> Research & Grants
> Grants Program
> Research Grants
> Research Grants Awarded
DNA Methylation Profiles for Diagnosis, Monitoring and Residual Disease Detection in Breast Cancer
Background: Despite significant progress in fight against breast cancer it still remains a deadly disease. Efficient detection and evaluation of tumor growth are important to correctly decide on treatment strategy and to monitor patient’s progress during therapy. A number of molecular markers are already in use, including expression of estrogen and progesterone receptors, Her2/neu, cancer antigen 15-3, ceruloplasmin, maspin and others. However, there is still substantial need for sensitive, specific and efficient multiparametric method of tumor detection and assessment. In this application we propose to evaluate the utility of a newly-developed DNA-based assay for breast cancer detection and monitoring.
The assay is based on analysis of tumor-specific gene silencing through promoter methylation, and has been developed to utilize a minimal amount of heterogeneous starting material in a high-throughput manner. While invaluable for research purposes, existing procedures of methylation status analysis usually require considerable pre-purification of tumor tissue from surrounding stroma and blood vessels, which makes their use for diagnostic purposes impractical. In most cases, these methods are also labor-intensive and/or involve relatively large amounts of starting material, impeding their use for routine clinical application. Our assay allows quick evaluation of methylation status of over 50 cancer-related genes, requires less than one thousand cells to perform, and can be finished within one week. The assay has been tested with breast cancer cell lines and tumor DNA from formaldehyde-fixed paraffin-embedded tissue (FFPET). The assay is flexible, and can be adapted to include additional genes of clinical importance.
Objective/Hypothesis: We will evaluate methylation status of breast cancer-related genes to identify genetic markers of different stages of breast cancer - from atypical hyperplasia (AH) to invasive ductal carcinoma (IDC) - and to correlate these findings with estrogen and progesterone receptor expression. We hypothesize that each stage of breast cancer development has a distinctive methylation profile of 50 genes (MP50) that will be used. We further hypothesize that such MP50’s are different for tumors with poor and favorable prognosis, and can be used for outcome prediction. Finally, we hypothesize that depending on tumor stage similar or even identical MP50 can be obtained from cell-free plasma DNA (cpfDNA) of breast cancer patients in order to continuously monitor tumor status and thus efficiency of therapy.
Aim 1. To determine MP50 for FFPET of normal breast (NB), AH, ductal carcinoma in situ (DCIS) and IDC using 20 estrogen- and progesterone receptor-positive and 20 estrogen- and progesterone receptor-negative samples from each group.
Aim 2. To evaluate feasibility of methylation profiling for cfpDNA from samples obtained from the same groups of patients.
Aim 3. To compare MP50 obtained in Aims 1 and 2, to identify consistent molecular markers specific for each group and to verify their validity by performing blinded assay on 20 samples of breast cancer and normal breast tissue.
Study design: Genomic DNA from FFPET and plasma will be isolated by the Pathology Core. MP50 for each group will be determined using previously developed method (manuscript submitted). Comparison of the data will be performed in collaboration with Biostatistical Group.
Potential Outcomes and Benefits of the Research: This study will provide the first comprehensive information about methylation of over 50 genes in breast cancer at different stages of disease. The study contains two parts: translational (aims 1 and 3) and developmental (aim 2). Translational component will provide data on molecular markers that can be directly used for cancer diagnosis, prognosis and monitoring. Developmental part, if successful, will allow the same assay to be performed in a minimally invasive manner.
Background. Despite significant progress recently in detection and treatment breast cancer still remains the major cause of death in women. One of the challenges for breast cancer therapy is the changing nature of the disease, which has to be constantly checked in order to appropriately adjust the treatment. The best known example of a biological marker used for cancer therapy selection is the estrogen receptor presence, which can change at different stages of disease.
In this proposal we will use a novel method of DNA analysis to identify diagnostic features for breast cancer detection and classification. There are many reasons to use DNA: first, DNA is chemically different in normal and malignant cells, and that difference is localized to specific regions. Certain DNA sequences in malignant cells contain many modified nucleotides (cytosines) that are tagged with methyl groups (methylated DNA). When this methylated DNA is present within regulatory regions of genes, such genes are no longer expressed, and if one of them is a tumor suppressor gene, the cell can easily become cancerous. There are several caveats with using methylated DNA for diagnosis. A major issue is that different genes are methylated in different tumors but there is no single gene that is always methylated in every tumor. Practically it means that if we concentrate on a single gene which is methylated in 70% of tumors, we will miss the other 30%. To overcome this quandary we will analyze not one but over 50 cancer-related genes in the same tumor sample, establishing a methylation profile of breast cancer.
There is another advantage in using DNA for cancer diagnosis, and that is the possibility to use the polymerase chain reaction or PCR. With PCR we can greatly increase – amplify - DNA available for analysis, which in practical terms means that we can start with truly miniscule amounts of diseased tissue in the first place. In our experience the DNA obtained from twenty cells is sufficient to do the assay, and we believe that we have not yet reached our lower limit.
In addition, the assay that we developed is fast: we can generate methylation profile of fifty genes per sample in a week. The assay can use patient’s samples where tumor cells are mixed with normal, and it can be done even using archived tissue that has been stored for many years.
Objective/Hypothesis. In this project we will use breast cancer and normal tissues of different groups of patients to build the methylation profiles characteristic for a specific stage of the disease. We will identify groups of genes specifically methylated in different types of breast cancer and will confirm that we can correctly distinguish between those types in a blind assay. We will also determine if DNA from patient’s blood will give us the same or similar result, because then the assay will require only a blood draw and will be useful for routine monitoring of patients during therapy.
Aim 1. To determine methylation profiles for different types of breast cancer.
Aim 2. To compare methylation profiles from blood to methylation profiles from tissue.
Aim 3. To confirm diagnostic results by doing blind assays of breast cancer methylation.
Study design. Normal DNA and DNA from tissues and blood of breast cancer patients will be provided by the Pathology Core. We will use tissues from patients with estrogen- and progesterone receptor-positive and patients with estrogen- and progesterone receptor-negative breast cancer diagnosed with different stages of cancer. We will also analyze DNA from blood in these same groups. Methylation profiles will be determined using M3A method (microarray-mediated methylation assay).
Potential Outcomes and Benefits of the Research. We will develop an exquisitely sensitive (no more than twenty cells per assay) approach to diagnose, classify and monitor breast cancer. Because of its sensitivity the assay will be useful at the earliest stages of disease when no currently available method is helpful; as we accumulate more data we will also establish correlations between methylation profiles with disease progression, response to therapy, and probable outcomes. This will help oncologists to design optimal therapy for each individual patient. When a blood-based assay is developed, oncologist will, with minimal discomfort or inconvenience to the patient, be able to monitor changes in tumor behavior throughout the treatment, to predict development of resistance to specific drugs and thereby modify therapy accordingly.