Real-World First-Line Use of Pertuzumab With Different Taxanes for Human Epidermal Growth Factor Receptor 2–Positive Metastatic Breast Cancer: A Comparative Effectiveness Study Using US Electronic Health Records

PURPOSE

On the basis of the results from CLEOPATRA, pertuzumab plus trastuzumab and chemotherapy is the first-line standard of care for human epidermal growth factor receptor 2 (HER2)-positive metastatic breast cancer (MBC). However, discrepancies have been reported between clinical trial and real-world outcomes. We report real-world outcomes for patients with HER2-positive MBC treated with first-line pertuzumab plus trastuzumab and a taxane in routine clinical practice in the United States.

METHODS

A retrospective analysis was conducted using electronic health record-derived deidentified data from the Flatiron Health database. Patients were grouped according to the first taxane received (paclitaxel/nab-paclitaxel or docetaxel). Median real-world progression-free survival (rwPFS) and overall survival (rwOS) was estimated using Kaplan-Meier methodology. Subgroup analyses were conducted in patients treated with docetaxel who met CLEOPATRA's key eligibility criteria.

RESULTS

We included 1,065 patients; 313 patients received paclitaxel/nab-paclitaxel and 752 received docetaxel. Patients who received paclitaxel/nab-paclitaxel were older, had a worse Eastern Cooperative Oncology Group Performance Status, and had more recurrent metastatic disease compared with the docetaxel group. After adjustment for potential confounders, similar median rwPFS (inverse probability of treatment weighted average treatment effect for the treated [IPTW-ATT] hazard ratio [HR], 1.09; 95% CI, 0.9 to 1.3; P = .365) and rwOS (IPTW-ATT HR, 1.23; 95% CI, 0.96 to 1.58; P = .101) was observed between treatment groups. In the subgroup of CLEOPATRA-eligible patients, median rwPFS and rwOS were 16.9 months and 57.8 months, respectively.

CONCLUSION

There was no statistically significant difference in real-world outcomes between patients treated with paclitaxel/nab-paclitaxel and those treated with docetaxel. Selecting patients using key CLEOPATRA eligibility criteria resulted in rwPFS and rwOS similar to those observed in CLEOPATRA, highlighting the importance of ensuring similar patient populations when comparing clinical trial and real-world data.

INTRODUCTION

CLEOPATRA demonstrated significantly improved progression-free survival (PFS) and overall survival (OS) when adding pertuzumab to trastuzumab and docetaxel for first-line treatment of patients with human epidermal growth factor receptor 2 (HER2)-positive metastatic breast cancer (MBC).^1-4 On the basis of these results, pertuzumab plus trastuzumab and chemotherapy is the first-line standard of care for patients with HER2-positive MBC.⁵ Preliminary results (52 months' median follow-up) from the PERUSE study of pertuzumab plus trastuzumab and a taxane (paclitaxel/nab-paclitaxel/docetaxel) in patients with HER2-positive MBC were consistent with CLEOPATRA's results and demonstrated that paclitaxel is a valid alternative to docetaxel with regards to efficacy and safety.⁶ Similar results were also observed at the final analysis (69 months' median follow-up).⁷

CONTEXT

• Key Objective

• To compare the effectiveness of using paclitaxel/nab-paclitaxel or docetaxel with first-line pertuzumab plus trastuzumab to treat patients with human epidermal growth factor receptor 2–positive metastatic breast cancer in routine clinical practice in the United States and to evaluate the impact of applying key trial inclusion/exclusion criteria on real-world outcomes.

• Knowledge Generated

• No statistically significant differences were observed in the outcomes of patients treated with pertuzumab plus trastuzumab and paclitaxel/nab-paclitaxel and those treated with pertuzumab plus trastuzumab and docetaxel. Subgroup analyses of patients treated with docetaxel who met CLEOPATRA's key eligibility criteria showed real-world progression-free survival and overall survival similar to those observed in the CLEOPATRA study.

• Relevance

• These results confirm those from the PERUSE study, which show similar outcomes between patients treated with docetaxel, paclitaxel, or nab-paclitaxel in combination with pertuzumab plus trastuzumab. They also highlight the importance of ensuring comparable patient populations when comparing clinical trial and real-world data.

Two studies using Ontario Cancer Registry data showed that patients with HER2-positive MBC treated with pertuzumab plus trastuzumab in routine clinical practice had shorter median OS than that observed in CLEOPATRA (39.2 and 43 months, respectively, v 56.5 months).^1,8,9 Discrepancies between clinical trial and real-world outcomes are often referred to as the efficacy-effectiveness gap (EEG).¹⁰ We conducted an analysis of patients with HER2-positive MBC treated with first-line pertuzumab plus trastuzumab and a taxane (paclitaxel/nab-paclitaxel/docetaxel) in routine US clinical practice. Here, we describe real-world outcomes for these patients and potential key contributors to the EEG.

METHODS

Objectives

The primary objective was to describe the real-world use and effectiveness of first-line paclitaxel versus docetaxel in patients with HER2-positive MBC. The secondary objective was to describe outcomes in a subgroup of patients receiving docetaxel who met CLEOPATRA's key eligibility criteria.

Patient Selection and Subgroups of Interest

This retrospective study used data from the deidentified, nationwide US, Flatiron Health electronic health record (EHR)–derived database. Figure 1 shows the study design. Patients who received an MBC diagnosis between June 2012 (the US launch date of pertuzumab) and January 2019, had HER2-positive disease, and were receiving first-line pertuzumab after MBC diagnosis, plus trastuzumab and a first taxane ± 28 (inclusive) days from first pertuzumab administration, were included. Exclusion criteria were >90 days between MBC diagnosis and first pertuzumab administration, use of oral or other injectable anticancer therapies (except endocrine therapies), or clinical trial enrollment between MBC diagnosis and first pertuzumab administration. Patients were grouped according to first taxane received (paclitaxel/nab-paclitaxel/docetaxel) with pertuzumab plus trastuzumab. As few patients were expected to receive nab-paclitaxel, they were included in the paclitaxel group for comparison with docetaxel-treated patients.

FIG 1.

Study design. ^aA cutoff of 2 years was selected as this approximated to 12 months of adjuvant treatment plus a 12-month disease-free interval from completion of systemic (neo)adjuvant therapy to metastatic diagnosis. ^bNumber of patients includes all patients with de novo MBC and patients with recurrent MBC with >2 years between EBC and MBC diagnosis. EBC, early breast cancer; ECOG PS, Eastern Cooperative Oncology Group Performance Status; EDM, enhanced data mart; EHR, electronic health record; IHC, immunohistochemistry; FISH, fluorescence in situ hybridization (per the ASCO/College of American Pathologists definition at the time the patient was treated); MBC, metastatic breast cancer; OS, overall survival; PFS, progression-free survival.

A subgroup of docetaxel-treated patients were defined using key CLEOPATRA eligibility criteria or their best approximation (Eastern Cooperative Oncology Group Performance Status [ECOG PS] of 0/1/missing; HER2 immunohistochemistry [IHC] 3+ or fluorescence in situ hybridization–positive disease per the ASCO/College of American Pathologists definition at the time of treatment; absence of CNS metastases; and ≤1 endocrine therapy between MBC diagnosis and first pertuzumab administration).³ Patients with <2 years between early breast cancer (EBC) and MBC diagnosis (if not de novo MBC), approximating to a disease-free interval of <12 months from completion of systemic (neo)adjuvant therapy (including HER2-targeted therapy) to metastatic diagnosis,³ were excluded. Although concurrent treatment with endocrine therapy was not permitted in CLEOPATRA, it was allowed in our study because of its use in routine clinical practice.

Study Variables and Outcomes

Baseline patient demographics, disease characteristics, planned treatment regimens, and dates of disease progression and last treatment administration were ascertained from Flatiron EHR data. Death was estimated as a composite mortality variable using data from EHRs, obituaries, and the Social Security Death Index.¹¹ Treatment history was abstracted for the purpose of this study. In patients with hormone receptor–positive disease, the cumulative incidence of endocrine therapy initiation within first-line treatment was estimated at 9 and 12 months. Patients who progressed, died, or were lost to follow-up before initiating endocrine therapy were censored on the date of progression or the last visit before loss of follow-up or death. Real-world PFS¹² (rwPFS; time in months from first pertuzumab administration to first documented evidence of disease progression or death from any cause, whichever occurred first and in the absence of an event, patients were censored on the date of their last clinical notes) was defined on the basis of clinicians' notes in EHRs referring to a distinct episode of tumor growth (radiologic or pathologic report or clinician determination) for the overall patient cohort and each subgroup of interest. Real-world OS¹³ (rwOS; time in months from first pertuzumab administration to death, regardless of cause and in the absence of an event, patients were censored at their last visit) was also defined for the overall patient cohort and each subgroup of interest. Time to last administration before discontinuation or death (TTLA) was measured as an exposure-derived end point. Treatment discontinuation was defined as missing >four doses of taxane while continuing to be followed up for >84 days or as death within 84 days after the last recorded taxane administration. As taxanes are administered at least once every 3 weeks, missing four doses was equivalent to an administration gap of >84 days.

Statistical Methods

Baseline patient demographics and disease characteristics were summarized for the overall patient cohort and each subgroup of interest. Categorical variables were reported as frequency and percentage, along with P values calculated using the Pearson's chi-squared test. Continuous variables were reported as medians, along with the IQR, and P values were calculated using the Mann-Whitney U test. Patients switching to a different taxane during treatment with pertuzumab plus trastuzumab were reported as frequency and percentage. Median rwPFS, rwOS, and TTLA were estimated using the Kaplan-Meier method, with 95% CIs. Hazard ratios (HRs) for patients treated with paclitaxel/nab-paclitaxel versus docetaxel were estimated using the Cox proportional hazards regression model. The inverse probability of treatment weighted (IPTW) method was used to estimate the average treatment effect for the treated (ATT) of initiating paclitaxel/nab-paclitaxel versus docetaxel, with the robust sandwich estimator used to estimate variance.¹⁴ Trimming (1%, 2%, and 3%) was applied as sensitivity analyses to remove observations with extreme weights; however, as the results were the same as those with no trimming applied, the presented model had no trimming, and the full sample was included in the analyses. HRs were weighted for prognostically important variables. Potential confounders and predictors of the outcomes were identified using directed acyclic graphs (available at DAGitty v3.0)¹⁵ in conjunction with a review of the subject (literature and expert opinion) and included age, body mass index, disease-free interval, number of metastatic sites, practice type, HER2 IHC 3+, hormone receptor status, and taxane treatment for EBC. Complete case analysis was performed. Differences in baseline characteristics included in the propensity score model between patients treated with pertuzumab plus trastuzumab and either paclitaxel/nab-paclitaxel or docetaxel were assessed by computing the standardized mean difference (SMD) before and after applying the IPTW-ATT method (Data Supplement [Table S1], online only). An SMD >10% was considered an indicator of covariate imbalance. Unweighted multivariable Cox models regressed over treatment and adjusted for all covariates included in the propensity score model were also used to estimate rwPFS and rwOS (Data Supplement [Table S2]).

RESULTS

Patients

A total of 1,065 eligible patients were included (Fig 1); 313 patients received paclitaxel/nab-paclitaxel and 752 received docetaxel (Fig 1). In the paclitaxel/nab-paclitaxel group, only 22 patients received nab-paclitaxel. The taxane administration schedule was weekly for 77% of paclitaxel/nab-paclitaxel–treated patients and every 3 weeks for 79% of docetaxel-treated patients. Table 1 shows baseline patient demographics and disease characteristics. The proportion of patients with de novo MBC was 54.6% (Table 1). At first pertuzumab administration, patients receiving paclitaxel/nab-paclitaxel were older and had a worse ECOG PS than patients receiving docetaxel (Table 1). There was also a higher proportion of patients with recurrent metastatic disease and those treated in the academic versus community setting in the paclitaxel/nab-paclitaxel group compared with the docetaxel group (Table 1). Although there were slight regional differences in chemotherapy selection, similar proportions of patients in each race group received paclitaxel/nab-paclitaxel or docetaxel (Table 1).

TABLE 1.

Baseline Patient Demographics and Disease Characteristics by Taxane Received

Treatment History

EBC Setting

Previous EBC taxane treatment was received by 292 patients; 98 patients received paclitaxel, two nab-paclitaxel, and 196 docetaxel (Table 2). Neoadjuvant endocrine therapy, trastuzumab, and pertuzumab was received by 12, 92, and 36 patients, respectively (Table 2). Adjuvant endocrine therapy, trastuzumab, and pertuzumab was received by 243, 237, and six patients, respectively (Table 2). Of the patients who received previous taxane treatment, 259 (88.7%) completed the treatment as planned. The proportion of patients who completed paclitaxel/nab-paclitaxel (87 of 99 [87.9%]) and docetaxel (172 of 196 [87.8%]) treatment was similar.

TABLE 2.

Treatment History by Taxane Received

MBC Setting

In the MBC setting, 51 patients received endocrine therapy before initiating pertuzumab plus trastuzumab (Table 2).

rwPFS

The median rwPFS in the paclitaxel/nab-paclitaxel group was 12.6 months (95% CI, 11.2 to 15.3) versus 15.0 months (95% CI, 13.4 to 17.0) in the docetaxel group (HR, 1.22; 95% CI, 1.02 to 1.44; P = .025; Fig 2A). After adjusting for potential confounders, median rwPFS was similar between the treatment groups (IPTW-ATT HR, 1.09; 95% CI, 0.9 to 1.3; P = .365).

FIG 2.

(A) rwPFS by taxane received, (B) rwOS by taxane received, and (C) TTLA by taxane received, before IPTW-ATT adjustment. IPTW-ATT, inverse probability of treatment weighted average treatment effect for the treated; rwOS, real-world overall survival; rwPFS, real-world progression-free survival; TTLA, time to last administration before discontinuation or death.

rwOS

The median rwOS in the paclitaxel/nab-paclitaxel group was 42.1 months (95% CI, 34.8 to 61.9) versus 49.4 months (95% CI, 44.0 to 60.2) in the docetaxel group (HR, 1.31; 95% CI, 1.05 to 1.64; P = .017; Fig 2B). After adjusting for potential confounders, the HR for rwOS was not statistically significant (IPTW-ATT HR, 1.23; 95% CI, 0.96 to 1.58; P = .101).

TTLA of Taxane

The median TTLA in the paclitaxel/nab-paclitaxel group was 3.7 months (95% CI, 3.3 to 3.9) versus 3.5 months (95% CI, 3.5 to 3.5) in the docetaxel group (HR, 0.87; 95% CI, 0.7 to 1.0; P = .072 [IPTW-ATT HR, 0.85; 95% CI, 0.73 to 1.0; P = .053]; Fig 2C).

Taxane Switch

Overall, 52 patients (4.9%) switched taxane treatment; 27 and 13 patients switched from docetaxel to paclitaxel and nab-paclitaxel, respectively, and 12 patients switched from paclitaxel to nab-paclitaxel.

Endocrine Therapy in the Overall Patient Population

Among the 746 patients with hormone receptor–positive disease, the cumulative proportion of patients initiating endocrine therapy with pertuzumab and trastuzumab within the first line was 53% and 59% at 9 and 12 months, respectively.

Subgroup Analyses in Docetaxel-Treated Patients Who Met Key CLEOPATRA Eligibility Criteria

A total of 615 patients (80%) who received docetaxel met key CLEOPATRA eligibility criteria. Of the 137 patients who did not meet the key criteria, the reasons were did not meet the criterion: Patients with >2 years between early breast cancer and metastatic breast cancer diagnosis (if not de novo metastatic breast cancer; n = 47 patients); did not meet the criterion: ECOG PS of 0/1/missing (n = 34); did not meet the criterion: HER2 IHC 3+ or fluorescence in situ hybridization–positive disease per ASCO/College of American Pathologists definition at the time of treatment (n = 54); did not meet the criterion: Absence of CNS metastases (n = 51); and did not meet the criterion: ≤1 endocrine therapy between metastatic breast cancer diagnosis and first pertuzumab administration (n = 6).

Baseline patient demographics and disease characteristics for the subgroup of patients who met key CLEOPATRA eligibility criteria are shown in Table 3, with similar data from the pertuzumab arm in CLEOPATRA.³ The median age of patients in the subgroup was 58 years (IQR, 49-66; Table 3). The number of patients who had received prior treatment with trastuzumab or with pertuzumab plus trastuzumab, respectively, was 95 (15.5%) and 18 (2.9%; Table 3). The number of patients in the subgroup who received endocrine therapy before initiating pertuzumab plus trastuzumab in the MBC setting was 24 (3.9%). Both the median age and the number of patients who had received previous HER2-targeted therapies was higher in the subgroup who met key CLEOPATRA eligibility criteria (58.0 years and 18.4% of patients, respectively) than in CLEOPATRA (54 years and 11.7%; Table 3).³ Other notable differences were the lower proportion of patients with IHC 3+ disease (62.6% v 87.1%) and the higher proportion of patients with hormone receptor–positive disease (73.1% v 47.0%) in the subgroup who met key CLEOPATRA eligibility criteria versus CLEOPATRA (Table 3).³ The median rwPFS in this subgroup was 16.9 months (95% CI, 15.0 to 20.9; Fig 3A), and the median rwOS was 57.8 months (95% CI, 49.1 to not reached; Fig 3B). rwPFS and rwOS for the patients who did not meet key CLEOPATRA eligibility criteria are shown in the Data Supplement (Fig S1).

TABLE 3.

Baseline Patient Demographics and Disease Characteristics of the Subcohort With Key CLEOPATRA Eligibility Criteria Applied and of the Pertuzumab-Treated Arm of the CLEOPATRA Study

FIG 3.

(A) rwPFS and (B) rwOS in the subcohort with key CLEOPATRA eligibility criteria applied. rwOS, real-world overall survival; rwPFS, real-world progression-free survival.

DISCUSSION

We describe for the first time the use of different taxanes with first-line pertuzumab plus trastuzumab in patients with HER2-positive MBC in routine US clinical practice. The results from this real-world cohort demonstrate the established use of paclitaxel/nab-paclitaxel in routine US care; 313 patients (29.4%) received paclitaxel/nab-paclitaxel alongside pertuzumab plus trastuzumab. Differences were observed between patients receiving paclitaxel/nab-paclitaxel or docetaxel, with patients receiving paclitaxel/nab-paclitaxel being older and having worse prognostic factors.

Median rwPFS and rwOS were shorter in patients in the paclitaxel/nab-paclitaxel group compared with the docetaxel group (12.6 v 15.0 months and 42.1 v 49.4 months, respectively). However, when the results were IPTW-ATT adjusted for potential confounders, there was no statistically significant difference in rwPFS and rwOS between the groups. Use of different taxanes with pertuzumab plus trastuzumab has similarly been investigated in PERUSE^6,7; however, extending the question to a real-world cohort allows confirmation of outcomes in patients with comorbidities (eg, worse ECOG PS or CNS metastases) or those receiving concurrent therapies (eg, concurrent endocrine therapy was permitted in our study to avoid selection bias for patients with hormone receptor–negative disease but was not permitted in CLEOPATRA³). Our results are consistent with those from PERUSE, which demonstrated similar PFS and OS in patients treated with pertuzumab plus trastuzumab and either docetaxel, paclitaxel, or nab-paclitaxel.^6,7

Although no safety data were collected, the real-world treatment duration of taxanes in this study was similar for patients receiving paclitaxel/nab-paclitaxel or docetaxel, which suggests a similar treatment tolerance between groups. Furthermore, median TTLA was similar between treatment groups, and the number of patients switching to a different taxane during pertuzumab plus trastuzumab treatment was low. However, the more favorable safety profile of paclitaxel and nab-paclitaxel versus docetaxel in PERUSE⁶ may mean that pertuzumab plus trastuzumab and paclitaxel/nab-paclitaxel is better tolerated in some patients.

The median rwPFS estimates of 12.6 and 15.0 months reported here align with previous results from Denmark (15.8 months)¹⁶; however, other studies have reported longer estimates (21.0-27.8 months).^17-19 Conversely, the median rwOS estimates of 42.1 and 49.4 months appear longer than previously reported (39.2-43.0 months),^8,9,16,20 possibly because of these previous studies having limited follow-up. Despite median OS being shorter than that in CLEOPATRA in both the pertuzumab and control arms in one real-world study, the significant OS benefit of pertuzumab observed in this study²⁰ highlights the importance of measuring comparative treatment effectiveness relative to an alternative treatment.

Median rwPFS in the subgroup of patients who met key CLEOPATRA eligibility criteria (16.9 months) was numerically longer than that observed in the overall patient cohort. The results from this subgroup, which represents approximately 80% of real-world patients, highlight the importance of including patients in clinical trials that will allow findings to be generalizable across the real-world population and of ensuring similar patient populations when comparing clinical trial and real-world data. Real-world outcomes should only be expected to align with those from clinical trials when the real-world patient population is selected to approximate that from the clinical trial. The median rwOS in the subgroup of patients who met key CLEOPATRA eligibility criteria (57.8 months) was numerically similar to that observed in CLEOPATRA (56.5 months),¹ which further supports the global generalizability of CLEOPATRA's results across the real-world population of patients with HER2-positive MBC.

Even after applying key CLEOPATRA eligibility criteria, patients in this subgroup were more likely to have hormone receptor–positive disease and less likely to have visceral metastases, compared with CLEOPATRA,³ and the use of endocrine therapy was common. These differences between real-world patients and those in CLEOPATRA were also observed in an Italian retrospective study¹⁹ and may lead to better real-world outcomes.²¹ Conversely, the subgroup who met key CLEOPATRA eligibility criteria was older and had fewer patients with IHC 3+ disease compared with the pertuzumab arm in CLEOPATRA,³ which may have promoted poorer outcomes. This may explain why a narrowed EEG remains and illustrates the need for examining individual patient-level data to fully explain the EEG. Previous exposure to HER2-targeted therapies was higher in our subgroup of patients who met key CLEOPATRA eligibility criteria compared with CLEOPATRA. This may be due to the real-world data (RWD) used in our study being more recent than data from CLEOPATRA. In addition, this analysis only reflects US data while CLEOPATRA was conducted across Asia, Europe, North America, and South America, where there may have been lower access to trastuzumab. The higher previous exposure to HER2-targeted therapies did not seem to negatively impact rwPFS and rwOS in the subgroup of patients who met key CLEOPATRA eligibility criteria after re-exposure.

In our analyses, patients receiving nab-paclitaxel were grouped with those receiving paclitaxel for comparison with docetaxel because of the small number of patients receiving nab-paclitaxel (n = 22). As weighting must be applied separately for each treatment group, splitting the weighted paclitaxel/nab-paclitaxel subgroup would not be valid. Therefore, no conclusions could be drawn from this study for nab-paclitaxel alone.

Strengths of this study were the inclusion of a heterogeneous patient population, the large representation of patients treated in the community setting (where most patients in the real world are treated and on which limited data have been reported to date), use of a curated data source (Flatiron) and rigorous covariate balance IPTW-ATT adjustment methods in the absence of random assignment, and assessment of previous (neo)adjuvant treatments. In addition, this study used a validated method for assessing rwOS¹³ and a method for assessing rwPFS that has been shown to replicate findings from randomized controlled trials.¹² Limitations included the restricted geographic representation of patients included in the Flatiron database, the limited representation of patients treated in the academic setting for comparison with CLEOPATRA, the potential for residual confounding after accounting for measured confounders, and population differences between the subgroup of patients who met key CLEOPATRA eligibility criteria in this study and patients included in CLEOPATRA (eg, the higher proportion of hormone receptor–positive disease, the use of endocrine therapy, and older age). Finally, as the majority of patients receiving paclitaxel/nab-paclitaxel and docetaxel received their taxane treatment weekly and every 3 weeks, respectively, there were more opportunities to observe progression events in the paclitaxel/nab-paclitaxel group.

In conclusion, there was no statistically significant difference in real-world outcomes between patients treated with pertuzumab plus trastuzumab and either paclitaxel/nab-paclitaxel or docetaxel. In addition, selecting real-world patients using key CLEOPATRA eligibility criteria resulted in rwPFS and rwOS similar to those reported for the pertuzumab arm of CLEOPATRA. In this way, RWD can be used to confirm findings and expand knowledge obtained through traditional clinical trials. Health care regulatory bodies are becoming increasingly accepting of RWD use in evidence generation. Indeed, the US Food and Drug Administration has introduced guidance for researchers using RWD in clinical investigations and has outlined their framework for implementing RWD in regulatory decision making.²² This should help to overcome the skepticism regarding RWD and facilitate more widespread adoption of it in drug development within sustainable and scientifically driven research frameworks, with new methodologies that address the current challenges of RWD use. RWD-based comparator arms in clinical trials,^23,24 for example, could accelerate anticancer drug development and save resources (human, time, and money).

AUTHOR CONTRIBUTIONS

Conception and design: Letizia Polito, Jinjoo Shim, Adam Knott, Yolande Du Toit, Eleonora Restuccia, Thibaut Sanglier, Sandra M. Swain

Collection and assembly of data: Letizia Polito, Jinjoo Shim, Thibaut Sanglier

Data analysis and interpretation: All authors

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Real-World First-Line Use of Pertuzumab With Different Taxanes for Human Epidermal Growth Factor Receptor 2–Positive Metastatic Breast Cancer: A Comparative Effectiveness Study Using US Electronic Health Records

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/op/authors/author-center.

Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).