Main content

    Da Costa International Funds for Breast Cancer Prevention

    Funded Da Costa Projects

    Genetics of Mammographic Density: A Pilot Study of a Whole Genome Association Study

    Dr. Elad Ziv
    • Dr. Karla Kerlikowske

    • Dr. John Shepherd

    • Dr. Steve Cummings

    Description of the Project:
    The aim of this study is to discover single nucleotide polymorphisms and haplotypes which are associated with mammographic density using a whole genome association approach. In particular, we will compare allele frequencies and haplotype frequencies among women with the lowest age and BMI adjusted mammographic density to women with the highest decile of age and BMI adjusted mammographic density. The sample size of this preliminary study will be 40 women, including 20 women from the high density group and 20 women from the low density group.

    Progress as of June 06:
    Our study requires use of genetic samples from women with high mammographic density and samples from women with low mammographic density sampled based on the SXA density measurement. To do this our plan is to obtain blood samples from the CPMC cohort from women with the highest age and BMI adjusted SXA measurement and women with the lowest age and BMI SXA measurement.

    To date we have completed several goals including (1) IRB approval of our protocol at UCSF and IRB approval at CPMC (2) negotiation and obtaining a price from Affymetrix that will allow us to increase the sample size to ~80-100 women total (40-50 from each extreme).

    Urinary 6-Sulphatoxy Melatonin Levels and Risk of Breast Cancer in Older Women

    Katie L Stone, PhD

    Description of the Project:
    Melatonin, a hormone secreted by the pineal gland, is involved in the circadian system. Melatonin has been shown to curtail cancer growth in animals and in cell cultures. Observational studies have found that levels of melatonin measured at night are lower in women with estrogen receptor (ER) positive breast cancer than in women with ER negative breast cancer or healthy controls. Two published studies of urinary melatonin levels and the risk of breast cancer produced conflicting results. We therefore proposed to conduct a case-cohort study within the multi-center Study of Osteoporotic Fractures (SOF), to test lower levels of 6-sulphatxymelatonin (aMT6s) using first void urine samples to determine whether they are associated with increased risk of breast cancer in older, postmenopausal women.

    Progress as of June 06:
    We were able to increase the sample to include 127 breast cancer cases and 384 non-cases (total sample size = 519). All assays of aMT6s and creatinine have been completed by the General Clinical Research Center at Oregon Health Sciences University.

    The preliminary analyses show no relationship between uncorrected or creatinine-corrected melatonin levels and risk of breast cancer in our case-cohort sample of older women. Sub-analyses to determine if urinary melatonin levels were associated with ER positive breast cancer were also negative. However, related to one of our secondary hypotheses, lower levels of uncorrected melatonin were found to be associated with greater risk of all-cause mortality. There also appears to be a strong positive relationship of creatinine-corrected melatonin levels and bone mineral density, a novel finding. Analyses of our specific aims will be completed by August.

    ATAC Mammographic Density Study

    Dr. Jack Cuzick

    Description of the Project:
    This study is testing the value of breast density for assessing the effect of anastrazole (an aromatase inhibitor) on breast density. We will assess the potential value of measurement of breast density for identifying women who respond best to treatment with aromatase inhibitors for reducing the risk of occurrence of breast cancer in the opposite breast after treatment for breast cancer. The Daniel and Phyllis Da Costa Fund is supporting the collection of mammograms and measurement of breast density.

    Progress as of May 2006:
    The collection of mammograms for this study is now well underway, concentrating initially on one of the largest ATAC centres in the UK. By focusing on several of the larger UK centres it is hoped that greater numbers of mammograms can be collected as efficiently as possible. This is also relatively cost effective, as once a system for collection is established at a particular hospital and the related administrative requirements are in place, it is easier to collect larger batches of films.

    It is intended that a minimum of 900 sets of sequential mammograms will be collected from the non-recurrent groups in order to assess the effect of tamoxifen and anastrozole (and combination) on mammographic breast density. In order to assess the predictive value of baseline mammographic density on the effectiveness of these hormonal therapies in terms of recurrence rate and new contralateral tumours, particular effort will be made to collect the films from a further 300 women whose cancer has recurred (200), or who have developed a contralateral tumour (100). To achieve these numbers it will be necessary to collect the original mammograms from outside the UK, which may prove more difficult.
    Once digitized, the mammograms are assessed for density on an ongoing basis by Dr. Ruth Warren, Consultant Radiologist at Addenbrooke's Hospital. However, it is intended that all mammograms will also be assessed using computer aided density measurements. The digital images are stored at the Cancer Research UK Centre for Epidemiology, Math and Statistics.

    The collection of mammograms from ATAC participants at two centres, based in Manchester, is now reaching completion. The scanned images of these mammograms have now been read for density by Dr. Ruth Warren, Addenbrookes Hospital, Cambridge, although it is intended that all images for collected for this study will later be assessed using computer aided density measurements.

    An initial analysis has also now been completed and an abstract will be submitted to the San Antonio Breast Cancer Symposium 2006.

    Hormone Levels and Mammographic Density; A Study of Twins

    Dr. John Hopper

    Description of the Project:
    Breast density is a strong risk factor for breast cancer, but the hormonal causes of high breast density have not been well studied. In this study, about 800 female and 600 sister twin pairs over age 40 have had a measure of breast density. Doctor Hopper and colleagues will collect blood for analysis of hormones that are likely to play an important role in the control of breast density, including IGF-1 and progesterone. This project also involves the development of a highly sensitive progesterone assay.

    Progress as of February 2006:
    Three batches of serum samples have been sent from The University of Melbourne to the Royal Marsden laboratory for analysis of hormone levels. This involves a total of 176 samples and the hormone analyses
    have been completed for the 102 samples sent in the first two batches (the third batch has just been dispatched). A total of 804 twin pairs and their sisters have already participated in the mammographic
    density study, and of these there are a further 258 subjects eligible for the study of hormone levels. A recruitment campaign is about to be launched to increase the number of twin pairs eligible for the
    hormone study.

    Development of a Breast Density Measure based on Electrical Conductivity (Bioimpedance)

    Dr. John Shepherd

    Description of the Project:
    Whole breast compositional density, measured via bioimpedance analysis (BIA), may be an equal or superior predictor of breast cancer risk to the mammographic density method. We are attempting to standardize breast positioning for all breast sizes and automate the BIA electrode placement by compressing the breast in a similar fashion as mammography. This preliminary study proposed to develop the in vivo device and method for measuring breast bioimpedance and to determine the reproducibility of the bioimpedance. We will extend this prototype for use in vivo; we hope to perform an efficient case/control cross sectional study of 25 women who recently have been diagnosed with cancer compare to 25 controls.

    Progress as of May 2006:
    Whole breast compositional density, measured via bioimpedance analysis (BIA), may be an equal or superior predictor of breast cancer risk to the mammographic density method. We are attempting to standardize breast positioning for all breast sizes and automate the BIA electrode placement by compressing the breast in a similar fashion as mammography. This preliminary study proposed to develop the in vivo device and method for measuring breast bioimpedance and to determine the reproducibility of the bioimpedance. Collection of BIA data is underway.

    Sex Hormones, Risk Factors and Risk of ER+ and ER- Breast Cancer in Postmenopausal Women

    Dr. Jennifer Lee
    Dr. Steve Cummings

    Description of the Project:
    A woman's endogenous level of estrogen (estradiol) is correlated with her risk of estrogen-sensitive (ER+) breast cancer. However, the role of estrogen in ER- cancer is not known. We will use serum and data from the Women's Health Initiative Observational Study Cohort to describe and compare the association between estradiol, testosterone, SHBG and risk of ER+ and ER- breast cancer. We will select 200 cases of ER+ and 200 cases of ER- cancer and a random sample of 600 controls from the cohort. We will use proportional hazards models to analyze combinations of sex hormone levels and risk factors for prediction of the two types of breast cancer.

    Progress as of May 2006:
    Cases and controls were selected and the samples of blood were pulled from the WHI repository. Half of the blood samples have been sent to Dr. Frank Stanzcyk's laboratory (USC) for sex hormone measurements. Results of the measurements are expected by August 2006.

    Breast Health Cohort

    Dr. Steve Cummings
    Dr. John Shepherd

    Description of the Project:
    This project will collect breast cancer risk factors and digitized mammograms, with phantoms, on all women attending CPMC Breast Health Center mammography. We will also collect serum and whole blood on those who volunteer. Cases of breast cancer will be ascertained and validated by SEER date collected through the ongoing SF Mammography Registry.

    Progress as of June 2006:
    Images on all patients undergoing digital mammography are being routinely collected. Over 35,000 images have been collected and about 8,000 women have provided serum and DNA that are archived for future research.

    Prospective Study of Predictive Value of a Proteomics Profile

    Dr. Jeff Tice

    Description of the Project:
    Current models for predicting breast cancer perform poorly in individuals and do not differentiate between estrogen receptor positive (ER+) cancer (the target of available preventive therapies) and estrogen receptor negative cancer. Serum proteomic profiling for cancer detection has generated tremendous excitement, but the studies have been criticized for selection bias, overfitting, and inadequate validation. We are undertaking a series of methodologically rigorous nested case-control studies to test whether proteomic profiles predict risk of breast cancer and to identify and validate serum proteomic patterns for prediction of breast cancer in postmenopausal women.

    Progress as of May 2006:
    SOF serum samples sent to EVMS lab, scraped, and returned to SOF repository. All proteomic spectra from the samples and pooled QC serum were generated by June 2005. The previously identified strong association between a proteomic pattern and the risk of ER+ cancer was not confirmed. No individual pattern was statistically significant. Additional analyses using additional chemical surfaces did not improve the strength of the association in validation data sets, however, non of the earlier strong associations have been validated. It appears that proteomic patterns, by the methods used in this study, are not useful for prediction of breast cancer in older postmenopausal women. Initial draft of negative results is in process.

    St. Gallens Meeting and St. Gallens Working Group

    Dr. Richard Santen

    Description of the Project:
    This grant funded a planning meeting which brought together 12 experts on risk factors for breast cancer to plan a project that will result in the development of a comprehensive breast cancer risk tool. The meeting took place at the Hotel Einstein in St.Gallens, Switzerland on January 26th, 2005.

    Progress as of May 2006 2005:
    The group agreed on a plan to obtain data on risk factors, sex hormone concentrations, and breast density from all major prospective cohort studies of breast cancer. The cohorts have been identified and data collection form has been developed. The feasibility of pooling the data remains to be determined. A report of the meeting has been submitted for publication.


    Estrogen Conference

    Dr. Mitch Dowsett
    Dr. Jonathan Middle

    Description of Project:
    The analysis of plasma estrogens is a critical part of studies to create a more rational algorithm for the assessment of breast cancer risk in postmenopausal women. Unfortunately the large majority of assays in use are particularly inaccurate in the postmenopausal range. This conference therefore brought together analysts, QA leaders, epidemiologists and representatives of diagnostic companies with representation from UK and US to assess the potential for creating more widespread accurate analysis of estrogens.

    Progress as of December 2005:
    The degree of inaccuracy of routine immunoassays was more striking than most were aware. It was concluded that tandem GCMS was the current gold standard and that any immunoassays for risk evaluation should be expected to closely approach this standard. While the diagnostic representation recognized the need for the first time it seems likely that the demand for better testing will need to be more vocal and widespread before they will focus on this and derive improved assays.

    UK Mammographic density review

    Dr. Mitch Dowsett

    Description of project:
    Mammographic density (MD) is expected to become an increasingly important component of risk evaluation particularly within screening units. There are however several methods for assessing MD which requires comparison of the methodologies before a method could be chosen for routine application. This meeting therefore brought together investigators from the 4 centres in the UK most involved in taking forward this technology - St Georges/Marsden, CRUK centre at St Bartholomews, Manchester, Cambridge together with Dr. John Shepherd from San Francisco.

    Progress as of December 2005:
    A strong will for collaboration was expressed and an outline plan was developed. This will be formalised into a stepwise program of research that will require application to funding agencies, including The Daniel and Phyllis Da Costa Fund, for support.

    Estradiol, testosterone and benefit of treatment with tamoxifen

    Dr. Mary Beattie

    Description of the Project:
    The P1 (Breast Cancer Prevention) Trial found that tamoxifen reduced the risk of invasive breast cancer by about 50% in women at high risk of breast cancer. This study tests the hypothesis that women who had the highest levels of estrogen or testosterone have the greatest reduction in risk of breast cancer during treatment with tamoxifen. The study will measure estradiol and testosterone levels in blood that was collected and stored before treatment was given. The original P1 Trial was conducted by the NSABP and supported by the National Cancer Institute. Dr. Cummings (Fund Director) helped design the study and is supporting the analysis and publication of the results.

    Completed July 2005:
    Surprisingly, this study found no significant association between levels of estradiol or testosterone and subsequent risk of ER+ breast cancer. Consequently, there was no greater reduction in risk of breast cancer in women with high levels of estradiol with tamoxifen. Thus, estradiol levels should not be used to identify women most likely to benefit from tamoxifen for prevention of breast cancer. Abstract of results were presented to San Antonio Breast Cancer meeting, December 2004. Article has been published by JNCI: 2006 Jan 18;98(2):110-5.