Baseline Characteristics, Treatment Patterns, and Outcomes in Patients with HER2-Positive Metastatic Breast Cancer by Hormone Receptor Status from SystHERs

Breast cancer comprises more than 25% of newly diagnosed cases of cancer in women worldwide (1). Of women in the United States with metastatic breast cancer (MBC), 14.8% have HER2-positive and hormone receptor (HR)-positive MBC, and 8.9% have HER2-positive and HR-negative disease (2). Among patients with HER2-positive MBC, clinical outcomes have been shown to differ between patients with HR-positive versus HR-negative disease. In an observational study that enrolled patients with HER2-positive MBC from 2007 to 2009, patients with HR-positive disease had longer median survival than those with HR-negative disease (34.4 vs 19.8 months; ref. 3).

Data from clinical trials have suggested that dual targeting of HR and HER2 is associated with better progression-free survival (PFS) in patients with HR-positive, HER2-positive MBC (4–6). Results from registHER, a registry study of patients with HER2-positive MBC enrolled from 2003 to 2006, showed improved PFS with the addition of endocrine therapy to first-line trastuzumab + chemotherapy [20.4 vs 9.5 months with trastuzumab + chemotherapy without endocrine therapy; hazard ratio, 0.53; 95% confidence interval (CI), 0.42–0.68] and lower risk of death (hazard ratio, 0.50; 95% CI, 0.36–0.70) in patients with HR-positive, HER2-positive MBC (7). Furthermore, overall survival (OS) was longer with the use of sequential (vs concurrent) chemotherapy and endocrine therapy (hazard ratio, 0.48; 95% CI, 0.26–0.89). However, there are limited recent data detailing treatments actually administered to this patient population in the real world. This gap is notable in light of the United States approval of several new HER2-targeted therapies for MBC, including lapatinib (2007), pertuzumab (2012), and trastuzumab emtansine (T-DM1; 2013). The combination of pertuzumab + trastuzumab + taxane is a current first-line standard of care for patients with HER2-positive MBC based on results from CLEOPATRA, in which it demonstrated significantly improved clinical outcomes relative to trastuzumab + taxane alone (8, 9). As clinical trials supporting FDA approval of modern HER2-targeted agents generally did not allow for use of endocrine therapy while patients were on study treatment, additional contemporary data are needed to understand treatment patterns and outcomes in patients with HR-positive, HER2-positive MBC within this treatment landscape.

In addition to the approval of several new HER2-targeted therapies in recent years, the accepted definition of HR positivity has shifted over the past decade. In 2010, the American Society of Clinical Oncology (ASCO) and the College of American Pathologists (CAP) proposed a cutoff of ≥1% of cells staining HR-positive [e.g., tumor cell immunoreactivity for the estrogen receptor (ER) and/or progesterone receptor (PR)] to differentiate HR-positive from HR-negative disease, replacing the previously common threshold of ≥10% (10). Little is known about the characteristics, treatment patterns, and outcomes of patients with MBC who also have low HR positivity (1%–9%), and whether this patient subgroup resembles those with moderate/high HR positivity (10%–100%) versus HR-negative MBC in the real world.

The Systemic Therapies for HER2-positive Metastatic Breast Cancer Study (SystHERs) was a fully enrolled (2012–2016), observational, prospective registry study of patients with recently diagnosed HER2-positive MBC in the United States. Representing a large contemporary population of patients with HER2-positive MBC, SystHERs is well-suited to characterize the management of patients with HR-positive, HER2-positive disease in the real-world setting. Here, we report baseline characteristics, treatment patterns, and clinical outcomes from SystHERs in patients with HR-positive and HR-negative disease, including subgroups of patients stratified by percentage of HR positivity [low (1%–9%) vs moderate/high (10%–100%)], MBC diagnosis type (de novo vs recurrent), age (<50 years vs 50–69 years), and race (black/African American vs white).

SystHERs (NCT01615068) was a prospective, multicenter, observational cohort study based in the United States. Patients were assessed and treated per their treating physician's standard practice. The primary study endpoint was treatment patterns; secondary endpoints included clinical outcomes. Additional study design information is included in the published study protocol (11).

At enrollment, eligible patients were aged ≥18 years and had an initial diagnosis of HER2-positive MBC within 6 months. Written informed consent for the use of medical records was provided by participating patients. SystHERs was conducted per FDA regulations, the International Conference on Harmonisation E6 Guidelines for Good Clinical Practice, the Declaration of Helsinki, and applicable local laws. Study sites obtained approval of the study protocol by the site's ethics committee or institutional review board (IRB). A central IRB was used for sites that did not have an IRB. Although initially planned to include at least 5 years of follow-up (11), the SystHERs registry was terminated early due to the sponsor's decision to prioritize evaluation of new therapies in patients with breast cancer.

HER2 status was determined locally by the treating physician based on the primary tumor or biopsy of recurrence. HR-positive disease was defined based on physician report, as captured on the case report form. Tumors were defined as HR-positive if any measurement from early breast cancer, MBC, or the primary breast tumor was recorded as ER-positive and/or PR-positive. In patients with HR-positive disease, the percentage of HR positivity was defined as the highest percentage of ER-positive or PR-positive cells. The percentage of ER-positive or PR-positive cells was not reported for patients considered by local investigators to have HR-negative disease.

Disease history, previous cancer-related treatment data, and baseline patient and disease characteristics (some data previously reported [12, 13]) were collected at enrollment. Treatments for MBC, disease progression, and clinical outcomes were captured quarterly from patient charts, diagnostic tests, laboratory findings, and clinical notes. Treating physicians identified metastatic sites at MBC diagnosis and on study. Patients who discontinued the study were provided the option to participate in quarterly survival follow-up.

Key patient cohorts were defined as those with HR-positive versus HR-negative disease. Secondary subgroups included those stratified by percentage of HR positivity (1%–9% and 10%–100% HR positivity; i.e., low and moderate/high HR positivity, respectively), MBC diagnosis type (de novo and recurrent disease, in which de novo MBC was defined as an initial breast cancer diagnosis with distant metastases observed concurrently, or confirmed within 90 days), age at enrollment (<50 years, 50–69 years, and ≥70 years), and race (black/African American and white).

Median follow-up was calculated as the median observation time from MBC diagnosis to data cutoff. First-line treatment was defined as any therapy received for MBC up to first disease progression. OS was calculated as the duration from MBC diagnosis to death. The Kaplan–Meier product-limit method was used to estimate OS, which was compared across cohorts using a log-rank test. Multivariate Cox regression was used to evaluate the association between baseline patient and disease characteristics and OS.

Patients in SystHERs were enrolled from June 2012 to June 2016. Among 977 patients who met eligibility criteria, 70.1% (685/977) had HR-positive and 29.9% (292/977) had HR-negative MBC. Within the HR-positive group, 60 (8.8%) had 1%–9% HR positivity and 550 (80.3%) had 10%–100% HR positivity; 75 (10.9%) patients were missing data for the percentage of HR positivity (Supplementary Fig. S1). At the February 22, 2018 data cutoff date, median follow-up duration was 27.6 months from MBC diagnosis for all patients, and 28.6 and 25.3 months in patients with HR-positive and HR-negative disease, respectively. The most common reasons for patient discontinuation were death [HR-positive: 26.9% (184/685) and HR-negative: 40.8% (119/292)] and sponsor decision to terminate the study [HR-positive: 54.3% (372/685) and HR-negative: 46.9% (137/292)].

When comparing patient baseline demographics and disease characteristics between the HR-positive and HR-negative cohorts, differences were observed in race, MBC diagnosis type, proportion of visceral disease, and metastatic site locations (Table 1). The HR-positive cohort (vs HR-negative cohort) had a lower proportion of black/African American patients [13.4% (92/685) vs 20.2% (59/292)] and a higher proportion of white patients [81.3% (557/685) vs 71.6% (209/292)]. The HR-positive cohort (vs HR-negative cohort) also had a lower proportion of patients with de novo MBC [46.3% (317/685) vs 58.2% (170/292)] and visceral disease [63.1% (432/685) vs 74.3% (217/292)]. Liver metastasis was present in 35.3% (242/685) of patients with HR-positive disease and 42.5% (124/292) of patients with HR-negative disease. Bone metastasis was present in 55.8% (382/685) of patients with HR-positive disease and 41.4% (121/292) of patients with HR-negative disease.

Within the HR-positive cohort, patients with low HR positivity (1%–9%) differed from those with moderate/high HR positivity (10%–100%) in the proportion of black/African American patients [20.0% (12/60) vs 12.9% (71/550), respectively]. Small disparities were also observed between the subgroups with low HR positivity (1%–9%) and moderate/high HR positivity (10%–100%) in the proportions of patients with de novo MBC [41.7% (25/60) vs 48.5% (267/550)], visceral disease [66.7% (40/60) vs 62.5% (344/550)], liver metastasis [40.0% (24/60) vs 35.6% (196/550)], and bone metastasis [50.0% (30/60) vs 56.7% (312/550); Supplementary Table S1].

In patients with HR-positive disease and HR status data available for both primary and metastatic tumors, concordance for HR positivity between primary and metastatic tissue was 81.0% (Supplementary Table S2A). Concordance was numerically lower in patients with low HR positivity (1%–9%) versus moderate/high HR positivity (10%–100%; 47.4% vs 87.1%, respectively; Supplementary Table S2B and S2C). In both HR-positive subgroups, patients with discordant HR status had HR-positive primary tumors and HR-negative metastatic tumors more commonly (low HR positivity: 25.0%; moderate/high HR positivity: 5.6%) than the converse; that is, HR-negative primary and HR-positive metastatic tumors (low HR positivity: 8.3%; moderate/high HR positivity: 3.1%).

At data cutoff, 97.8% (670/685) of the HR-positive cohort and 95.2% (278/292) of the HR-negative cohort had received first-line systemic treatment for MBC. In the HR-positive cohort, the most common regimen was HER2-targeted therapy + chemotherapy + endocrine therapy, administered to 310 (45.3%) patients, with the sequential administration of endocrine therapy (i.e., chemotherapy induction followed by maintenance endocrine therapy) more common than concurrent administration [34.9% (239/685) vs 10.4% (71/685), respectively; Table 2]. The second most common regimen was HER2-targeted therapy + chemotherapy without endocrine therapy, received by 247 (36.1%) patients. In addition, 75 (10.9%) patients received HER2-targeted therapy + endocrine therapy (no chemotherapy), 3 (0.4%) received chemotherapy + endocrine therapy, 17 (2.5%) received endocrine therapy alone, and 18 (2.6%) received HER2-targeted therapy alone or chemotherapy alone (no endocrine therapy).

Overall, 59.1% (405/685) of patients with HR-positive disease had any first-line exposure to endocrine therapy, most frequently aromatase inhibitors [46.3% (317/685)]. Endocrine therapy was administered with HER2-targeted therapy in 56.2% (385/685) of patients, which included trastuzumab + pertuzumab in 40.4% (277/685). Of note, 56 patients with HR-positive breast cancer progressed, withdrew, or were lost to follow-up within 9 months on study, while 142 patients were still in the first-line treatment phase without any endocrine treatment at data cutoff (56 of those patients were still receiving first-line chemotherapy). As such, the number of patients planned to receive or actually administered first-line endocrine therapy may be underestimated.

In the HR-negative cohort, the most common first-line regimen was HER2-targeted therapy + chemotherapy [87.7% (256/292); Table 2]. Patients with HR-positive and HR-negative disease received HER2-targeted therapies in similar proportions [94.5% (647/685) vs 94.5% (276/292)], but patients with HR-positive disease were treated with chemotherapy less commonly than those with HR-negative disease [82.2% (563/685) vs 91.8% (268/292)]. A regimen of trastuzumab + pertuzumab + taxane (with or without endocrine therapy) was administered to a slightly lower proportion of patients with HR-positive than HR-negative disease [65.5% (449/685) vs 69.2% (202/292)].

Within the HR-positive cohort, patients with tumors showing low HR positivity (1%–9%) received first-line HER2-targeted therapy + chemotherapy + endocrine therapy at a notably lower proportion than those with moderate/high HR positivity [10%–100%; 25.0% (15/60) vs 49.1% (270/550)] and were less likely to receive endocrine therapy overall [33.3% (20/60) vs 64.2% (353/550); Supplementary Table S3]. Rather, patients with tumors showing low HR positivity (1%–9%) more commonly received HER2-targeted therapy + chemotherapy without endocrine therapy compared with patients whose tumors had moderate/high HR positivity [10%–100%; 60.0% (36/60) vs 32.0% (176/550)].

Patients with de novo HR-positive MBC were treated with first-line chemotherapy [86.1% (273/317) vs 78.8% (290/368)] and/or endocrine therapy [65.6% (208/317) vs 53.5% (197/368)] more commonly than those with recurrent HR-positive disease (Supplementary Table S4). HER2-targeted therapy + chemotherapy + endocrine therapy was administered to a higher proportion of patients with de novo versus recurrent disease within the HR-positive cohort [55.2% (175/317) vs 36.7% (135/368)], as was the administration of trastuzumab + pertuzumab + taxane [with or without endocrine therapy; 71.0% (225/317) vs 60.9% (224/368)].

Among patients with HR-positive disease, use of first-line endocrine therapy was more common with increasing age [<50 years: 56.8% (104/183), 50–69 years: 62.8% (221/352), and ≥70 years: 77.4% (65/84)], while use of chemotherapy became less common [<50 years: 90.2% (165/183), 50–69 years: 85.2% (300/352), and ≥70 years: 63.1% (53/84)]. Black/African American and white patients received HER2-targeted therapy and chemotherapy at similar rates within each HR cohort, but in patients with HR-positive disease, use of endocrine therapy was slightly more common in black/African American patients [68.3% (56/82) vs 61.6% (309/502)].

By data cutoff, 30.4% (208/685) of patients in the HR-positive cohort and 42.1% (123/292) of patients in the HR-negative cohort had died. Estimated median OS from MBC diagnosis was longer in patients with HR-positive disease (53.0 months) than those with HR-negative disease (43.4 months; hazard ratio, 0.70; 95% CI, 0.56–0.87; Fig. 1). A multivariate analysis adjusting for baseline patient and disease characteristics indicated that HR-positive status is associated with better survival outcomes than HR-negative status (Supplementary Fig. S2). Median OS in patients with low HR-positive tumors (1%–9%) was shorter compared with patients with moderate/high HR-positive tumors (10%–100%; 40.1 vs 53.8 months) and similar to patients with HR-negative disease (43.4 months; Fig. 2).

In both the HR-positive and HR-negative cohorts, patients with de novo MBC had longer median OS than those with recurrent MBC (HR-positive: de novo, not reached vs recurrent, 45.1 months; HR-negative: de novo, 54.6 months vs recurrent, 30.8 months; Fig. 3). Relative to patients 50–69 years old, patients <50 years old at MBC diagnosis had numerically longer OS in both the HR-positive (49.3 vs 56.8 months, respectively) and HR-negative cohorts (42.7 vs 54.6 months, respectively; Supplementary Fig. S3). Black/African American and white patients had similar median OS in the HR-positive cohort (56.8 vs 53.0 months, respectively), but in the HR-negative cohort, black/African American patients had numerically shorter median OS compared with white patients (36.4 vs 46.7 months; Supplementary Fig. S4).

Within the HR-positive cohort, patients who received any first-line endocrine therapy had a median OS of “not reached” (95% CI, 51.7–not reached), compared with 40.1 months (95% CI, 35.9–53.0 months) in patients who did not receive first-line endocrine therapy. Median PFS for patients in the HR-positive cohort who received any first-line endocrine therapy was 20.4 months (95% CI, 18.4–23.3 months) compared with 10.2 months (95% CI, 9.0–11.9 months) in patients in this cohort who did not receive first-line endocrine therapy.

Our analysis of patients with HER2-positive MBC from the real-world SystHERs study indicated that 56.2% of patients with HR-positive disease received first-line treatment with HER2-targeted therapy + endocrine therapy, which was commonly administered as a regimen of HER2-targeted therapy + chemotherapy + sequential endocrine therapy (34.9% of HR-positive patients). First-line endocrine therapy was used less commonly in patients in the HR-positive cohort who had tumors with low (1%–9%) versus moderate/high HR positivity (10%–100%), possibly due to variable use of the revised cutoff from ASCO/CAP (10) for considering tumors HR positive. Patients with HR-positive disease had longer median OS compared with those with HR-negative disease (53.0 vs 43.4 months), potentially reflecting less aggressive disease in the HR-positive cohort and also possibly due to the availability of another targeted therapy (i.e., endocrine therapy) for these patients. Survival outcomes were also better for patients with de novo (rather than recurrent) MBC in both HR subgroups, presumably due to the lack of acquired drug resistance due to prior adjuvant therapy.

Current ASCO guidelines for use of endocrine therapy in HR-positive MBC recommend treatment with endocrine therapy in patients with MBC and any percentage of HR positivity (6). In patients with HER2-positive MBC that is also HR positive, initial first-line treatment with HER2-targeted therapy and chemotherapy is recommended on the basis of demonstrated benefits in PFS and OS with dual antibody therapy, specifically pertuzumab + trastuzumab in combination with a taxane (8, 9), with the potential use of maintenance endocrine therapy following cessation of chemotherapy (6, 14). In this investigation from SystHERs, we found that 65.5% of patients with HR-positive disease received trastuzumab + pertuzumab + taxane (with or without endocrine therapy) and 69.2% of patients with HR-negative disease also received trastuzumab + pertuzumab + taxane. ASCO guidelines further state that for patients with HER2-positive and ER/PR-positive breast cancer, clinicians may recommend either standard first-line therapy or, for selected patients, endocrine therapy plus HER2-targeted therapy or endocrine therapy alone (14). Such recommendations reflect evolving preferred treatment patterns in the absence of data from randomized, double-blind, controlled trials (14), which may be due, in part, to availability of results from prospective outcome studies such as registHER (6, 7). For example, observations from the registHER study demonstrated that the addition of endocrine therapy to trastuzumab + chemotherapy was associated with longer median PFS and OS compared with HER2-targeted therapy + chemotherapy alone (7). Furthermore, there was a trend toward improved OS with sequential use of chemotherapy + endocrine therapy versus concurrent administration (7). Despite such data and guidelines, in SystHERs we found that only 45.3% of patients with HR-positive, HER2-positive MBC received first-line HER2-targeted therapy + chemotherapy + endocrine therapy (34.9% sequential and 10.4% concurrent). Most of the remaining patients received HER2-targeted therapy + chemotherapy only (36.1%) or HER2-targeted therapy + endocrine therapy (10.9%). Overall, first-line endocrine therapy was administered to 59.1% of patients in the HR-positive subgroup. Of note, first-line maintenance endocrine therapy may have been underestimated due to the limited follow-up time. In addition, as some patients progressed prior to completing a course of first-line chemotherapy, the actual administration of maintenance endocrine therapy may have been lower than planned administration.

It remains to be seen whether maintenance first-line endocrine therapy following pertuzumab + trastuzumab + taxane treatment, the current standard of care, confers a survival advantage in patients with HR-positive, HER2-positive disease. The PERTAIN study investigated first-line trastuzumab + an aromatase inhibitor (AI) with or without pertuzumab and with or without induction chemotherapy in patients with HER2-positive, HR-positive MBC or locally advanced breast cancer (15). Data from PERTAIN showed that in patients who received induction chemotherapy (i.e., trastuzumab + induction chemotherapy + AI with or without pertuzumab), PFS was similar between those who received pertuzumab (16.89 months) and those who did not (16.85 months). For patients who did not receive induction chemotherapy (i.e., trastuzumab + AI with or without pertuzumab), PFS was longer in patients who received pertuzumab (21.72 months) compared with those who did not (12.45 months). Importantly, PERTAIN was not designed to show differences between patients with and without induction chemotherapy.

In this study, patients with HR-positive disease and tumors with moderate/high HR positivity exhibited longer OS than those with tumors with low HR positivity or HR-negative disease. OS did not notably differ between patients in the low HR positivity versus HR-negative disease subgroups. However, it is unclear whether low versus high HR positivity are truly prognostic indicators for survival, or if the disparities we observed result from differences in baseline characteristics or treatments administered between the two groups. Despite ASCO/CAP recommendations that tumors with 1%–9% HR positivity be considered HR-positive with regard to disease management (10), our analysis of data from SystHERs suggests that these patients may have different treatment patterns compared with the 10%–100% HR-positive group. Only 33.3% of patients with 1%–9% HR-positive disease received endocrine therapy, versus 64.2% of patients with tumors with 10%–100% HR positivity. Data from recent retrospective studies in early breast cancer suggest that patients with primary breast tumors expressing ER in 1%–9% of cells may derive limited or no benefits from endocrine therapy (16, 17). While the number of patients with 1%–9% HR positivity in SystHERs (n = 60) is too limited to assess the association of specific treatment regimens with clinical outcomes, future studies should investigate whether low HR positivity in patients with HER2-positive MBC is predictive of clinical benefits from endocrine therapy.

Among patients with HR-positive disease stratified by MBC diagnosis type, age, or race, those with de novo MBC, older patients, and black/African American patients received regimens containing first-line endocrine therapy more commonly than those with recurrent MBC, younger patients, and white patients, respectively. Chemotherapy was also more commonly administered to patients with de novo (vs recurrent) MBC, and was administered to older patients less frequently than younger patients. Chemotherapy use was similar between black/African American and white patients. Patients with HR-positive breast cancer had consistently longer OS than those with HR-negative cancer. This difference may potentially be attributable, in part, to the availability of endocrine therapy for these patients, as well as in those patients with intrinsically less aggressive biology.

Among patients with HER2-positive MBC in the SystHERs registry, 70.1% were identified as having HR-positive disease compared with 55.0% of patients in registHER, which enrolled patients from 2003 to 2006 (7). Statistical modeling of data from the NCI's Surveillance Epidemiology and End Results (SEER) Program projected that the proportion of patients with HR-positive breast cancer would increase between 2009 and 2016, particularly in younger women (18). As of 2010, data from SEER suggested that among patients with known HR and HER2 status, 5,240 of 7,568 patients (69.2%) with HER2-positive breast cancer had HR-positive status (2). This increase could potentially be due to the ASCO/CAP-advised changes to the HR-positive cutoff criteria in 2010 (10), raising the number of patients effectively classified as HR positive. This increase could also potentially be due to differences in adjuvant treatments (e.g., trastuzumab) that could have influenced the natural history and characteristics of patients who develop MBC and/or bias in the time to recurrence between HR-positive versus HR-negative disease. Given the availability of endocrine therapies used for the treatment of HR-positive, but not HR-negative breast cancer, such a trend may ultimately increase survival rates in patients with breast cancer overall.

Patient sampling in SystHERs was methodologically rigorous, with a very low rate of refusals over the enrollment period, and the patient population was nationally representative. Therefore, our patient population likely provides an accurate depiction of real-world practice and treatment patterns over the study period. Limitations of this analysis include short follow-up time in some patients, potentially underestimating the percentage of patients with HR-positive disease who received maintenance first-line endocrine therapy; a relatively small number of patients in the low HR-positive subgroup; and missing percentages for HR positivity in 10.9% of patients from the HR-positive subgroup. Furthermore, as the percentage of ER-positive or PR-positive cells was not reported for patients considered by local investigators to have HR-negative disease, it is possible that some patients with low HR positivity may have been included in the HR-negative subgroup. Also, treatment decisions based on aggressiveness of clinical features and patient performance statuses could have introduced selection bias in outcomes associated with specific treatments. For example, among patients receiving induction chemotherapy, those with more aggressive disease might progress and move on to other chemotherapy regimens rather than sequential endocrine therapy. This removes such patients with aggressive disease from the group receiving sequential treatment, thereby potentially selecting for better outcomes in the patients who remain.

In summary, data from SystHERs suggest that some patients with HR-positive, HER2-positive breast cancer in the real world may be undertreated despite real-world experience that the addition of maintenance endocrine therapy to maintenance HER2-targeted therapy, following HER2-targeted therapy + chemotherapy, is associated with improved survival. Furthermore, we observed that most patients' tumors showing low HR positivity were not administered endocrine therapy, potentially influenced by studies indicating limited benefits for endocrine therapy in this population. To optimize disease management in the potentially growing number of patients with HR-positive, HER2-positive breast cancer, future studies should assess (i) the efficacy of various treatment and sequencing combinations with endocrine therapy, including whether the increased administration of first-line HER2-targeted therapy, chemotherapy, and maintenance endocrine therapy improves outcomes in patients with HR-positive, HER2-positive MBC in the real world; and (ii) the utility of endocrine therapy in the population of patients with tumors expressing low HR positivity.

M. Cobleigh holds an ownership interest (including patents) in Genomic Health and is an unpaid consultant/advisory board member for Genentech, Marcogenics, Immunomedics, PUMA, Genomic Health, and GlaxoSmithKline. D.A. Yardley is a paid consultant/advisory board member for Novartis, Genentech/Roche, Daiichi Sankyo/Lilly, Eisai, Celgene, Biotheranostics, NanoString Technologies, and Bristol-Myers Squibb; reports receiving speakers bureau honoraria from Novartis and Genentech/Roche; reports receiving commercial research grants via her institution from AstraZeneca, Genentech/Roche, Syndax, Novartis, MedImmune, Lilly, Medivation, Pfizer, Eisai, Tesaro, Macrogenics, Abbvie, Immunomedics, Daiichi Sankyo, Merck, Clovis Oncology, Oncothyeron, and InventisBio; and reports receiving other remuneration from Novartis and Genentech/Roche. A.M. Brufsky is an unpaid consultant/advisory board member for Roche. S.M. Swain reports receiving commercial research grants to her institution from Genentech/Roche and Pfizer; is an unpaid consultant/advisory board member for Cardinal Health, Daiichi-Sankyo, Eli Lilly & Company, Genentech/Roche, Genomic Health, Inivata Ltd., Molecular Therapeutics, Novartis, Pieris Pharmaceuticals, and Tocagen; and reports receiving other remuneration from AstraZeneca, Bristol-Myers Squibb, Cardinal Health, Caris Life Sciences, Daiichi-Sankyo, Eli Lilly & Company, Genentech/Roche, Inivata Ltd., NanoString Technologies, Novartis, and Roche. P.A. Kaufman is a paid consultant for and reports receiving commercial research grants from Roche/Genentech. D. Tripathy is an unpaid consultant/advisory board member for Puma Biotechnology. S.A. Hurvitz reports receiving commercial research grants from Amgen, Ambrx, Bayer, Daiichi Sankyo, Genentech/Roche, GlaxoSmithKline, Immunomedics, Lilly, Macrogenics, Novartis, OBI Pharma, Pfizer, Pieris, PUMA, Radius, Sanofi, and Seattle Genetics; and is an unpaid consultant/advisory board member for Novartis, Glaxo-SmithKline, Amgen, Macrogenics, and Seattle Genetics. J. O'Shaughnessy is a paid consultant for Abbvie, Agendia, Amgen Biotechnology, AstraZeneca, Bristol-Myers Squibb, Celgene Corporation, Eisai, Genentech, Genomic Health, GRAIL, Immunomedics, Heron Therapeutics, Ipsen Biopharmaceuticals, Jounce Therapeutics, Lilly, Merck, Myriad, Novartis, Ondonate Therapeutics, Pfizer, Puma Biotechnology, Prime Oncology, Roche, Seattle Genetics, and Syndax Pharmaceuticals. V. Antao is an employee of Genentech and holds an ownership interest (including patents) in Hoffmann La Roche. H. Li is an employee of F. Hoffmann La Roche. L. Chu is an employee of and reports receiving other remuneration from Genentech/Roche. M. Jahanzeb is a paid consultant for and reports receiving commercial research grants from Genentech. No potential conflicts of interest were disclosed by the other authors.

Conception and design: M. Cobleigh, A.M. Brufsky, H.S. Rugo, S.M. Swain, P.A. Kaufman, D. Tripathy, S.A. Hurvitz, J. O'Shaughnessy, V. Antao, L. Chu, M. Jahanzeb

Development of methodology: M. Cobleigh, D.A. Yardley, A.M. Brufsky, H.S. Rugo, S.M. Swain, D. Tripathy, V. Antao, L. Chu, M. Jahanzeb

Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): M. Cobleigh, D.A. Yardley, A.M. Brufsky, H.S. Rugo, S.M. Swain, P.A. Kaufman, D. Tripathy, S.A. Hurvitz, J. O'Shaughnessy, L. Chu, M. Jahanzeb

Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): M. Cobleigh, D.A. Yardley, A.M. Brufsky, H.S. Rugo, S.M. Swain, P.A. Kaufman, D. Tripathy, J. O'Shaughnessy, V. Antao, H. Li, L. Chu, M. Jahanzeb

Writing, review, and/or revision of the manuscript: M. Cobleigh, D.A. Yardley, A.M. Brufsky, H.S. Rugo, S.M. Swain, P.A. Kaufman, D. Tripathy, S.A. Hurvitz, J. O'Shaughnessy, G. Mason, V. Antao, H. Li, L. Chu, M. Jahanzeb

Administrative, technical, or material support (i.e., reporting or organizing data, constructing databases): V. Antao

Study supervision: M. Cobleigh, A.M. Brufsky, S.M. Swain, V. Antao, M. Jahanzeb

SystHERs was funded by F. Hoffmann-La Roche/Genentech, Inc. The authors are grateful to the patients, families, and investigators who participated in SystHERs. We would also like to thank: Musa Mayer for her work as part of the SystHERs Steering Committee; the entire SystHERs team, including clinical operations leads Michelle Usher (F. Hoffmann-La Roche/Genentech, Inc.) and Sandy Lam (F. Hoffmann-La Roche/Genentech, Inc.); Bongin Yoo (F. Hoffmann-La Roche/Genentech, Inc.) for his contributions to the statistical analysis; Allen Lee (Everest Clinical Research Services, Inc.) for his assistance with the statistical analysis; and Bokai Xia (F. Hoffmann-La Roche/Genentech, Inc.) for his statistical programming expertise. Support for third-party writing assistance was provided by Sabrina Hom, PhD, of CodonMedical, an Ashfield Company, part of UDG Healthcare plc, and funded by F. Hoffmann-La Roche/Genentech, Inc. F. Hoffmann-La Roche/Genentech, Inc. funded the SystHERs study and participated in the study design, data collection, data analysis, data interpretation, and writing of this article.

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.