Neoplastic Angiogenesis

Neoplastic Angiogenesis — Not All Blood Vessels Are Created Equal

PERSPECTIVE

Neoplastic Angiogenesis — Not All Blood Vessels Are Created Equal Isaiah J. Fidler, D.V.M., Ph.D., and Lee M. Ellis, M.D.

The primary cause of death from cancer is the progressive growth of metastases that are resistant to conventional therapies. The site of metastasis from primary neoplasms is not random. Indeed, in 1889, Stephen Paget proposed that the process of metastasis did not occur by chance but, rather, that certain favored tumor cells (the “seed”) had a special affinity for the growth-enhancing milieu within specific organs (the “soil”). He concluded that metastases developed only when the seed and the soil were compatible. Our current version of the seedand-soil hypothesis consists of two principles. First, neoplasms are heterogeneous: they consist of cells with different biologic properties. Cancer cells are genetically unstable, and by the time of diagnosis, malignant neoplasms contain multiple subpopulations of cells with different biologic properties, including varying potential for invasion and metastasis. Second, the outcome of the growth and spread of cancer depends on multiple interactions (“crosstalk”) of tumor cells with homeostatic mechanisms, and tumor cells are likely to usurp the mechanisms. The induction and maintenance of the vasculature in neoplasms is a good example of the way in which tumor cells appropriate a physiologic process for their advantage. Cells cannot survive if they lack adequate supplies of oxygen and nutrients or are unable to dispose of toxic molecules. Oxygen can diffuse from capillaries to a distance of only 150 to 200 µm; when cells are farther away from a blood supply, they die. Thus, to become clinically relevant, a tumor requires persistent neovascularization, or angiogenesis, for its growth and survival. The extent of angiogenesis is determined by the balance between the positive and negative regulating molecules that tumor cells and host cells release into the microenvironment of the neoplastic tissue. Angiogenesis can occur through either sprouting or nonsprouting processes. Sprouting angiogenesis involves the branching of new capillaries from preexisting vessels. Nonsprouting angiogenesis results from the enlargement, splitting, and

n engl j med 351;3

www.nejm.org

fusion of preexisting vessels, a process that is mediated by the proliferation of endothelial cells within the wall of a vessel. Furthermore, hematopoietic precursor cells can be incorporated into the growing vascular bed. Since the genetic instability and biologic heterogeneity of neoplasms are the principal causes of failure of systemic antitumor therapy, targeting the neovasculature of tumors has been explored as a way of attacking a genetically stable and essential component of tumors. Recent studies, however, suggest that tumors can influence changes in the microenvironment of the involved tissue, and now we are learning that tumors may even affect the genetic composition of host cells within the tumor vasculature (see Figure). During the past 30 years, endothelial cells have been at the center of angiogenesis research. In the embryo, these cells arise from the splanchnic mesoderm, which differentiates into a multitude of vascular structures that have different functions in specific organs. In adults, hematopoietic precursor cells can differentiate into endothelial cells and become incorporated into a growing vascular bed. Endothelial cells that line the blood vessels in different organs are phenotypically distinct. Indeed, a considerable effort has been made to understand the molecular mechanisms that regulate the growth, maturation, and function of endothelial cells, because all these mechanisms are critical to the design of antivascular therapy for cancer. Recently, investigators have come to appreciate the importance of the types of cells other than endothelial cells that compose the vasculature, including pericytes, the supportive vascular smoothmuscle cells that are essential for the optimal function and survival of endothelial cells. There are relatively specific markers of vascular components, such as CD31 (platelet-endothelial adhesion molecule, or PECAM), CD34, and von Willebrand factor for endothelial cells and desmin, NG2, and alpha– smooth-muscle actin for pericytes. Morphologic evaluation is equally important in identifying vascular components, but the gold standard for the

july 15, 2004

Downloaded from www.nejm.org on September 2, 2005 . This article is being provided free of charge for use in Argentina. Copyright © 2004 Massachusetts Medical Society. All rights reserved.

215

Neoplastic Angiogenesis — Not All Blood Vessels Are Created Equal

PERSPECTIVE r Microenvironment Tumo

Angiogenic sprout

Mosaic vessel

Vascular mimicry

Pericytes

Tumor cells

Endothelial cells

Normal blood vessel

Endothelial cells with tumor DNA

Figure. Tumor Vasculature.

identification of cell types within the vascular network is still electron microscopy. It is commonly believed that endothelial cells within tumor vessels are genetically stable, diploid cells and thus differ from genetically unstable tumor cells. However, tumor vessels are much more complex than this belief suggests. In the vasculature of normal tissues, endothelial cells are typically covered by several layers of vascular smoothmuscle cells that regulate blood flow and vascular permeability. At the capillary level, the continuity of the pericyte layer is broken at intervals, thus increasing vascular permeability. The structure and function of small and large vessels differ in different tissues and in tumors growing in different sites.

216

Several years ago, it was proposed that tumor cells themselves can form vascular channels within tumors (e.g., melanoma), and it has been demonstrated that tumor vessels can have a mosaic composition in which endothelial cells line the vessel, along with interspersed tumor cells. In this issue of the Journal, Streubel et al. (pages 250–259) report that microvascular endothelial cells within B-cell lymphomas express markers of the neoplasm. This novel finding again demonstrates the complex and unpredictable nature of the tumor microvasculature. Using multiple markers, these investigators found that cells lining the microvasculature possess both endothelial-cell characteristics and the genetic aberrations of the surrounding B-cell lymphoma. The authors discuss four possible explanations for these findings. One is that pluripotent stem cells are the precursors of both the lymphoma and the endothelial cells. Indeed, recent research has demonstrated the ability of hematopoietic precursor cells to differentiate along multiple pathways. A second possibility is that the tumor cell that originates from the lymphoid lineage differentiates to an endothelial-cell phenotype under the influence of environmental stimuli. The third proposal is that lymphoma cells and endothelial cells fuse. The fourth is that endothelial cells (which are cells of the reticuloendothelial system) engulf apoptotic bodies containing genetic information from tumor cells. Although all these hypotheses are reasonable, the important message from this insightful study is that the tumor microvasculature is much more complex and unpredictable than it was initially perceived to be. Recognizing the complex cellular and molecular mechanisms that regulate vessel development in tumors should allow us to develop more appropriate antivascular therapies that may vary among tumor types (seeds) that grow in different organ microenvironments (soil). From the Departments of Cancer Biology (I.J.F., L.M.E.) and Surgical Oncology (L.M.E.), University of Texas M.D. Anderson Cancer Center, Houston.

n engl j med 351;3

www.nejm.org

july 15, 2004

Downloaded from www.nejm.org on September 2, 2005 . This article is being provided free of charge for use in Argentina. Copyright © 2004 Massachusetts Medical Society. All rights reserved.