Clinical Significance of PIK3CA and ESR1 Mutations in Circulating Tumor DNA: Analysis from the MONARCH 2 Study of Abemaciclib plus Fulvestrant

This article is featured in Highlights of This Issue, p. 1475

Breast cancer is the most common cancer among women worldwide and the leading cause of cancer death in women (1). Of those diagnosed with breast cancer, hormone receptor–positive (HR+), HER2-negative (HER2−) is the most common subtype, accounting for nearly 70% of all metastatic breast cancer (2). Within this population, resistance to endocrine therapy (ET) is common, and most patients ultimately succumb to disease, leading to interest in development of more precise therapeutic approaches (3, 4).

Endocrine resistance evolves from many mechanisms, including genetic dysregulation, posttranslational modifications, and altered cell signaling promoting ligand-independent activation of the estrogen receptor (ER) and decreased sensitivity to antiestrogens (5–7). A major mechanism of resistance is mutation of the ERα gene, ESR1, frequently observed after treatment with aromatase inhibitors (8, 9). Most somatic ESR1 mutations occur at either D538 or Y537, within the ESR1 ligand-binding domain. The clinical implications of these frequently observed mutations, with regard to fulvestrant response, remain unclear (6, 9, 10).

In addition to ESR1-mediated endocrine resistance, studies have also identified a prominent role for PI3K pathway dysregulation in metastatic breast cancer progression and endocrine resistance (4, 11). PIK3CA, which encodes phosphatidylinositol-4, 5-bisphosphate 3-kinase catalytic subunit α, is mutated in approximately 40% of HR+, HER2− advanced breast cancers (ABC), resulting in a gain-of-function phenotype and increased downstream signaling and oncogenesis (4, 12). Most mutations occur at E542K or E545K of exon 9 (helical domain of p110α), and H1047R/L of exon 20 (activation loop in the kinase domain), representing “hotspot” mutations (4). Prior clinical studies have demonstrated HR+ breast cancers with mutations in either exon 9 or exon 20 exhibited decreased response to ET (4). To address this clinically, several targeted therapies have been developed to inhibit the PI3K pathway (4). Efficacy of the PI3Kα inhibitor alpelisib in an endocrine-resistant population with HR+, HER2− ABC was demonstrated in the phase III SOLAR-1 study, demonstrating improved median PFS in combination with fulvestrant in patients with PIK3CA-mutant tumors, although no statistical benefit in OS was observed (13).

CDK4 and 6 have also been identified as actionable therapeutic targets in the setting of endocrine resistance (4, 14). Abemaciclib is an oral, potent, and selective CDK4 and 6 inhibitor, dosed on a continuous schedule and approved as monotherapy and in combination with either fulvestrant or an aromatase inhibitor for HR+, HER2− ABC (3). MONARCH 2 was a phase III randomized, double-blind, placebo-controlled study that evaluated the efficacy of abemaciclib in combination with fulvestrant in patients with HR+, HER2− ABC whose disease had progressed on ET. This trial demonstrated improved PFS (median 16.4 months vs. 9.3 months; HR, 0.553; 95% CI, 0.449–0.681; P < 0.001) and overall survival (OS; median 46.7 months vs. 37.3 months; HR, 0.757; 95% CI, 0.606–0.945; P = 0.01) with a tolerable safety profile (3, 15).

Here we report the results of an exploratory analysis to assess the PFS, OS, time to chemotherapy (TTC), chemotherapy-free survival (CFS), and time to second disease progression (PFS2) in patients with PIK3CA- or ESR1-wild-type or -mutant tumors who received abemaciclib plus fulvestrant or placebo plus fulvestrant in the MONARCH 2 study.

MONARCH 2 was a randomized, double-blind, placebo-controlled, phase III study of abemaciclib plus fulvestrant or placebo plus fulvestrant in women with HR+, HER2− ABC who had disease progression on ET.

Eligible patients were adult females (≥18 years) with any menopausal status (pre- or perimenopausal women received a gonadotropin-releasing hormone agonist), diagnosed with HR+, HER2− ABC and disease progression while receiving neoadjuvant or adjuvant ET, within 12 months after adjuvant ET, or while receiving first-line ET for ABC. Patient performance status must have been ≤1 on the Eastern Cooperative Oncology Group (ECOG) scale and patients must have had measurable disease defined according to the RECIST (version 1.1) or nonmeasurable bone-only disease (i.e., blastic, lytic, or mixed lytic). Patients must not have had prior treatment with more than one line of ET or any prior chemotherapy for ABC. Exclusion criteria also included prior treatment with fulvestrant, everolimus, or any CDK4 and CDK6 inhibitors, presence of visceral crisis, or evidence or history of central nervous system metastasis.

The protocol was approved by ethical and institutional review boards and the study was conducted in compliance with the Declaration of Helsinki. All patients provided written informed consent prior to joining the study.

Dosing was as previously described (3, 15). Briefly, 669 patients [intent-to-treat (ITT) population] were randomized 2:1 to receive either abemaciclib (150 mg) or placebo twice daily on days 1 to 28 of a 28-day cycle plus fulvestrant (500 mg intramuscularly) on days 1 and 15 of cycle 1, and on day 1 of each subsequent cycle. Treatment continued until disease progression, death, or patient withdrawal (15). Randomization was stratified by metastatic site (visceral, bone-only, or other) and sensitivity to ET (primary or secondary resistance).

Efficacy and safety assessments were conducted as described previously (3, 15). Droplet digital PCR (ddPCR) was used to evaluate four mutations each in PIK3CA (E542K, E545K, H1047L, and H1047R) and ESR1 (D538G and Y537C/S/N) via the Bio-Rad QX200 Droplet Digital PCR system (Asuragen) in baseline plasma samples. Samples were considered positive if at least one mutation tested positive and defined the translational research (TR) population. End-of-treatment samples were not collected.

PFS, the primary endpoint, was measured from the time of randomization to the date of objective progression or death. Additional endpoints included OS (time from randomization until death), TTC (time from randomization to first postdiscontinuation chemotherapy, death prior to initiation was treated as censored), CFS (time from randomization to initiation of first postdiscontinuation chemotherapy or death), and PFS2 (time from randomization to the discontinuation of first subsequent postdiscontinuation therapy or death).

Analyses were conducted on those patients with valid baseline samples. PFS, OS, TTC, CFS, and PFS2 were analyzed with a data cut-off of June 20, 2019. The median PFS, OS, TTC, CFS, and PFS2 estimated using the Kaplan–Meier method are reported along with their 95% confidence intervals (CI). HR and 95% CI were derived from the unstratified Cox regression model.

Lilly provides access to all individual participant data collected during the trial, after anonymization, with the exception of pharmacokinetic or genetic data. Data are available to request 6 months after the indication studied has been approved in the United States and European Union and after primary publication acceptance, whichever is later. No expiration date of data requests is currently set once data are made available. Access is provided after a proposal has been approved by an independent review committee identified for this purpose and after receipt of a signed data sharing agreement. Data and documents, including the study protocol, statistical analysis plan, clinical study report, blank or annotated case report forms, will be provided in a secure data sharing environment. For details on submitting a request, see the instructions provided at www.vivli.org.

From August 2014 to December 2015, 669 patients were randomized 2:1 to abemaciclib plus fulvestrant (N = 446) or placebo plus fulvestrant (N = 223; Supplementary Fig. S1; ref. 3). Baseline plasma from 334 patients was analyzed for PIK3CA mutations in ctDNA by ddPCR, with 219 of 334 (65.6%) samples successfully analyzed and defined as the PIK3CA translational research (TR) population. In this population, PIK3CA mutations were detected in 96 (43.8%) of 219 patients (26.5% abemaciclib arm, 17.4% placebo arm). The most frequently observed mutation was H1047R (33.3%), followed by E545K (6.1%), E542K (4.3%), and H1047 L (1.4%; Supplementary Table S1). Eight of 219 patients (3.7%) had both H1047R/L and E545K/E542K mutations. At baseline, 248 of 334 (74.3%) samples were successfully analyzed and defined as the ESR1 TR population. ESR1 mutations were detected in 147 (59.3%) of 248 patients (36.7% abemaciclib arm, 22.6% placebo arm). The most frequently observed mutation was D538G (45.1%), followed by Y537C (15.8%), Y537S (7.8%), and Y537N (7.5%; Supplementary Table S1). Forty-four of 248 patients (17.7%) had both D538G and Y537C/S/N mutations. Seventy-one patients had both PIK3CA and ESR1 mutations. Baseline characteristics were generally similar between the TR population and the rest of the ITT population, as well as between PIK3CA- and ESR1-mutant and wild-type subgroups (Table 1). A slight increase in the percentage of patients with visceral disease was observed in the PIK3CA-mutant compared with PIK3CA-wild-type subgroup, whereas a slightly higher percentage of bone-only disease was observed in the wild-type subgroup, although differences in metastatic sites were not statistically significant. A higher prevalence of patients with secondary endocrine resistance was observed in the ESR1-mutant subgroup, whereas primary endocrine resistance was more prevalent in the ESR1-wild-type subgroup. Progesterone receptor–positive tumors were more common in the ESR1-mutant subgroup compared with ESR1-wild-type.

Median PFS in the ITT population was 16.9 months compared with 17.1 months in the PIK3CA TR population and 17.2 months in the ESR1 TR population in patients treated with abemaciclib plus fulvestrant (Supplementary Fig. S2). Overall, PFS was improved in the abemaciclib arm compared with placebo, regardless of PIK3CA or ESR1 mutation status, consistent with results in the ITT population.

Patients with and without PIK3CA mutations in the abemaciclib arm demonstrated improved PFS compared with the placebo arm (wild-type: median 16.9 months vs. 12.3 months; HR, 0.51; 95% CI 0.33–0.78; mutant: median 17.1 months vs. 5.7 months; HR, 0.53; 95% CI, 0.33–0.84; Fig. 1). In addition, irrespective of the specific PIK3CA mutation (E545K/E542K or H1047R/H1047L), patients in the abemaciclib arm demonstrated numerically improved PFS compared with the placebo arm (E545K/E542K: median 17.1 months vs. 13.9 months; HR, 0.60; 95% CI, 0.24–1.52; H1047R/H1047L: median 17.2 months vs. 4.8 months; HR, 0.50; 95% CI, 0.29–0.83; Supplementary Fig. S3). For patients who received placebo plus fulvestrant, median PFS was shorter in the PIK3CA-mutant population (5.7 months) compared with patients without PIK3CA mutations (12.3 months). The numerical improvement in median PFS with abemaciclib plus fulvestrant was greater in patients with PIK3CA-mutant tumors (17.1 months vs. 5.7 months), compared with wild-type (16.9 months vs. 12.3 months).

Similarly, regardless of ESR1 mutation status, patients in the abemaciclib arm showed improved PFS compared with the placebo arm (wild-type: median 15.3 months vs. 11.2 months; HR, 0.44; 95% CI, 0.27–0.71; mutant: median 20.7 months vs. 13.1 months; HR, 0.54, 95% CI, 5.7–16.9; Fig. 2) and PFS was extended regardless of which ESR1 mutation (D538G or Y537C/S/N) was present (D538G: median 26.0 months vs. 13.1 months; HR, 0.45; 95% CI, 0.29–0.69; Y537: median 16.6 months vs. 4.8 months; HR, 0.48; 95% CI, 0.28–0.83; Supplementary Fig. S4). For patients harboring D538G-mutant tumors who received placebo + fulvestrant, median PFS was similar to wild-type tumors (13.1 months vs. 11.2 months). In contrast, median PFS was shorter in patients harboring Y537-mutant tumors compared with wild-type who were treated with placebo + fulvestrant (4.8 months vs. 11.2 months; Supplementary Fig. S4). In addition, the numerical improvement in median PFS with abemaciclib plus fulvestrant was greater in patients with ESR1-mutant tumors (20.7 months vs. 13.1 months) compared with wild-type (15.3 months vs. 11.2 months).

Median OS in the ITT population was 46.7 months compared with 47.7 months in the PIK3CA TR population in patients treated with abemaciclib plus fulvestrant and was not reached in the ESR1 TR population. Overall, abemaciclib plus fulvestrant prolonged median OS compared with the placebo arm regardless of PIK3CA or ESR1 mutation status (Supplementary Fig. S5). Median OS was numerically extended in the abemaciclib arm versus the placebo arm in patients with PIK3CA-mutant tumors (44.5 months vs. 33.8 months) and in the PIK3CA-wild-type subgroup (55.5 months vs. 39.7 months) tumors (Fig. 3). Likewise, in patients harboring either the E545K/E542K or H1047R/L mutations median OS was numerically prolonged in the abemaciclib arm compared with the placebo arm (E545K/E542K: 48.8 months vs. 34.8 months; H1047R/H1047L: 44.5 months vs. 33.8 months; Supplementary Fig. S6).

Median OS was also numerically extended in the abemaciclib arm versus the placebo arm in patients with either ESR1-mutant (not reached vs. 42.2 months) or ESR1-wild-type tumors (52.2 months vs. 29.4 months; Fig. 4). Interestingly, in this analysis, median OS was longer in patients with ESR1-mutant versus ESR1-wild-type tumors receiving placebo plus fulvestrant (42.2 months vs. 29.4 months; Fig. 4). This was particularly pronounced for the D538G-mutant subgroup compared with wild-type ESR1 (Supplementary Fig. S7).

At the time of the data cut-off, all patients in the abemaciclib arm and the placebo arm of the TR population had received a first subsequent postdiscontinuation therapy. Therapies were generally well balanced across PIK3CA- and ESR1-mutant vs. wild-type subgroups and across treatment arms. Use of chemotherapy as first subsequent postdiscontinuation therapy was higher in the placebo arm (PIK3CA: 37.2%; ESR1: 38.5%) compared with the abemaciclib arm (PIK3CA: 25.6%; ESR1: 28.0%). ET was more common in both the PIK3CA- and ESR1-wild-type subgroups of the placebo arm compared with the mutant subgroups (Supplementary Tables S2 and S3).

TTC was extended in patients treated with abemaciclib plus fulvestrant compared with placebo plus fulvestrant regardless of PIK3CA mutation status (wild-type: median not reached vs. 18.6 months; HR, 0.47; 95% CI, 0.29–0.78; mutant: median 39.2 months vs. 19.2 months; HR, 0.64; 95% CI, 0.36–1.15; Supplementary Fig. S8A). It should be noted that in the TTC analysis, patient death was not counted as an event.

TTC was also extended in patients treated with abemaciclib plus fulvestrant compared with placebo plus fulvestrant regardless of ESR1 mutation status (wild-type: median 50.2 months vs. 12.8 months; HR, 0.45; 95% CI, 0.26–0.78; mutant: median 51.5 months vs. 26.8 months; HR, 0.65; 95% CI, 0.41–1.03; Supplementary Fig. S8B). CFS was likewise improved in patients treated with abemaciclib plus fulvestrant compared with placebo plus fulvestrant, regardless of PIK3CA or ESR1 mutation status (Supplementary Fig. S9). Interestingly, TTC and CFS were prolonged in the placebo arm for patients with ESR1-mutant tumors compared with wild-type.

PFS2 was extended in patients receiving abemaciclib compared with placebo regardless of PIK3CA or ESR1 mutation status. PFS2 in PIK3CA-wild-type patients was prolonged in the abemaciclib arm compared with the placebo arm (median 31.7 months vs. 19.7 months; HR, 0.52; 95% CI, 0.34–0.80). In the PIK3CA-mutant population, PFS2 was also numerically extended in the abemaciclib arm versus the placebo arm (median 23.1 months vs. 17.8 months; HR, 0.72; 95% CI, 0.45–1.16; Supplementary Fig. S10A). PFS2 was extended in patients with wild-type ESR1 treated with abemaciclib plus fulvestrant compared with placebo plus fulvestrant, (median: 25.5 months vs. 16.6 months; HR, 0.44; 95% CI, 0.27–0.70), as well as in patients with mutant ESR1, with a late separation of the curves after 30 months (26.6 months vs. 26.5 months; HR, 0.66; 95% CI, 0.44–0.97; Supplementary Fig. S10B).

This exploratory analysis from MONARCH 2 demonstrated that abemaciclib extended PFS and OS when added to fulvestrant in women with HR+, HER2− ABC that had progressed on ET, regardless of PIK3CA or ESR1 mutation status in ctDNA. In addition, other exploratory endpoints, TTC, CFS, and PFS2, were also improved with abemaciclib plus fulvestrant.

The MONARCH 2 patient population, HR+, HER2− ABC, represents a large component of ABC clinically, and, importantly, included patients with disease progression on ET (3). The MONARCH 2 study demonstrated that patients treated with abemaciclib plus fulvestrant consistently derived benefit with regards to PFS and OS compared with placebo plus fulvestrant across both the ITT population and clinically relevant subgroups, including patients with primary ET resistance (3).

Previous studies have highlighted a role for mutations in PIK3CA as a mediator of endocrine resistance and association with worse clinical outcome (4, 12, 13). Consistent with this hypothesis, we report shorter median PFS (5.7 months vs. 12.3 months) and OS (33.8 months vs. 39.7 months) in patients with mutant PIK3CA compared with wild-type when treated with placebo plus fulvestrant.

Somatic mutations in the PI3K pathway occur in more than 70% of breast cancers, with the majority occurring in PIK3CA (nearly 40% of HR+, HER2− MBC; refs. 4, 12). Specifically targeting PIK3CA has resulted in improved patient benefit with manageable safety profile, as demonstrated in the SOLAR-1 clinical study of alpelisib plus fulvestrant in patients with HR+, HER2− ABC that had progressed on previous ET. SOLAR-1 reported statistically improved PFS in patients harboring PIK3CA-mutant tumors when treated with alpelisib plus fulvestrant compared with placebo plus fulvestrant (median 11.0 months vs. 5.7 months; ref. 13), although statistically improved OS was not observed (16). Here we report median PFS of 5.7 months in patients harboring PIK3CA-mutant tumors treated with placebo plus fulvestrant, whereas the addition of abemaciclib to fulvestrant improved median PFS to 17.1 months. Similarly, median OS was 33.8 months in patients harboring PIK3CA-mutant tumors treated in the placebo arm, and treatment with abemaciclib plus fulvestrant improved median OS to 44.5 months.

Interestingly, although we observed little change in PFS in tumors harboring ESR1 mutations when treated with placebo plus fulvestrant, compared with wild-type, OS was unexpectedly prolonged in the ESR1-mutant subgroup compared with wild-type. Previous studies have demonstrated minimal effect of ESR1 mutation status on the PFS of patients receiving fulvestrant (17, 18), whereas OS tended to be shorter in ESR1-mutant subgroups (19, 20). The reason for this difference, as well as the increased prevalence of ESR1 mutations observed in this analysis, is not clear, but may be due to the sample size, patient population (inclusion criteria and endocrine-resistant population evaluated in this study), or methodology used (including differences in assay used and sensitivity).

Prior preclinical studies characterizing ESR1 mutants demonstrated that D538G continued to respond to single-agent fulvestrant therapy, with only modest differences compared to wildtype (6, 8–10). In contrast, cells harboring the Y537S mutation were less responsive to single-agent fulvestrant therapy (6). In our study, analysis of specific ESR1 mutations following fulvestrant treatment (placebo arm) demonstrated that the median PFS was similar between the ESR1-wild-type and D538G mutant subgroups, and shorter for the Y537 subgroup, consistent with these prior studies indicating relative resistance to fulvestrant (6, 9). Importantly, regardless of ESR1 mutation status or which ESR1 mutation was present, tumors treated with abemaciclib plus fulvestrant demonstrated improved PFS and OS compared with placebo plus fulvestrant. In agreement with our observations, the PALOMA-3 clinical trial reported improved OS in patients treated with palbociclib plus fulvestrant compared with placebo plus fulvestrant, regardless of ESR1 (mutant: 27.7 months vs. 20.2 months; wild-type: 32.8 months vs. 28.0 months) and PIK3CA (mutant: 27.7 months vs. 18.3 months; wild-type: 32.8 months vs. 26.6 months) mutation status (21).

In conclusion, abemaciclib plus fulvestrant improved both PFS and OS, regardless of PIK3CA or ESR1 mutation status. The numerical improvement in median PFS relative to placebo plus fulvestrant was greater in patients with either PIK3CA-mutant tumors (17.1 months vs. 5.7 months), compared with wild-type (16.9 months vs. 12.3 months) or ESR1-mutant tumors (20.7 months vs. 13.1 months) compared with wild-type (15.3 months vs. 11.2 months). Similarly, median OS was prolonged in the abemaciclib arm compared with placebo arm in patients harboring either PIK3CA-mutant tumors (44.5 months vs. 33.8 months) compared with wild-type (55.5 months vs. 39.7 months), or ESR1-mutant tumors (not reached vs. 42.2 months) compared with wild-type (52.2 months vs. 29.4 months). Although benefit was observed in the abemaciclib plus fulvestrant arm regardless of mutation status, this study is limited by the small sample size, as a subgroup of the MONARCH 2 study population. These exploratory data are hypothesis-generating and provide insight on abemaciclib treatment for tumors with and without PIK3CA or ESR1 mutations. While acknowledging the limitations of cross study comparisons, the similarity of outcomes for the single-agent fulvestrant control arms in PIK3CA-mutant cancers between MONARCH 2 and SOLAR-1 and the magnitude of improvement in PFS and OS demonstrated in MONARCH 2 also highlights a potential therapeutic strategy for treatment of HR+, PIK3CA-mutant metastatic breast cancer and support further evaluation in prospective, and suitably powered, clinical trials.

S.M. Tolaney reports grants and personal fees from Eli Lilly during the conduct of the study. S.M. Tolaney also reports grants and personal fees from AstraZeneca, Merck, Nektar, Novartis, Pfizer, Genentech/Roche, Gilead, BMS, Eisai, Nanostring, Sanofi, Odonate, and Seattle Genetics; grants from Exelixis and Cyclacel; and personal fees from Puma, Athenex, Daiichi Sankyo, OncoPep, Kyowa Kirin Pharmaceuticals, Samsung Bioepsis Inc., CytomX, Certara, Mersana Therapeutics, Ellipses Pharma, 4D Pharma, OncoSec Medical Inc., Chugai Pharma, BeyondSpring Pharmaceuticals, OncXerna, Zymeworks, and Zentalis outside the submitted work. M. Toi reports grants, personal fees, and other support from Eli Lilly during the conduct of the study. M. Toi also reports grants, personal fees, and other support from AstraZeneca, Daiichi Sankyo, and Kyowa-Kirin; other support from Athenex Oncology, Bertis, Kansai Medical Net, Terumo, and BMS; grants and personal fees from Chugai, Eisai, Nippon Kayaku, Pfizer, Shimadzu, Taiho, Takeda, and Yakult; personal fees from Novartis, Devicore Medical Japan, and Exact Science; grants from GL Science, The JBCRG association, Shionogi, Astellas, and The KBCRN association; and grants and other support from Luxonus outside the submitted work. M.Toi has a patent for publication number: 20210205822 pending, a patent for patent number: 10766040 issued, a patent for patent number: 10638934 issued, a patent for publication number: 20190032103 pending, a patent for publication number: 20180275131 pending, a patent for publication number: 20180136212 pending, a patent for patent number: 9873916 issued, a patent for patent number: 9370307 issued, a patent for publication number: 20140350358 pending, a patent for patent number: 8865663 issued, a patent for patent number: 8778332 issued, a patent for publication number: 20140073533 pending, a patent for publication number: 20130085372 pending, a patent for patent number: 8362027 issued, a patent for publication number: 20120201797 pending, a patent for publication number: 20110295189 pending, a patent for patent number: 8034824 issued, a patent for publication number: 20110015144 pending, a patent for publication number: 20090281063 pending, and a patent for publication number: 20060183794 pending. In addition, M. Toi reports employment with Organization for Oncology and Translational Research, The Japan Breast Cancer Research Group Association, The Japanese Onco-Cardiology Society, The Kyoto Breast Cancer Research Network Association, Asian Journal of Breast Surgery, Asian Journal of Surgery, Breast Cancer Research and Treatment, British Journal of Cancer, Cancer Science, Frontiers in Women's Cancer, International Journal of Oncology, and Scientific Reports. P. Neven reports other support from Novartis, Lilly, and Pfizer outside the submitted work. J. Sohn reports grants from MSD, Roche, Novartis, AstraZeneca, Lilly, Pfizer, GSK, Daiichi Sankyo, Sanofi, and Boehringer Ingelheim outside the submitted work. E.-M. Grischke reports nonfinancial support from Lilly during the conduct of the study. A. Llombart-Cussac reports grants, personal fees, and nonfinancial support from Eli Lilly during the conduct of the study. A. Llombart-Cussac also reports grants from AGENDIA; grants and personal fees from Exact Science and Novartis; grants, personal fees, and nonfinancial support from Daiichi Sankyo, Roche, Pfizer, and AstraZeneca; and personal fees from Seagen, Gilead, and Pierre Fabre outside the submitted work. In addition, A. Llombart-Cussac is a co-founder of MedSIR and has ownership of stock in MedSIR. H. Soliman reports personal fees from Novartis, Eisai, AstraZeneca, and Seattle Genetics during the conduct of the study. S. Wijayawardana reports personal fees from Eli Lilly during the conduct of the study, as well as personal fees from Eli Lilly outside the submitted work. V.M. Jansen reports other support from Eli Lilly during the conduct of the study, as well as other support from Mersana Therapeutics and Elevation Oncology outside the submitted work. L.M. Litchfield reports personal fees from Eli Lilly and Company during the conduct of the study, and has ownership of stock in Eli Lilly and Company. G.W. Sledge Jr reports employment with AstraZeneca and Novartis. No disclosures were reported by the other authors.

S.M. Tolaney: Conceptualization, supervision, investigation, writing–review and editing. M. Toi: Investigation. P. Neven: Investigation, writing–review and editing. J. Sohn: Investigation. E.-M. Grischke: Investigation. A. Llombart-Cussac: Investigation. H. Soliman: Investigation. H. Wang: Conceptualization, formal analysis, methodology, writing–original draft. S. Wijayawardana: Formal analysis. V.M. Jansen: Conceptualization, supervision, project administration, writing–review and editing. L.M. Litchfield: Conceptualization, supervision, project administration, writing–review and editing. G.W. Sledge Jr: Investigation, writing–review and editing.

The authors are grateful to the patients, their families, and caregivers for participating in MONARCH 2. Authors thank the study investigators and site staff for their participation. Writing and editing provided by Nicholas Pulliam, employee of Eli Lilly and Company; funding provided by Eli Lilly and Company.

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.