This phase II trial evaluates response-guided low-dose tamoxifen for reducing breast density in women who are at higher than average risk for breast cancer. Increasing breast density is a well established risk factor for breast cancer. Tamoxifen is a selective estrogen receptor modulator. It works by blocking the effects of the hormone estrogen in the breast. Tamoxifen has been shown to reduce breast density, even at reduced dosages, and is approved for the prevention of breast cancer.
Age under 50 years and regular menstruation (most recent period within the past 3 months)
Age under 50 years and continuous hormonal contraception use and at least one intact ovary
Women who are not postmenopausal based on serum hormone levels. Women with estradiol > 30 pg/mL and follicle-stimulating hormone (FSH) < 30 IU/mL are eligible
A history of unilateral estrogen receptor (ER) positive ductal carcinoma in situ (DCIS) with local therapy completed (as determined by treating physician recommendation and patient acceptance) at least 1 month prior to study entry. (The untreated breast will be the study breast, for both imaging and optional biopsy)
BIRADS breast density category A
Use was less than 6 months duration in the past 5 years and not used in the 1 year prior to enrollment OR
Use was no greater than 2 months duration in the past 1 year and not used in the 6 months prior to enrollment
History of invasive breast cancer
Single group
Biopsy Procedure
Procedure/SurgeryUndergo biopsy
Biospecimen Collection
Procedure/SurgeryUndergo collection of blood samples
Mammography
Procedure/SurgeryUndergo mammography
Questionnaire Administration
Other interventionAncillary studies
Tamoxifen
DrugGiven PO
Response at any time point
Tamoxifen response is defined as absolute dense area reduction (aDAR) of \>= 10% on mammogram compared to baseline mammogram. A one sample exact binomial test for one proportion will be used, and the exact two-sided p-value will be reported. Response rate for the entire study population, as well as for the group of women who were dose-escalated, will be estimated and reported with the corresponding 95% confidence interval. Absolute and relative changes in dense area will be estimated using descriptive statistics (e.g., mean and standard deviation, or median and interquartile range), and, as a sensitivity analysis, a paired t-test or the signed rank test will be used to determine whether there was a reduction in dense area. Similar descriptive statistics will be performed in each dose sequence group for both the response rate and dense area changes.
Plasma levels of major tamoxifen (TAM) metabolites
Will assess the association of plasma levels of major tamoxifen metabolites with tamoxifen dose and breast density changes from baseline. Metabolite levels will be summarized using standard descriptive statistics at each time point, and change patterns examined using graphical tools for longitudinal data. Will also examine the association of TAM metabolite levels in plasma with aDAR over time using exploratory graphical analyses of aDAR versus (vs.) concurrent metabolite level by time point and dose level. The overall analytic approach to determine whether metabolite levels are associated with aDAR will use linear mixed models.
Longitudinal change in serum biomarkers of tamoxifen response
Will evaluate longitudinal change from baseline in serum biomarkers of tamoxifen response at each dose level: sex hormone binding globulin, insulin like growth factor 1 and C-reactive protein. Biomarker values and changes from baseline will be summarized using standard descriptive statistics at each time point by response status. Differences in baseline levels of these biomarkers between different dose sequence groups will be analyzed using linear regression models with biomarker value as the outcome, and dose cohort as the predictor. Will also use linear mixed models to examine biomarker change patterns over time.
Baseline dense area
Will assess the association of baseline dense area (continuous variable) with tamoxifen response. Will be compared using linear regression models between 5 mg responders and non-responders at 6 months, and between dose sequence groups.
Patient-reported symptoms
Will evaluate the impact of tamoxifen dose on participant-reported symptoms using the Breast Eight Symptom Scale. Will examine patient reported outcome (PRO) patterns over time using graphical tools for longitudinal data, and will use descriptive statistics to summarize PRO values at various time points for groups of patients based on their more recent TAM dose level. Changes in PRO from baseline will be summarized similarly. Linear mixed models will be used.
Breast tissue-based biomarkers
Will evaluate breast tissue-based biomarkers and their association with tamoxifen response in participants who consent to the optional research biopsies. Will be compared between 5mg responders vs. non-responders at 6 months using a two-sample t-test or Wilcoxon rank sum test. In addition, logistic regression models may be used to determine the joint association between multiple baseline characteristics and tamoxifen response to 5 mg at 6 months.
Single nucleotide polymorphisms (SNPs)
Will evaluate the association between SNPs that overlap between risk of breast cancer and dense area of breasts. Others that relate to efficiency of tamoxifen metabolism will be explored.
Change in breast cancer risk estimates
Will evaluate change in breast cancer risk estimates as assessed by an artificial intelligence tool and compare changes by dose group. Changes will be compared between dose sequence groups using a two-sample t-test. In addition, analysis of covariance models may be fitted with 12-month score as the outcome, and dose group and baseline score as predictors.
Click on the lead sponsor to view all of their trials.
