Anastrozole Dose Escalation for Optimal Estrogen Suppression in Postmenopausal Early-Stage Breast Cancer: A Prospective Trial

The third-generation aromatase inhibitors (AI) anastrozole, exemestane, and letrozole all play a major role in the adjuvant therapy of postmenopausal women with early-stage estrogen receptor-α–positive (ER+) breast cancer (1–3). The mechanisms by which all three AI inhibit tumor growth are through inhibition of aromatase, an enzyme encoded by the CYP19A1 gene and decreased estrogen production from androgenic precursors (4–6). Although all three AI are potent in terms of inhibition of in vivo aromatization (7–9), letrozole is the most potent (10). However, large phase III adjuvant clinical trials do not indicate any difference in efficacy among the three AI. Specifically, the MA.27 trial showed no significant difference between anastrozole and exemestane (11) in terms of disease-free survival (DFS), and the FACE trial showed no difference between anastrozole and letrozole for DFS and overall survival (12).

We have previously demonstrated that there was substantial interindividual variability in the metabolism of anastrozole and its pharmacodynamic effects on estrone (E1) and estradiol (E2) levels (13). To explore whether the degree of estrogen suppression was associated with breast cancer recurrence in women receiving adjuvant AI, we subsequently conducted a matched case-control study of postmenopausal women with ER+ breast cancer receiving adjuvant AI in the MA.27 (anastrozole and exemestane) and PreFace (letrozole) trials. The primary aim was to assess whether E1 and/or E2 levels after 6 months of treatment were associated with an increased risk of disease recurrence within 5 years of AI initiation [early breast cancer (EBC) event; ref. 14]. We found a statistically significant threefold increase in the risk of an EBC event for those taking anastrozole with both E1 ≥1.3 pg/mL and E2 ≥0.5 pg/mL, but a similar association was not found for exemestane or letrozole. Of note, in the multivariate analysis assessing E1 and E2 together, conditional logistic regression found that EBC events for women with only one of either E1 <1.3 pg/mL or E2 <0.5 pg/mL did not differ significantly from those with both E1 <1.3 and E2 <0.5 [OR = 1.05; 95% confidence interval (CI), 0.69–1.58; P = 0.83]. Thus, adequate estrogen suppression (AES) was achieved if E1 and/or E2 levels were below these designated thresholds.

In this investigation, we performed a prospective pharmacodynamic study to evaluate the impact of standard-dose anastrozole (1 mg/day, ANA1) on E1 and E2 levels and among those with inadequate estrogen suppression (IES, E1 ≥1.3 pg/mL and E2 ≥0.5 pg/mL), to evaluate the safety and pharmacodynamic efficacy of high-dose anastrozole (10 mg/day, ANA10). ANA10 was chosen, as prior trials evaluating ANA1 and ANA10 showed them to be equipotent in suppressing estradiol and comparable in terms of clinical outcomes and tolerability (15–18). The primary objective was to estimate the proportion of women who have adequate E1 and/or E2 suppression after 8 to 10 weeks of adjuvant ANA10, having had inadequate E1 and E2 suppression after 8 to 10 weeks of ANA1.

Postmenopausal women with primary stage I to III ER-positive/HER2-negative breast cancer were eligible upon completion of all planned breast and axillary surgery, adjuvant radiotherapy, and adjuvant chemotherapy. For those without prior bilateral oophorectomy and age <60 years, amenorrhea for longer than 12 consecutive months and follicle-stimulating hormone (FSH) level in the postmenopausal range were required; however, for those who received (neo)adjuvant chemotherapy, this criterion was required to be met at the time of diagnosis. ER+ disease was defined per American Society of Clinical Oncology/College of American Pathologists guidelines as ≥1% of cells with positive nuclear staining (19). Additional registration criteria included adequate blood counts and chemistries, Eastern Cooperative Oncology Group performance status 0 to 2, no prior ovarian function suppression use, no prior history of contralateral ductal carcinoma in situ or invasive breast cancer, no prior history of prevention therapy with an AI or selective estrogen receptor modulator, and no concurrent active malignancy or history of malignancy within 3 years of registration.

Patients were recruited from all Mayo Clinic tertiary campuses (Rochester, MN; Jacksonville, FL; and Scottsdale, AZ) and registered to receive 8 to 10 weeks of anastrozole administered as 1 mg orally once daily (cycle 1: ANA1). Blood samples were taken before starting ANA1 and at the completion of ANA1 to ascertain plasma E1 and E2 levels. If E1 was <1.3 pg/mL and/or E2 <0.5 pg/mL after ANA1, the patient stopped participating in the study, having achieved AES. Further treatment was at the discretion of the patient and medical team. Those with IES, defined as E1 ≥1.3 pg/mL and E2 ≥0.5 pg/mL, after 8 to 10 weeks of ANA1, could either withdraw from the study or be re-registered to receive 8 to 10 weeks of anastrozole administered as 10 mg orally once daily (cycle 2, ANA10) to evaluate whether an increased dose of anastrozole would result in AES. Regardless of post-ANA10 E1 and E2 levels, all patients were then switched to 8 to 10 weeks of letrozole administered at 2.5 mg orally once daily (cycle 3). The sequence of ANA10 to letrozole was selected as ANA10 is not FDA approved, and letrozole would be the logical choice for an alternative AI in women with IES on ANA1 given its potency (10). At completion of letrozole, E1 and E2 levels were determined. Patients then ended their participation in the study, and further treatment was at the discretion of the patient and medical team.

This protocol was approved by the Mayo Clinic Institutional Review Board (19-006637) and registered at ClinicalTrials.gov (NCT04294225). All participants provided signed written informed consent before registration. The Mayo Clinic Comprehensive Cancer Center Data Safety and Monitoring Board monitored the trial every 6 months. The study was conducted in accordance with the ethical guidelines of the Declaration of Helsinki.

Adverse events (AE) were documented using Common Terminology Criteria for AE v.5. A safety run-in period was conducted for the first six patients who began ANA10. If two or more of the first six patients enrolled in the safety run-in experienced a dose-limiting toxicity (DLT), as defined in Supplementary Table S1, that was definitely, probably, or possibly related to ANA10, enrollment would be halted to allow the study team to review toxicity data thoroughly and make decisions about changes to the protocol. If not, enrollment would continue.

Pre‐ and post‐AI treatment E1 and E2 levels were measured by Clinical Laboratory Improvement Amendments–approved LC/MS-MS assays in the Immunochemical Core Laboratory at Mayo Clinic. Details of the methodology have been published (13, 20). Intra-assay coefficients of variations (CV) for E1 were 17.8%, 7.5%, and 6.1% at 0.30, 0.50, and 0.84 pg/mL, respectively. Intra-assay CV for E2 were 11.8%, 7.3%, and 6.0%, at 0.23, 0.50, and 0.74 pg/mL, respectively. Interassay CV for E1 were 12.0%, 9.5%, and 7.9% at 0.25, 0.51, and 0.85 pg/mL, respectively. Interassay CV for E2 were 10.8%, 8.5%, and 6.9% at 0.29, 0.50, and 0.77 pg/mL, respectively. The lower limit of quantification (LLQ) was 1.0 pg/mL for E1 and 0.3 pg/mL for E2.

Anastrozole and its metabolites were measured in the laboratory of Zeruesenay Desta, Ph.D., at Indiana University School of Medicine, using ultrahigh performance LC/MS-MS assays. Details about sample preparation, LC/MS-MS methodology, and calibration curve preparation have been published (13). The calibration standards and quality controls were judged for batch quality based on the FDA guidance for the industry about bioanalytical method validation. The LLQ of anastrozole was 0.07 ng/mL. The linearity of anastrozole was in the range of 0.24 to 500 ng/mL. The assay demonstrated an intraday precision of 7.1% CV and an interday precision of 14.5% CV across the tested concentration ranges and quality controls. The accuracy of the assay ranged from 92.9% to 107.1% intraday and from 92.9% to 115% interday.

The primary aim of this study was to assess whether patients with IES after 8 to 10 weeks of ANA1 would achieve AES after receiving 8 to 10 weeks of ANA10. Two analytic approaches were taken to address this objective. The first approach focuses on the set of eligible patients who documented taking at least 80% (45–84 days) of the planned dose of ANA10 (adherent cohort), and the second approach focuses on all eligible patients who started an 8- to 10-week course of ANA10 (intention-to-treat cohort). Patients who did not provide a blood sample after discontinuing ANA10 were considered not to have achieved AES.

Specifically, the study was designed to assess whether the proportion of patients in the adherent cohort who achieve AES with ANA10 after IES with ANA1 was at most 25% against the alternative that it was at least 50%. With a sample size of 29 patients in the adherent cohort, a one-sided binomial test of proportions with a significance level of 0.05 had an 87% chance of rejecting the null hypothesis that the AES with ANA10 is at most 25% when the true AES with ANA10 is at least 50%. Specifically, the null hypothesis was rejected if 12 or more among these 29 patients achieve AES with ANA10. A 90% binomial CI was constructed for the AES rate for each treatment (to be consistent with the one-sided α level of our hypothesis test).

Secondary objectives were as follows: to estimate the proportion of women with IES after 8 to 10 weeks of adjuvant ANA1; to estimate the proportion of women with IES after 8 to 10 weeks of adjuvant ANA1 that persisted after 8 to 10 weeks of adjuvant ANA10; to examine the toxicity profile of ANA1, ANA10, and letrozole over each 8- to 10-week cycle of treatment; to examine concentrations of anastrozole at both the ANA1 and ANA10 dose levels; and to examine E1 and E2 levels in patients receiving letrozole (following ANA10).

Data lock occurred on November 30, 2023. Analyses were conducted using SAS 9.4.

The data generated in this study are available within the article and its Supplementary Data Files. A structured repository of data does not exist. Data are available upon request to the corresponding author. T.C. Haddad received the required authorization from the Data Review Committee of the author’s institution. All data requests must comply with the Personal Information Protection Act to ensure the privacy and confidentiality of patient data. The corresponding author T.C. Haddad will provide guidance on the privacy-preserving procedures required for data access.

The trial began on April 28, 2020. Enrollment was discontinued on June 1, 2022, before obtaining the planned 29 patients with IES after 8 to 10 weeks of ANA1, as the threshold for rejecting the null hypothesis that the AES rate after 8 to 10 weeks of ANA10 among those with IES after 8 to 10 weeks of ANA1 was 25% at most.

A total of 161 patients were registered. Among them, three patients withdrew consent before starting protocol treatment, and two patients were ineligible (Fig. 1 CONSORT diagram). Patient and disease characteristics of the 156 eligible patients who started ANA1 are presented in Table 1. Pre-ANA1 E1 and E2 levels were such that 152 (97.4%) patients had E1 ≥1.3 pg/mL (high E1) and E2 ≥0.5 pg/mL (high E2); 1 (0.6%) had E1 <1.3 pg/mL (low E1) and E2 <0.5 pg/mL (low E2); 1 (0.6%) had high E1 and low E2; and 2 (1.3%) patients did not provide a pre-ANA1 blood sample.

Among the 156 eligible patients who began ANA1 treatment, 131 (84.0%) patients were adherent [documented taking at least 80% (for 45–84 days) of the planned dose] and provided a blood sample for E1 and E2 testing. One patient (0.6%) was adherent but did not provide a blood sample for E1 and E2 testing because of side effects. Fifteen patients (9.6%) returned for their post-ANA1 visit without their pill diary but did provide a blood sample for E1 and E2 testing, and nine patients (5.8%) discontinued ANA1 because of toxicity (four patients) or refusal (five patients).

IES was observed in 44 patients post-ANA1 because of the persistence of high E1 and high E2 levels (43 patients) or an increase from low E1 and low E2 levels pre-ANA1 to high E1 and high E2 levels post-ANA1 (one patient).

AES was observed in 102 patients post-ANA1 due to several hormone level changes: a decrease from high E1 and high E2 levels pre-ANA1 to low E1 and low E2 levels post-ANA1 (38 patients); decrease from high E2 levels pre-ANA1 to low E2 levels post-ANA1 but persistent high E1 levels post-ANA1 (59 patients); decrease from high E1 levels pre-ANA1 to low E1 levels post-ANA1 but persistent high E2 levels post-ANA1 (four patients); or persistence of high E1 and low E2 levels post-ANA1 (one patient).

The degree of estrogen suppression could not be determined in 10 patients who did not provide a post-ANA1 blood sample. Thus, estrogen suppression post-ANA1 was adequate in 102 (65.4%), inadequate in 44 (28.2%), and not assessed in 10 (6.4%) patients. Among the 132 patients of the adherent cohort, AES was achieved in 91 (68.9%) patients, IES occurred in 40 (30.3%) patients, and estrogen levels were not assessed in one (0.8%) patient.

The patient and disease characteristics of the patients by estrogen suppression status post-ANA1 are provided in Table 1. These two patient cohorts were similar in terms of age, body mass index (BMI) classification, and E1 levels at registration. However, pre-ANA1 E2 levels among patients with AES post-ANA1 [median, 29; IQR (25th–75th percentile): 21–43] were significantly lower than pre-ANA1 E2 levels among patients with IES post-ANA1 [median, 34; IQR (25th–75th percentile): 24–51; Wilcoxon rank-sum test P = 0.0007]. E1 and E2 levels pre- and post-ANA1 are provided in Fig. 2A and B.

By study design, 122 patients did not proceed to the ANA10 phase of the trial because of AES (102 patients) or because of early discontinuation of ANA1 (nine patients) or refusal to undergo blood draw post-ANA1 (one patient). Eleven patients with IES after ANA1 did not proceed to the ANA10 cycle because of refusal (five patients) or toxicities (six patients). Thus, 33 patients with IES (30 adherent to ANA1) proceeded to the ANA10 portion of the trial.

Supplementary Table S2 provides the toxicities which led to discontinuation of ANA1 or refusal to switch to ANA10 following a finding of IES.

Per protocol design, tolerability was assessed among the first six patients treated with ANA10. None of these six patients reported a DLT after 28 days of treatment. Supplementary Table S1 provides the definition of DLT. Thus, patients with IES after ANA1 were allowed to switch to ANA10 with continued safety monitoring.

Of the 33 patients who began the ANA10 cycle, 25 patients were adherent (reported taking ANA10 for 45–84 days) and underwent post-ANA10 E1 and E2 testing. One patient returned unopened medication and declined to continue on the study, and seven patients discontinued ANA10 early because of toxicity (one patient experienced grade 2 urinary tract infection and one patient experienced grade 2 breast and abdominal pain with grade 1 vaginal hemorrhage and palpitations) and patient decision (five patients) as per Fig. 1. Toxicities regardless of attribution for all 32 patients who began ANA10 are reported in Table 2. No severe (Common Terminology Criteria for AE grade 3–5) toxicities were reported.

Considering all 32 eligible patients who began ANA10, AES was achieved in 19 (59.4%) patients, IES occurred in six patients (18.8%), and estrogen suppression status was not assessed in seven patients (21.9%). Thus, the post-ANA10 AES rate was 59.4% (90% CI, 43.4%–74.0%) among these 32 patients.

Among the 25 patients in the ANA10 adherent cohort, AES was achieved in 19 (76.0%; 90% CI, 58.1%–89.0%) patients. and IES persisted in six patients (24.0%). The changes in E1 and E2 levels observed among the 19 adherent patients with AES were such that there was a decrease to E1 low and E2 low levels in 13 patients, a decrease to E1 low but persistent high E2 in two patients, and persistent high E1 but low E2 in four patients (Fig. 2C and D).

All patients who completed ANA10 were eligible to switch to letrozole 2.5 mg/day for 8 to 10 weeks by study design. One patient with IES did not enter the letrozole cycle of the trial because her FSH level was in the premenopausal range.

Of the 24 patients who began letrozole, 22 (91.7%) patients were adherent (reported taking letrozole for 45–84 days) and underwent post-letrozole E1 and E2 testing. One (4.2%) patient recorded taking letrozole for 45 to 84 days but E2 could not be ascertained because of laboratory error, and one patient (4.2%) did not return her pill diary but underwent post-letrozole E1 and E2 testing. No patient discontinued letrozole early.

E1 and E2 levels at the completion of letrozole (Fig. 2C and D) among 19 patients who had post-ANA10 AES were such that 17 patients (16 adherent; one no pill diary returned) maintained AES, but two patients (both adherent) were found to have IES. Levels following letrozole among the five patients who had post-ANA10 IES were available for four patients (all adherent), and all four were found to have AES. Thus, 21 (91.3%) of the 23 letrozole-adherent patients maintained or attained AES.

Plasma drug concentration levels post-ANA1 were available for 129 of the 132 adherent ANA1 patients and were not found to differ between the 90 patients with AES and the 39 patients with IES (Wilcoxon rank-sum P = 0.5744; Fig. 2E). Additionally, anastrozole concentrations had a weak association with both E1 and E2 levels post-ANA1 (Spearman rank correlation: −0.13 and −0.30, respectively).

There were 24 ANA1-adherent patients who were also adherent to ANA10 and had post-ANA10 plasma drug concentration levels ascertained. All but three of these patients (87.5%) had a fivefold or higher increase in anastrozole after ANA10. The fold change in anastrozole plasma concentration levels post-ANA10 was not found to differ between those with AES post-ANA10 and those with IES post-ANA10 (Wilcoxon rank-sum P = 0.7389; Fig. 2F); between those with E1 low and E1 high post-ANA10 (Wilcoxon rank-sum P = 0.9533); or between those with E2 low and E2 high post-ANA10 (Wilcoxon rank-sum P = 0.1779).

During the development of the nonsteroidal AI anastrozole, doses of 1 and 10 mg were shown to be equipotent inhibiting aromatase and suppressing estrogen levels (7, 15). Both doses were associated with comparable clinical and safety outcomes, leading to the 1-mg dose becoming the standard of care (21). However, at an individual level, there is considerable variability in side effects and clinical outcomes, which prompted the question of whether the 1-mg dose is appropriate for all patients.

In this pharmacodynamic trial, we prospectively demonstrated that approximately 30% of postmenopausal women with operable ER+, HER2-negative breast cancer taking standard-dose anastrozole 1 mg daily experienced IES, as defined by levels of E1 ≥1.3 pg/mL and E2 ≥0.5 pg/mL. These findings were consistent with a prior evaluation of estrogen levels in patients taking anastrozole 1 mg daily for 4 weeks, in which E1 and E2 concentrations were above the lower limits of quantitation for 30% and 21% of patients, respectively (13). Although the clinical significance of this finding remains unknown, our prior matched case-control study (14) suggests IES after 6 months of ANA1 is associated with a threefold increased risk of early recurrence.

Although the proper threshold for suppressing E1 and E2 has not been prospectively validated, our research indicates that E1 and E2 levels may be critical biomarkers of prognosis and AI efficacy. Further strengthening this assessment are findings from the IBIS-II prevention trial that found anastrozole to provide a 49% reduction in the risk of a primary breast cancer event among postmenopausal, high-risk women (22). A post hoc secondary analysis of these data showed the anastrozole benefit to be confined to those women with baseline medium or high ratios of serum estradiol to sex hormone–binding globulin, whereas there was no significant effect in those with a baseline ratio in the lowest quartile (23). Additionally, in the neoadjuvant POETIC trial, higher baseline plasma estradiol concentrations were reported to be associated with a significantly greater response to nonsteroidal AI, as measured by the 2-week change in tumor Ki67 (24). Although the collective data highlight the emerging value of baseline E2 and other sex hormone levels on breast cancer prognosis and AI efficacy, dynamic changes in these levels after treatment initiation, as evaluated in our prospective trial, may improve their prognostic and predictive value.

Notably, in the present study, 76% of patients with IES post-ANA1 who subsequently adhered to ANA10 achieved AES. Consistent with prior clinical trials in postmenopausal women with metastatic breast cancer, the 10-mg dose of anastrozole was well tolerated in the adjuvant setting with a comparable toxicity profile to standard-dose treatment (17). Additionally, there were no DLT or grade 3 or higher AE observed.

AES on ANA10 was maintained in 17 of 19 (89.5%) patients after completing 8 to 10 weeks of letrozole. Importantly, all four patients with IES at both dose levels of anastrozole went on to achieve AES with letrozole. These findings may be due to letrozole being the most potent suppressor of both plasma and tissue estrogen levels of the three FDA-approved AI (9, 10). Although adjuvant letrozole was not found to be clinically superior to anastrozole in the FACE trial (12), it is notable that the trial was terminated prematurely because of lower-than-expected DFS events. Although the current study findings are not practice-changing, given the choice of available AI, letrozole may be preferred as it can overcome IES with anastrozole in most patients.

It is unknown whether the 10-mg dose of anastrozole will be considered for further development given the availability of other AI. However, there are no data beyond what is presented in this study on the pharmacodynamic effects of switching from anastrozole to another AI. It is also important to note that our preclinical studies suggest that anastrozole is a ligand for the ER, and at higher concentrations, it may induce ER degradation in addition to inhibiting aromatase (14).

BMI is known to have an impact on breast cancer risk and prognosis, and obese postmenopausal women are at increased risk for breast cancer, disease recurrence, and death when compared with women who have a normal BMI (25–27). It has been postulated that increased total body aromatization and elevated estrogen levels in obese postmenopausal women may contribute to this observation. Notably, BMI has been shown to have an impact on AI efficacy as an adjuvant treatment for breast cancer, particularly for anastrozole (28, 29). One trial demonstrated a nonsignificant but numerically lower E2 plasma level in nonobese patients compared with obese postmenopausal patients after 3 months of anastrozole or letrozole; however, serum FSH levels were significantly lower in obese patients compared with nonobese patients (30). In our study, we observed a comparable distribution of BMI among postmenopausal patients who did or did not adequately suppress E1 and E2, although the study was not powered to assess for this difference. Additionally, we utilized the ultrasensitive quantitation of estrogen levels by LC/MS-MS, which differs from methods in some prior studies and should be considered when evaluating findings across trials.

Of note were the weak associations anastrozole plasma concentrations had with both E1 and E2 levels post-ANA1. The findings in this study were consistent with our previous report such that there were large interindividual variations in plasma anastrozole concentrations and their effect on E1 and E2 concentrations in postmenopausal women treated with adjuvant anastrozole (31). We hypothesized that this interindividual variability may be due to genetic variability, as we previously identified a single-nucleotide polymorphism–based model that had excellent performance characteristics for predicting E1 and E2 concentrations above the thresholds used in the current study (32).

Strengths of this study include its prospective design and the evaluation of serial E1 and E2 levels in patients receiving two different nonsteroidal AI and two dose levels of anastrozole as adjuvant therapy for breast cancer. Additionally, there were minimal exclusion criteria, thus making the study cohort representative of a real-world setting. Several of our findings also prospectively validated prior study results—that 30% of patients taking adjuvant ANA1 experience IES, ANA10 is tolerable, letrozole is potent and can overcome IES with anastrozole, and there is substantial interindividual variability in anastrozole concentrations.

Although the study size was relatively small, it was sufficient to achieve the primary objective. As this trial was conducted at a single institution across its three geographically distinct campuses (in Minnesota, Florida, and Arizona) and the majority (90%) of trial participants were non-Hispanic White, our findings may not be broadly applicable to other health systems or underrepresented racial and ethnic minority patient populations (Supplementary Table S3). Although it would have been of interest to randomize women with IES on ANA1 to ANA10 or letrozole and evaluate clinical outcomes, as ANA10 is not currently FDA approved and sample size was not feasible, demonstration of its safety and tolerability, as well as its efficacy in achieving AES, was prioritized for this trial. Quantitation of estrogen levels by LC/MS-MS in our Immunochemical Core Laboratory, while currently the most sensitive method for detection (with LLQs of 1.0 pg/mL for E1 and 0.3 pg/mL for E2), is not routinely available clinically, which may limit the implementation of our trial findings in routine clinical care. However, Mayo Clinic plans to make this assay available clinically.

There are accumulating data that anastrozole results in IES in a substantial proportion of patients. Further study is required to validate the proper threshold for estrogen suppression in patients taking anastrozole across the spectrum of prevention, (neo)adjuvant, and metastatic breast cancer settings. In line with this, additional research is needed to determine the best approach to managing patients with IES, either by increasing the dose of anastrozole or switching to a different AI, and whether these approaches are associated with better suppression of tumor proliferation and growth and/or lead to a decrease in breast cancer recurrence.

T.C. Haddad reports grants from NCI during the conduct of the study, research funding to the institution from Takeda Oncology, and other support to the institution from Puma Biotechnology outside the submitted work. L. Wang reports grants from NCI during the conduct of the study. S. Chumsri reports research funding to the institution from Merck & Co., Pfizer, Salix Pharmaceuticals, Rebiotix Inc, Novartis, and BriaCell Therapeutics and other support to the institution from AstraZeneca, Daiichi Sankyo, Immunomedics, Biotheranostics, Novartis, Athenex, Syndax, Puma Biotechnology, Eisai, and Seagen Rebiotix Inc. M.P. Goetz reports grants from NCI during the conduct of the study and research funding to Mayo Clinic from Lilly, Pfizer, Sermonix, Loxo, AstraZeneca and ATOSSA Therapeutics; consulting fees to the institution from ARC Therapeutics, AstraZeneca, Biotheranostics, Blueprint Medicines, Lilly, Novartis, RNA Diagnostics, Sanofi Genzyme, Seattle Genetics, Sermonix, Engage Health Media, Laekna and TerSera Therapeutics; and personal fees (CME) from Research to Practice, Clinical Education Alliance, Medscape, MJH Life Sciences, Total Health Conferencing, and Curio Science. C.C. O’Sullivan reports research funding to the institution from Genentech, Bavarian Nordic, Sermonix, Seagen, Tesaro and Nference, and other support to the institution from AstraZeneca Enhertu Steering Committee, Medscape Presentation, Seagen HER2CLIMB-05 Clinical Study Steering Committee, Edith Sanford Presentation, Seagen Breast Cancer Advisory Board, AstraZeneca Dato-DXd Advisory Board, and AstraZeneca HER2 Steering Commitee. K.V. Giridhar reports other support to the institution from Lilly, Novartis, AstraZeneca, and Puma Biotechnology.

T.C. Haddad: Conceptualization, resources, data curation, formal analysis, supervision, funding acquisition, validation, investigation, methodology, writing–original draft, writing–review and editing. V.J. Suman: Conceptualization, resources, data curation, formal analysis, supervision, validation, investigation, methodology, writing–original draft, writing–review and editing. K.V. Giridhar: Investigation, writing–review and editing, analysis. K. Sideras: Investigation, writing–review and editing, analysis. D.W. Northfelt: Investigation, writing–review and editing, analysis. B.J. Ernst: Investigation, writing–review and editing, analysis. C.C. O’Sullivan: Investigation, writing–review and editing, analysis. R.J. Singh: Resources, data curation, formal analysis, validation, investigation, methodology, writing–review and editing. Z. Desta: Resources, data curation, formal analysis, validation, investigation, methodology, writing–review and editing. P.P. Peethambaram: Investigation, writing–review and editing, analysis. T.J. Hobday: Investigation, writing–review and editing, analysis. S. Chumsri: Investigation, writing–review and editing, analysis. R.A. Leon-Ferre: Investigation, writing–review and editing, analysis. K.J. Ruddy: Investigation, writing–review and editing, analysis. S. Yadav: Investigation, writing–review and editing, analysis. J.L. Taraba: Resources, supervision, investigation, writing–review and editing. B. Goodnature: Supervision, investigation, writing–review and editing. M.P. Goetz: Investigation, writing–review and editing, analysis. L. Wang: Conceptualization, resources, data curation, supervision, funding acquisition, investigation, writing–review and editing. J.N. Ingle: Conceptualization, resources, data curation, formal analysis, supervision, funding acquisition, investigation, methodology, writing–review and editing.

We are grateful to the patients who generously volunteered to participate in this study. We thank the Mayo Clinic investigators, research nurses and coordinators, laboratory technicians, and clinical care teams for their support of this clinical trial. This work was supported by the Mayo Clinic Breast Cancer SPORE (P50CA116201; belonging to funded authors: T.C. Haddad, V.J. Suman, M.P. Goetz, L. Wang, and J.N. Ingle), which receives funding from the NCI at the NIH and the George M. Eisenberg Foundation for Charities (belonging to the funded author J.N. Ingle).