Roche-Genentech; grants and personal fees from Macrogenics, Eli Lilly, and Polyphor; grants from Sanofi and Pfizer; and other from Amgen outside the submitted work. Y. Yuan reports grants and other from Eisai and Merck during the conduct of the study. Y. Yuan also reports grants and other from Novartis, Genentech, and Pfizer; other from Immunomedics, AstraZeneca, and Daiichi Sankyo; and grants from Puma Biotechnology outside the submitted work. G. Wang reports grants from Eisai, Daiichi Sankyo, Seattle Genetics, Pharmacyclics, Eli Lilly, Immunomedics, OBI Pharma, Sermonix, Odonate Therapeutics, and MSK/Puma Biotechnology during the conduct of the study. V. Karantza reports personal fees from Merck outside the submitted work, and owns MRK stocks. S. Diab reports other from Gilead, Karyopharm, G1 Therapeutics, Bluebird, and Mustang Bio outside the submitted work; S. Diab also reports consulting and speaking fees for Pfizer, Novartis, Eli Lilly, Genentech, Clovis Oncology, AstraZeneca, Seattle Genetics, and Puma Biotechnology. No disclosures were reported by the other authors.

S.M. Tolaney: Conceptualization, formal analysis, investigation, writing–original draft, writing–review and editing. K. Kalinsky: Conceptualization, formal analysis, investigation, writing–review and editing. V.G. Kaklamani: Conceptualization, formal analysis, investigation, writing–review and editing. D.R. D'Adamo: Conceptualization, formal analysis, investigation, writing–review and editing. G. Aktan: Conceptualization, formal analysis, investigation, writing–review and editing. M.L. Tsai: Conceptualization, formal analysis, investigation, writing–review and editing. R.M. O'Regan: Conceptualization, formal analysis, investigation, writing–review and editing. P.A. Kaufman: Conceptualization, formal analysis, investigation, writing–review and editing. S.T. Wilks: Conceptualization, formal analysis, investigation, writing–review and editing. E. Andreopoulou: Conceptualization, formal analysis, investigation, writing–review and editing. D.A. Patt: Conceptualization, formal analysis, investigation, writing–review and editing. Y. Yuan: Conceptualization, formal analysis, investigation, writing–review and editing. G. Wang: Conceptualization, formal analysis, investigation, writing–review and editing. C. Savulsky: Conceptualization, formal analysis, investigation, writing–review and editing. D. Xing: Conceptualization, data curation, software, formal analysis, validation, investigation, methodology, writing–review and editing. E. Kleynerman: Conceptualization, formal analysis, investigation, writing–review and editing. V. Karantza: Conceptualization, formal analysis, investigation, writing–review and editing. S. Diab: Conceptualization, formal analysis, investigation, writing–review and editing.

This work was supported by Eisai Inc., Woodcliff Lake, NJ, and Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ.

Medical writing support was provided by Tarah M. Connolly, PhD, of Oxford PharmaGenesis Inc., Newtown, PA, with funding provided by Eisai Inc., Woodcliff Lake, NJ, and Merck Sharp & Dohme Corp., a subsidiary of Merck & Co., Inc., Kenilworth, NJ.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.