Foreign Occupiers

Metastatic tumors cause the vast majority of cancer deaths. Bone metastases are particularly debilitating. OSUCCC – James researchers are working to understand how they happen and how to block them.


Inject metastatic prostate cancer cells into an animal model and within 15 minutes tumor cells localize to the brain, eyes, lung, kidneys and bone. All will soon die except those in bone.

Thomas Rosol, DVM, PhD, professor of Veterinary Biosciences, and his laboratory tracked these events in a mouse model using cancer cells carrying the protein that makes fireflies glow. “The metastatic cells probably interact with the bone microenvironment and create a fertile environment that enables them to grow,” Rosol says.

Most invasive cancer cells don’t cause metastatic tumors. “Most cancer cells die after entering the bloodstream,” he says. Twenty days into the experiment, only two cells capable of causing bone metastases remained, one in the leg and one in the back.

“That’s two prostate cancer cells out of 100,000,” Rosol says. “We want to learn what’s special about those few surviving cells, and we want to learn how to stop them.”

Roughly nine in 10 cancer deaths are caused by tumors that spread—metastasize—to organs often far removed from the primary tumor. Metastatic cancer cells are models of the adage, “What doesn’t kill you makes you stronger.” They have survived chemotherapy, radiation and target agents, selective pressures that leave them resistant to further therapy. 

Treating bone metastases is particularly difficult. Surgical control is generally not an option because bones are usually not sites for surgical removal, Rosol says. “In general, once cancer metastasizes to bone, it is incurable.”

Lung, thyroid and renal cancers often metastasize to bone, and sometimes melanoma. Prostate and breast cancer and multiple myeloma commonly spread to bone. Rosol has personal experience with the latter. His father died of multiple myeloma at home under hospice care.

“Bone metastases can be devastating and painful,” he says. “They can destroy bone and lead to fractures. In the spinal column, they can cause vertebrae to collapse and lead to paralysis. In the leg, arm or ribs, they can cause fractures that won’t heal because of the cancer present. They can be very debilitating.”

Rosol is working to understand bone metastases in both human and veterinary patients. “A handful of spontaneous cancers are relatively common in dogs and cats, and we treat them at our Veterinary Medical Center,” he says.

The cancers include:

  • Breast cancer – occurs in both dogs and cats;
  • Prostate cancer, osteosarcoma and lymphoma – occurs in dogs;
  • Oral cancer – cats commonly develop an aggressive, invasive oral cancer.

Rosol’s lab develops cell lines from these cancers for studies in animal models.

The animal models enable translational research that has application for both human and animal patients.

Process and Patterns

While benign tumors tend to remain localized and removable, malignant tumors develop a capacity to invade neighboring tissue and to metastasize, a process that requires several steps (see sidebar).

Research has shown that many of the genes involved in invasion and metastasis are associated with a cell program called the epithelialmesenchymal transition, a pathway normally activated during early embryogenesis and wound healing. In malignancy, the pathway gives epithelial cells features of mesenchymal cells, enabling them to detach, move through the extracellular matrix, interact with endothelial cells and other stromal cells, and enter the blood and lymphatic systems.

The site of metastasis depends on the type of cancer. Most metastatic cells are transported by the bloodstream, and the preferred metastatic site is often the next organ downstream. But why breast and prostate cancers metastasize frequently to bone is not understood, Rosol says.

The idea that metastases of certain cancers prefer particular tissues Occupiers was recognized long ago. In 1889, the British surgeon Sir James Paget published a paper in the journal Lance proposing the “seed and soil” hypothesis, which likens tumor cells to seeds that require a particular kind of “soil,” or organ.

“Today, we think of this as the dynamic interplay between cancer cells and the tumor microenvironment,” Rosol says.

Breast and prostate cancers can affect the bone microenvironment in opposite ways: Breast cancer cells destroy bone, a process called osteolytic metastasis, while prostate cancer cells can cause bone production, or osteoblastic metastasis, a pathology that also causes pain and disability (see illustration).

Modeling Metastasis

Research by Rosol and others has provided insights into how bone metastases happen. In 1987, Rosol was part of a group that discovered parathyroid hormone-related protein (PTHrP), which is released by some types of cancer and can cause dangerously high levels of blood calcium, a condition called hypercalcemia.

PTHrP stimulates osteoclasts, which are cells that occupy and dissolve bone. Normally, osteoclasts work in coordination with bone-building cells called osteoblasts to remodel bones and accommodate changes in mechanical stress. Together, they replace an estimated 10 percent of bone mass annually, or the entire skeleton every decade.

It turns out, Rosol says, that the release of PTHrP by metastatic breast cancer cells induces osteoclasts to resorb bone. “So cancer cells themselves don’t destroy bone; they direct normal bone cells to destroy the bone,” he says.

Unfortunately, investigating the molecular mechanism underlying these events is notoriously difficult for two key reasons: a lack of animal models (other mammals develop spontaneous cancers, but the tumors produce few metastases), and the difficulty of seeing bone metastases on X-rays.

Rosol and his colleagues have made progress against both problems. A break came in the mid-90s when the technology of bioluminescence was discovered by Christopher Contag, PhD, at Stanford. Rosol immediately recognized its potential for studying metastasis, and he visited Contag’s laboratory to learn the technique.

The method uses genetic engineering to add the gene for luciferase, the enzyme that enables fireflies to glow, to lines of human and animal cancer cell lines. “This endows the cells with light just like fireflies,” Rosol says. They use a high-sensitivity camera to visualize the glowing cells inside the bones of immunodeficient mice even at a very early stage.

“This was a tremendous advance,” Rosol says. “We can count the number of metastases, calculate their growth rate and investigate different treatments to see if they reduce the number or growth of metastases.”

Rosol now has an inventory of 10 bioluminescent cell lines of canine, feline and human breast cancer; canine and human prostate cancer; canine and human osteosarcoma; and feline and human oral cancer.

“Oral cancer will destroy facial bones,” Rosol says. “It’s not metastasis; it’s local invasion into the bones, and it will destroy the jaw or face. The mechanism of destruction is very similar to that of metastatic cancer, even though the process to get to the bone is different.”

Breast Cancer – Osteolytic Metastasis

Breast cancer research by Rosol’s lab involves mainly human tumors. The researchers hypothesize that metastatic breast cancer invades bone through a vicious cycle of tumor-induced bone resorption.

Their studies and those of others have revealed a chain of events that leads to bone breakdown. It begins when breast cancer cells release PTHrP. This stimulates osteoblasts in the bone to release a cytokine called RANKL (receptor activator of nuclear factor kappa-B ligand), a protein important for bone metabolism. RANKL, in turn, induces osteoclasts to resorb bone. This breakdown releases both calcium, resulting in hypercalcemia, and a variety of growth factors from the bone. “Those growth factors nurture the proliferation of the cancer cells,” Rosol says.

Prostate Cancer – Osteoblastic Metastasis

Ahmad Shabsigh, MD, assistant professor of Urology at Ohio State and a collaborator of Rosol’s, knows what metastatic disease means for prostate cancer patients. “The vast majority of men diagnosed with prostate cancer in the United States have local disease, but still, about 32,000 men died from prostate cancer in 2010, and they all died from metastatic disease,” Shabsigh says.

Dogs are the only other mammal known to develop spontaneous prostate cancer, and their disease also metastasizes to bone, Rosol notes. “The biology of prostate cancer metastasis is amazing,” he says. “It induces osteoblasts to produce excessive, poor quality bone that can lead to spinal cord compression, pathological fractures and pain.”

PTHrP plays a role here, too, he says, but it involves a different portion of the molecule. In vitro evidence suggests that prostate-specific antigen (PSA), the protein used to detect prostate cancer, is actually an enzyme, he says. “We think PSA destroys the part of PTHrP that induces osteoclasts, and that the remaining truncated molecule induces bone formation.”

In 2007, Rosol had a breakthrough in prostate cancer modeling. A dog with prostate cancer was brought for treatment to Ohio State’s College of Veterinary Medicine Hospital for Companion Animals. The disease had metastasized to one leg, which was amputated. Rosol and his team obtained malignant cells from the animal and have developed a cell line that could lead to new treatments for prostate cancer.

“Most models of metastatic prostate cancer don’t induce new bone,” he says. “Amazingly, the cell line we developed using this tumor induces new bone just like human disease.” They have since published three papers on the cell line, which they call ACE-1. They’ve shipped the cell line to universities around the United States and the world and licensed it to a major pharmaceutical company for research.

A Need for Inhibitors

When cancer cells enter bone marrow, they often lie dormant there for years before metastatic tumors develop. “We don’t know why it takes so long, or how they come out of dormancy,” Rosol says.

At least 30 percent of people with primary breast or prostate tumors and no evidence of metastases have individual metastatic cells in their bone marrow, Rosol says. “They’re very rare, but they are there. Some will never develop a metastasis and some will. If we can learn what triggers their growth, we might be able to prevent it. It’s an important question but very hard to model.”

Currently, no drugs exist that block metastasis, but a growing number of agents are available to inhibit bone loss, he says. Most of these belong to a class of drugs called bisphosphonates, which were originally developed to inhibit bone loss due to osteoporosis. They work by inhibiting osteoclasts, and they’ve recently been used to treat patients with bone metastases.

“Their effect on metastasis is still unclear, but they do reduce bone destruction, and that is beneficial for a woman with breast cancer,” Rosol says. “They are also being investigated for prostate cancer, but we don’t know what they will do there.”

Newer agents that influence bone loss are in development or have been approved for clinical use by the Food and Drug Administration. They include denosumab, an antibody-based drug that inhibits osteoclasts by targeting RANKL.

Looking Foward

Preventing metastasis remains a daunting challenge. “One strategy is to prevent activation of those disseminated tumor cells,” Rosol says. “There is evidence that the tumor cells displace bone marrow stem cells and compete for the same location, so one idea being considered in the field is to repopulate bone marrow with new stem cells to evict the cancer cells.”

In the case of prostate cancer, Rosol is working with Shabsigh; Tarik Khemees, MD, a clinical fellow of Shabsigh’s; Michael Knopp, MD, PhD, professor of Radiology and Novartis Chair of Imaging Research; and Michael Tweedle, PhD, professor and Stefanie Spielman Chair in Cancer Imaging, to improve treatment of localized disease and prevent metastasis from occurring.

“There is no cure for advanced prostate cancer or for bone metastatic disease, but no patients die from localized disease,” Shabsigh says. “If we can do a better job of treating localized disease, perhaps we can prevent metastasis from occurring.”

Their goal is to improve the imaging of localized prostate cancer and the focal treatment of the disease using cryotherapy or high-frequency ultrasound.

“We want to develop an animal model that will help us develop techniques to identify exactly where the tumor is within the prostate,” Shabsigh says. “That may enable us to treat only the areas of the prostate with cancer and prevent the removal of the whole gland.”

A better understanding of the mechanisms of metastasis is essential for developing effective treatments for advanced disease. “The ability to model and image metastasis in mice with human and animal tumors has been an important advance to enable cancer research that will improve the lives of those with late-stage cancers,” Rosol says. “We will not be able to prevent cancer deaths until we understand why and how to prevent metastasis and metastatic tumor growth in patients.”

Foreign Occupiers-2

Breast cancer cells that metastasize to bone (see figure) release a cytokine called parathyroid hormone-related protein (PTHrP), which causes nearby bone cells called osteoblasts to release a cytokine called RANKL. RANKL, in turn, stimulates bone-absorbing cells called osteoclasts to dissolve the bone. Along with weakening the bone, this action releases growth factors from the bone that stimulate proliferation of the cancer cells and calcium that can lead to hypercalcemia. Prostate cancer cells that metastasize to bone (right side of figure) also release PTHrP, but it’s thought that the protein is cleaved by prostate-specific antigen (PSA), and that the resulting truncated PTHrP molecule and other growth factors activate abnormal bone production by osteoblasts.

Metastasis Requires Several Steps

  • Cancer cells invade neighboring tissue, break from the primary tumor and enter the circulation.
  • Most circulating tumor cells die; a few are caught in capillaries in a new organ.
  • At the secondary site, cancer cells may enter the tissues and experience one of several outcomes:
    • Solitary tumor cells may die, lie dormant or proliferate.
    • Nonvascular micrometastases may die, grow or lie dormant (i.e., cell death and cell proliferation occur equally).
    • A micrometastasis may grow and become a vascularized, clinically detectable metastatic tumor.

Sign up to receive publications from the OSUCCC – James

Sign Up Now

Contact The James

General Inquiries




460 W. 10th Avenue
Columbus, OH 43210

Contact Us

Meet the Leaders

Leadership & Executive Cabinet at The James

Learn More