Study Design and Patients. A phase II single arm, interventional clinical trial was completed to evaluate changes in molecular biology and tumor responses in patients treated with short-term NET. The primary objective of the study was to assess NET-induced changes in HER1-4 protein expression levels in tumors, and their association with post-NET cancer cell proliferation rate dichotomized Ki-67 positivity as low (< 10%) or high (≥ 10%). Secondary objectives included assessment of radiographic responses and other molecular markers. We enrolled women aged 18 years or older with HR+/HER2- BC with histologically confirmed operable and clinically or radiographically measurable lesions (cT1-T3, cN0, cM0). Estrogen and/or progesterone receptor positivity was defined as ≥ 1% nuclear staining by immunohistochemistry (IHC) and HER2/neu-negative by IHC or fluorescence in situ hybridization (FISH) according to the current ASCO/CAP guidelines21,22. Patients had to have an ECOG performance status of ≤ 2. Patients with bilateral BC and those with multifocal or multicentric disease were eligible. Patients were excluded if they had lymph node-positive or distant metastatic BC, purely noninvasive BC (i.e., ductal carcinoma in situ, lobular carcinoma in situ), HER2-positive BC, history of any malignancy except non-melanomatous skin cancer or carcinoma in situ of the cervix within 2 years, or were pregnant or actively breast feeding. All patients provided written informed consent. This study was performed in line with the principles of the Declaration of Helsinki. The Institutional Review Board and the Protocol Review and Monitoring Committee of the Medical College of Wisconsin approved the study.
Procedures. Patients were treated with NET for 4 weeks (+/- 1 week) prior to surgery, with dosing continuing until the day of surgery (+/- 2 days). Choice of NET regimen [tamoxifen or aromatase inhibitors (AIs)], based on menopausal status, medical conditions and patient preference, was at the discretion of the treating physicians. Standard daily dosing was used without any dose reductions or modifications (tamoxifen 20mg, anastrozole 1mg, letrozole 2.5mg, exemestane 25mg). Ovarian suppression with gonadotropin-releasing hormone (GnRH) analogues was allowed for premenopausal women. Cytochrome P450 2D6 (CYP2D6) inhibitors were prohibited with tamoxifen use. Initial diagnostic core biopsies were used for pre-treatment biomarker assessments and surgical tumor specimens were used for post-treatment assessments. Study Schema is shown in Fig. 1. Patients were followed for 30 days after surgery to record any adverse events from treatment. Adverse events were assessed according to Common Terminology Criteria for Adverse Events version 5.0. After tumor resection, patients were treated with radiation and/or chemotherapy according to standard-of-care treatment guidelines. Pre-NET biopsy specimens were used for OncotypeDx testing to aid in adjuvant chemotherapy decision making when indicated. ET will continue in the adjuvant setting for a period of at least 5 years. Patients will be followed for at least five years from completion of study to assess patient outcomes, i.e., ipsilateral, contralateral or distant recurrences.
Clinical Assessments. Patients underwent diagnostic breast mammography and ultrasonography at the time of diagnosis and again within 5 days prior to surgery. For clinical radiographic tumor measurements, the largest bi-dimensional measurements and when possible, three-dimensional measurements were recorded. Radiographic tumor responses were assessed using World Health Organization Response Evaluation Criteria in Solid Tumors (WHO): Complete response (disappearance of tumor), Partial response (≥ 50% decrease in the product of the bi-dimensional measurements of the tumor), No change (50% decrease in total tumor size cannot be established nor has a 25% increase in the size of the lesion been demonstrated), or Progressive disease (≥ 25% or greater increase in the total tumor size).
Pathology Assessments. Among the 37 patients, one patient had pathological complete response (pCR) and a second patient had too few residual cancer cells post-NET to measure HER proteins. Analyses of paired molecular tumor data on pre- and post-NET specimens were therefore limited to 35 patients. Standardized immunohistochemistry (IHC) protocols for estrogen receptor (ER) (6F11, Leica Biosystems, Buffalo Grove, IL), progesterone receptor (PR) (16, Leica Biosystems, Buffalo Grove, IL), HER1 (EGFR H11, Agilent, Santa Clara, CA), HER2 (4B5, Ventana, Roche Diagnostics, Indianapolis, IN), HER3 (HER3, Cell signaling, Danvers, MA) and HER4 (erbB4/HER4, MilliporeSigma, Burlington, MA) as well as HER2 dual probe FISH (PathVysion, Abbott, Abbott Park, IL) testing if HER2 IHC score was equivocal (2+), were performed in a College of American Pathologists (CAP)-accredited laboratory and all stains were visually evaluated by a board certified breast pathologist (JMJ) to assess biomarker status in the pre- and post-treatment tumor specimens. IHC for HER1 (EGFR), HER3 and HER4 was interpreted visually, with HER1 IHC interpreted as positive if there was membranous staining and HER3 and HER4 interpreted as positive if there was cytoplasmic and/or membranous staining, with 0 being negative (< 1%) and 1+, 2 + or 3 + scores matching the intensity and cutoffs (10% tumor staining) used by the current HER2 guidelines22. HER1-4 protein upregulation was defined as an increase of ≥ 1 in IHC score (ordinal 0,1,2,3), whereas downregulation conversely was defined as a decrease of ≥ 1 in IHC score.
Histopathological responses were assessed by change in tumor proliferation and cellularity. Tumor proliferation protein Ki-67 (MIB-1, Agilent, Santa Clara, CA) on pre- and post-treatment specimens was assessed by quantitative IHC. For Ki-67 quantification image analysis, QuPath (version 0.3.0, build 2021) software (Queen’s University Belfast, Belfast, Northern Ireland; https://qupath.github.io) was used for the cell detection, segmentation, objective classification (cancer and stroma) and determination of the percent of Ki-67-positive cancer cells. Tumor cellularity and tumor infiltrating lymphocytes (TILs) in pre- and post-treatment tumor specimens were assessed visually by pathologist according to Residual Cancer Burden guidelines23 and the International TILs Working Group guidelines24, respectively
Statistical analyses. We planned a sample size of 37 patients to achieve at least 80% power at significance level of 0.05, when testing the one-sided one sample hypothesis that the proportion of tumors with NET-associated upregulation of one or more HER proteins (HER1-4) is at least 50% versus the null hypothesis that the proportion is no larger than 30%. One patient had two ipsilateral HR + tumors with distinct histologies. Since two tumors in one patient cannot be regarded as independent, biostatistician blinded to marker data determined that the tumor with highest post vs. pre-NET HER1-4 protein change should be included in the analyses, with larger tumor size serving as a second selection criterion in case of a tie. Chi-square, Wilcoxon rank-sum tests and t-tests were performed where appropriate. Linear regression multivariable analysis was performed to evaluate the association of covariates (age, BMI, menstrual status, tumor grade, histology, pre- and post-NET changes in tumor size, Ki67, ER%, PR%, cellularity and TILs) with the primary outcome of changes in HER1-4 protein levels. All analyses were performed using SAS version 9.4 (Cary, NC).