Metastasis Research Center: Current Projects

Finding New Treatments for Rhabdomyosarcoma

Mary Baylies, PhD
Rhabdomyosarcoma (RMS) is the most common pediatric soft tissue sarcoma, representing 3 to 5 percent of all childhood cancers. RMS is classified into two subtypes, alveolar rhabdomyosarcoma (ARMS) and embryonal rhabdomyosarcoma (ERMS).

ARMS is notoriously aggressive and associated with metastasis. The outcome for patients with metastatic or recurrent disease remains dismal with an estimated five-year survival of  less than 30 percent. Approximately 14 percent of children with RMS have metastatic disease at the time of initial diagnosis. These data emphasize a critical need to identify novel therapeutic agents that limit RMS metastasis and improve survival rates, which is the key goal of this proposal.

With between 250 and 350 new cases diagnosed each year, RMS is a rare cancer lacking major investments in treatment development. Diseases with comparatively low numbers of newly diagnosed cases can profit from drug-repurposing screens of compounds approved by the US Food and Drug Administration. Drugs identified in these screens can be prioritized for clinical trials

We have adopted a relatively high-throughput protocol for drug screens in the Drosophila system and found three compounds that reduce ARMS metastasis. We hypothesize that compounds from an FDA-approved library will selectively inhibit ARMS development or metastasis in this efficient and novel in vivo model. Positives recovered from this screen will be tested for efficacy against human RMS cell lines.

Contralateral Breast Cancers: Independent Cancers or Metastases?

Colin Begg, PhD
Women diagnosed with breast cancer frequently develop cancer in the opposite breast, either simultaneously or subsequently. These cancers have traditionally been classified as new, independent primaries. However it is possible that a proportion of the tumors represent metastases from the original primary to the contralateral breast.

Indeed, using modern molecular pathologic technology investigators have demonstrated in other cancer sites, notably lung cancer, that many tumors classified as independent are in fact metastases. A resolution of this issue is important in breast cancer to allow individualization of the extent of surgery on the contralateral breast and axilla based on a diagnosis of metastatic disease versus new primary cancer, and to direct the choice and duration of systemic therapy in patients diagnosed with contralateral breast cancer (CBC).

Additionally, evidence of good survival outcomes after resection of limited metastatic sites such as the contralateral breast and axilla, followed by aggressive systemic therapy, would have significant implications for the management of some cancers. This includes patients presenting with stage IV breast cancer and an intact primary tumor, or those who develop metastases to isolated, surgically respectable sites in this era of improved targeted therapies.

Although a number of studies have examined the clonal relatedness of CBCs, these investigations have been based exclusively on small sample sizes and employed technology with limited resolution and diagnostic accuracy. We have the opportunity to provide resolution to the issue by assembling tumor specimens from both breasts for a large number of CBC cases from our tumor archives and other available sources, and using up-to- date sequencing technology and specialized statistical methods to classify the tumor pairs as either independent or of clonal origin.

ARMS is associated with chromosomal translocations generating the fusion oncoprotein PAX3/7- FOXO1. This fusion protein acts as an aberrant transcription factor, altering protein networks in muscle cells and ultimately causing tumor formation and metastasis. Targets of PAX7-FOXO1 are therefore candidates for the development of new drugs and could serve as additional biomarkers for the disease.

Our preliminary data has identified one such target. To identify additional target genes, we will carry out a limited RNAi screen in Drosophila based on comparative transcriptome analysis between fly and human. Top candidates from the fly screen will be validated in human tumor cell lines.

If successful, these investigations will identify compounds for clinical trials and novel drug targets or protein networks for ARMS treatment.

Neutrophil Mediated Antimetastatic Therapy

Robert Benezra, PhD
Cancer-related mortality is usually associated with metastasis, the spreading of cancer from a primary tumor to distant organs. Previous studies suggest that looking at the events that determine the site of future metastasis have shown that the primary tumor can modify distant organs and prepare them for the arrival of tumor cells.

In addition, our preliminary data suggests that the primary tumor can induce an antimetastatic immune response and delay the formation of distant metastases. This response is mediated by neutrophils — a subset of white blood cells that play a role in inflammation and immune responses to infections.

Neutrophils are mobilized by the primary tumor and accumulate in the pre-metastatic lungs prior to the arrival of tumor cells. Tumor-entrained neutrophils (TENs) but not naïve neutrophils can induce tumor cell death in vitro by initiating physical contact with the tumor cells and generating H2O2. In vivo, depletion of neutrophils results in enhanced metastatic progression while transfusion of tumor-entrained neutrophils can dramatically reduce the development of lung metastases.

We have identified the tumor-secreted factors that stimulate the antimetastatic neutrophil response and shown that while these factors may contribute to primary tumor progression, they concomitantly stimulate an anti-metastatic response at the metastatic site. To gain additional insight into the antimetastatic role of neutrophils and explore their therapeutic potential, we will further characterize tumor-entrained neutrophils and test their antimetastatic potential in a variety of preclinical settings.

Cell Adhesion and Signaling in Tumor Invasion, Dormancy, and Reactivation at Metastatic Sites

Filippo Giancotti, MD, PhD
Dominant mutations in oncogenes and recessive mutations in tumor suppressor genes disrupt the regulatory circuits that govern cells, endowing them with the ability to survive, proliferate, and invade independently of contextual constraints. Our laboratory studies the molecular basis of tumor initiation and progression to metastasis with emphasis on the role of adhesion signaling in these processes. We are currently studying various steps of the invasion-metastasis cascade in the following projects:

  1. We use mouse genetics and biochemistry to study the mechanisms by which oncogenic mutations hijack cell adhesion and signaling pathways to promote the loss of contact inhibition and anchorage dependence, thereby enabling local tissue invasion.
  2. We study the molecular basis of novel oncogenic mutations that promote tumor invasion and dissemination by inducing epithelial-mesenchymal transition (EMT). This work involves biochemistry, mouse genetics, and analysis of clinical samples.
  3. We have developed novel gain-of-function genetic screens in the mouse to identify genes that enforce dormancy or induce reactivation of disseminated cancer cells. The genes we have identified are components of signaling pathways that regulate adult stem cells, and we are now characterizing their mechanism of action and validating their relevance to human cancer.

View our metastasis-related publications on PubMed.

Real-Time In Vivo Marker Detection for Rapid Measurement of Metastasis Onset

Daniel A. Heller, PhD
The study of metastatic cancer presents several challenges that may be addressed using improved sensor technologies. These challenges include detecting metastasis at early stages and discerning indolent versus aggressive tumors. Both tasks require simple, reliable, inexpensive, and non-invasive methods for detection.

We will perform real-time and multiplexed in vivo quantification of metastatic cancer markers. The elevated serum levels of several antigenic biomarkers have been shown to mark the onset of metastasis. Using biofunctionalized single-walled carbon nanotubes (SWCNTs), we will develop unique sensors to detect the concentration of these biomarkers in real-time, with high sensitivity and selectivity, and from within the organism.

Developing Novel Strategies to Target Tumor Cell-Macrophage Interactions in Brain Metastasis

Johanna A. Joyce, PhD
Metastasis to distant sites, including the brain, remains the single most lethal aspect of breast cancer. Despite important advances in the treatment of localized breast cancer, few therapies succeed at combating disseminated disease.

Indeed, the critical need for improved therapy for women with metastatic breast cancer is underscored by the fact that only 5 to 10 percent of these patients survive five or more years after their diagnosis. This is a particularly dire problem for breast cancer patients whose disease has spread to the brain, as there are currently no effective therapies for treating brain metastasis.

In addition, as the incidence of brain metastasis often exceeds 30 percent in both HER2-positive and triple-negative breast cancer subtypes, a significant number of breast cancer patients are affected by this disease annually in the United States. Clearly, novel insights are required to improve treatment of breast to brain metastases.

One new paradigm comes from the observation that noncancerous stromal cells in the tumor microenvironment can make critical contributions to promote malignancy. While the tumor microenvironment has emerged as an important regulator of cancer progression in other organ sites, and hence is considered a potential therapeutic target, our knowledge of the brain tumor microenvironment remains very limited.

Specific Aims

We will address this knowledge gap and specifically focus our investigation on tumor-associated macrophages and microglia in the brain. We have interesting preliminary data supporting the tumor-promoting functions of these cell types.

We hypothesize that macrophages in the brain microenvironment significantly contribute to promoting the extravasation, seeding and outgrowth of metastatic breast cancer cells in this secondary site. Targeting these cells, or the tumor-promoting factors they produce, might lead to novel mechanistic insights into brain metastasis.

We will use animal models to evaluate different small molecule inhibitors targeted against these factors in preclinical trials. This project has the potential to rapidly translate new findings into innovative strategies to successfully treat metastatic breast cancer.

Mutant p53 and metastasis

Scott Lowe, PhD
The tumor suppressor p53 is the most frequently mutated gene observed in human cancers. In most tumors, p53 is inactivated through a two-hit mechanism involving a missense mutation in one allele and loss of the other allele though a deletion event linked to loss of heterozygosity. Interestingly, mice harboring common p53 missense mutations develop tumors that are more aggressive and metastatic than those arising in p53-null mice, suggesting the tumor cells harbor gain-of-function activities that influence metastasis. Our laboratory explores the mechanisms by which these mutations promote metastasis in pancreas cancer and other aggressive tumor types with the hope that this research will provide insight into p53 biology and point toward therapeutic targets for limiting cancer dissemination.

View our metastasis-related publications on PubMed.

Metastasis Genes and Pathways

Joan Massagué, PhD
To better understand and treat metastasis, it is critical to identify the mechanisms that turn disseminated tumor cells (DTCs) into metastasis-initiating seeds. Our group has recently identified a number of genes that amplify cell survival and stem cell pathways in DTCs, promoting metastasis colonization in breast, lung, and renal cancers.

Working on breast cancer, we found that the tyrosine kinase protein SRC primes triple-negative breast cancer DTCs for a robust response of the PI3K-AKT pathway in limiting levels of pathway activators upon infiltrating the bone marrow. The leukocyte-tethering receptor VCAM-1 was found to do the same for DTCs in the lungs. The extracellular matrix protein tenascin-C forms niches that amplify WNT and NOTCH signaling in metastasis-initiating breast cancer cells. 

We also found that the CXCL1-S100A8 paracrine loop protects DTCs from the stresses of metastasis and chemotherapy in lung and breast cancers. These genes and mechanisms prime tumor cells for metastasis in animal models and are associated with relapse in patients. Pharmacologic inhibitors of PI3K, AKT, and CXCL1 can eradicate micrometastatic disease in mice.

These findings point at a general strategy whereby DTCs resort to diverse mechanisms for the purpose of amplifying the signal output of a common set of pathways. The identification of metastasis suppressor micro-RNAs in breast cancer and epigenetic regulators of metastasis in renal and breast cancers adds to this growing knowledge.

In addition, we addressed the long-standing question of how organ-specific metastatic traits emerge in primary tumors. We recently found that mesenchymal stromal signals resembling those of a distant organ, the bone marrow, select for cancer cells that are primed for metastasis in that organ, illuminating how metastatic traits may evolve in a primary tumor and its distant metastases. In collaboration with Larry Norton, Deputy Physician-in-Chief for Breast Cancer Programs, we elucidated the phenomenon of tumor “self-seeding,” in which circulating metastatic cells infiltrate tumor masses for clonal expansion. The ability of tumors to self-seed may provide an explanation for the commonly observed clinical link between tumor growth and aggressiveness.

Focusing on the problem of brain metastasis, we developed models of breast cancer and lung adenocarcinoma and identified mediators of cancer cell infiltration in the brain. Using these models, we found that reactive astrocytes in the brain parenchyma produce plasminogen activator (PA) and generate plasmin that mobilizes the pro-apoptotic cytokine FasL to kill the vast majority of intruding cancer cells. Brain metastatic cells express high levels of anti-PA serpins that thwart this astrocyte-mediated defense. Moreover, the surviving brain metastatic cells express the neural cell adhesion molecule L1CAM for spreading on brain capillaries as a requirement for colony outgrowth.

We are currently investigating the broader implications of these finding for metastasis to other organ sites and by different tumor types, and using preclinical models to explore various strategies for the translation of our findings into the clinic.

View our metastasis-related publications on PubMed.

Metastasis in the Zebrafish System

Richard White, MD, PhD
Metastasis exerts its physiologic effects via an interaction between the tumor cells and the host cells within individual organs. Because of this, an understanding of the factors that promote metastasis will require increasingly sophisticated in vivo models that recapitulate each step of metastasis.

In recent years, the zebrafish has emerged as an important tool in cancer research because of its capacity for high-resolution in vivo imaging coupled with rapid and large-scale genetic manipulation. We have developed a zebrafish model of melanoma that closely resembles the human disease at histological and gene expression levels.

We will use the zebrafish as a platform to discover new genes and pathways associated with metastasis. We will first develop quantitative assays for melanoma metastasis using a transparent adult zebrafish strain known as casper. With these tools, we will then characterize intratumoral and metastatic heterogeneity using the Brainbow fate mapping system. Tumor heterogeneity is an essential factor underlying treatment resistance and failure during disseminated disease.

In addition, we will investigate whether stress-induced mutation via upregulation of dinB/polK mechanistically underlies the generation of heterogeneity during tumor progression. This would have broad implications across different cancer types, since mechanisms generating heterogeneity would be a rational therapeutic target in metastatic disease.

Our discoveries in the zebrafish will be complemented by investigation of parallel mechanisms in human tissues, leveraging the strengths of each system to investigate fundamental dynamics of metastatic progression.