Altering the Microenvironment to Promote Dormancy of Metastatic Breast Cancer Cell in a 3D Bone Culture System

The 5-year cure rate for breast cancer is high, provided it does not metastasize. However, it is estimated that 85% of individuals with advanced disease harbor metastases. New, sensitive technologies reveal that as many as one-third of patients diagnosed at M0 tumor-node-metastasis tumor stage had breast cancer cells in their bone marrow. While ductal carcinoma in situ, when detected early, is 98% curable, metastases growing in the bone are basically incurable. On a positive note, the efficiency of metastasis is estimated to be low. Dissemination of tumor cells from the primary tumor mass into various organs is not sufficient to cause formation of '...overt, vascularized, clinically detectable metastases.' Some cancer cells may exist for long periods in organs as single dormant cells and can remain occult and asymptomatic for many years. These disseminated tumor cells apparently survived chemotherapy, radiation, and adjuvant therapy. The dormant cells may have long periods of latency. In some cases, active bone marrow metastases occur decades after the removal of the primary tumor. The possibility of these cells sitting quietly for years and then suddenly proliferating into a tumor mass remains a threat for patients and survivors. What causes some cancer cells to enter a dormant state? What stimulates them to grow? The answers to these questions could relegate metastatic disease to a manageable chronic condition.

Trying to understand and dissect out the mechanisms that maintain dormancy is difficult due to the lack of model systems that can mimic the in vivo behavior of these processes. There are cancer cell lines that mimic aggressive metastasis in mice. There are a few lines that mimic dormancy in that they traffic in the body but do not grow. To study bone metastasis, it is important to use a bone-relevant model. The microenvironment of the bone, its extracellular matrix (ECM), and associated cytokines and growth factors are believed to be key players. We propose to use a novel 3D culture system that permits us to grow authentic bone ECM. Using this system, we found that human metastatic breast cancer cells, MDA-MB-231, attach, proliferate, invade, and remodel the matrix such that the osteoblasts and the cancer cells align in arrays similar to those seen in pathological tissues but not in standard tissue culture. A metastatic suppressed variant of this line, MDA-MB-231BRMS1, forms primary tumors in mice and traffics to the bone but does not grow there. Likewise in the 3D culture, it loosely attaches but does not invade the matrix or grow. We propose to use these as pairs of metastatic and dormant lines in the 3D culture with a matrix formed by human osteoblasts. In a parallel set of experiments, murine cell lines D2A1 and D2.OR will be tested with murine ECM. This system allows the future genetic manipulation of the mice as a source of osteoblasts.

The hypothesis is that the ECM and cytokines of the bone microenvironment are critical in determining whether metastatic breast cancer cells will grow or become dormant. The specific aims are: (1) to determine how modification of the composition and structure of the ECM affects proliferation and dormancy. In order to manipulate the ECM in ways that mimic the in vivo status, we will deprive osteoblasts of estrogen, expose the osteoblast-produced ECM to substances that cause oxidative stress, and incubate the osteoblastic tissue with bone-degrading osteoclasts to partially degrade the matrix. The major endpoint will be the proliferation or dormancy of cancer cells in co-culture with the osteoblast tissue. Cytokines and growth factors are important components of the bone marrow as well as the bone matrix. There is anecdotal evidence from humans that inflammation due to injury or arthritis can trigger metastatic growth. In Aim 2, we plan to determine how bone remodeling and inflammatory cytokines in the microenvironment produced by the osteoblasts and/or cancer cells affects proliferation and dormancy of cancer cells. We will test cytokines associated with bone remodeling as well as cytokines associated with arthritis. In addition, we have found that some of these same cytokines are secreted by osteoblasts as part of an inflammatory stress response to metastatic breast cancer cells. Combinations of these cytokines or neutralizing antibodies to them will be added to the cultures of osteoblast tissue in combination with proliferating/dormant cancer cells.

The bioreactor is not a rapid screen but will serve as a preclinical model to test drugs that may promote dormancy or inadvertently break it. Commonly used anti-inflammatory molecules might be useful if applied under the appropriate conditions. Newer drugs can be tested in the bioreactor to determine if they might inadvertently stimulate dormant cells. At present there are no good models to study dormancy in vitro.