National Cancer Institute
Member ListGroupsResource SharingFunding OpportunitiesArticlesClinical TrialsMeetingsWorkshopsJobs & TrainingTarp in the News
National Institute of Neurological Disorders and Stroke

Juvenile Diabetes Research Foundation International

National Eye Institute


Trans-Institute Angiogenesis Research Program (TARP) Workshop, May 10-12, 2004

The workshop was held to assess the state of current knowledge about angiogenesis, to define areas of research need, and to make recommendations to expand on successes and close gaps. Discussions were divided into Basic and Clinical sessions, although there was substantial crosstalk between the two domains. Indeed, the recommendations were sufficiently similar that they are largely combined in this summary.

Because the Basic discussions tended to range across the topics provided, some of the ideas expressed have been rearranged to conform to the outline.

Control of neovascularization is a mix of environmental cues, soluble factors, genetics, and other properties of tissues, vessels and of the extracellular matrix. "Angiogenesis" is many different processes, yet is at the same time too narrow a term to encompass various relevant forms of neovascularization, particularly, differences between normal and pathological states. Normal angiogenesis results from balanced expression in appropriate amounts and sequence of many factors, yielding hierarchical, evenly spaced, well-differentiated vessels. In tumors, wound healing, and inflammation, unbalanced expression of a small subset of factors yields vessels that are highly abnormal in organization, structure and function. Normal responses, for example, to ischemia, are lost. There are primary and secondary protein mediators of angiogenesis. Permeability is altered in disease - breakdown of barriers yields a state more like a wound. This compromise of barriers with chronic exposure to growth factors is in contrast to development, where the barriers stabilize with maturity. We do not know what signal is sensed, or which factors are important.

Heterogeneity is a major challenge. There is controversy about the source of cells in pathological angiogenesis, complicated by different pathologies and genetic differences among animal models. These differences extend to number, location and function of endothelial precursor cells (EPC)*; markers, and mobilizing factors. Elements beyond the endothelium itself, including pericytes, smooth muscle cells, monocytes/macrophages and platelets, further contribute to angiogenesis. This complicates comparisons among different disease states, organs, and model systems. Inflammatory and immune cells may be benign or hostile, depending on context. More information is needed on the source and role of all these cells. The basement membrane and extracellular matrix also exhibit heterogeneity. Developmental biology can inform us about initiation, maturation, migration, patterning and remodeling, including endothelial apoptosis. However, in pathologic models, the focus is on up- and down-regulation of factors, while in development, it is more on patterns of signaling. Diabetes, asthma, and aging offer potentially fruitful models.

Discussion of the clinical issues mapped more neatly to the questions posed. The group identified some successful strategies, including revascularization in breast cancer lymphangiogenesis, Avastin for colorectal cancer (albeit only in combination with conventional chemotherapy), anti-VEGF in eye diseases, alpha-interferon, and thalidomide; but with the caveat that results vary by site. Also mentioned were Vitaxin, endogenous antiangiogenic proteins, kinase inhibitors, and cox-2 inhibitors. The best strategies will be those based on strong rationales and solid preclinical data. We should avoid tweaking trials that are not founded in mechanistic science. Basic research is critical to clinical success, and clinical findings must cycle back to the bench. Developments on the horizon include anti-VEGF antibody fragments, VGF traps, permeability enhancers, and inhibitors of VEGF production. Modes of delivery known to be safe in one disease can be adapted to others. We can also use new modes of delivery for existing drugs. We need to address whether we should be normalizing or maturing pathological angiogenesis. The role of lymphangiogenesis is understudied - we do not know if angiogenesis therapies affect lymphangiogenesis responses, whether lymphangiogenesis contributes to tumor dissemination, and whether it complements angiogenesis. We need to better understand the shape of dose-response curves, to be able to measure drug action in tissues, and evaluate the toxicity of sustained inhibition of angiogenesis. The severity of the disease and age of the patient play a role. Other diseases to be considered are emphysema, pulmonary hypertension, rheumatoid arthritis, and psoriasis. Vascular targeting to specific "zip codes" should be exploited. The concept of pro-angiogenic treatment in Alzheimer's suggests that brain function should be followed in all patients on antiangiogenesis drugs.

Disease-disease transfer will benefit if the FDA and NIH partner to facilitate the efforts of industry. A phase I trial in oncology, where the barriers to entry are lower, could be used as the basis for Phase II trials in other diseases. Oversight processes could be streamlined without loss of stewardship. Shared databases for pro-and anti-angiogenesis trials in all body sites, especially toxicity data, would promote safety and reduce redundancy. Clinical trial networks to set standards and share expertise would be cost-effective, especially if industry could be attracted to use them. A collateral benefit would be data across body sites, for example, ocular responses of patients being treated for tumors.

More basic research, especially developmental biology, is needed on the interaction of angiogenesis and transplantation. Tumors are not a good model for growing mature vasculature in transplants. There will be site-specific issues affecting vessel phenotype in transplanted tissues. One promising angle is pretreatment of islets with matrix metalloproteases to recapitulate the 3-D network. An interesting question is how statins blunt rejection of heart transplants.


  1. Broaden the mission to Vascular Biology - encompassing development, maintenance, activation, remodeling, and pathology. A Vascular Health Initiative has potential for major public health impact.

  2. Break down organizational barriers and bridge the ICs. This would have two forms - an intramural research group with shared space, and an advisory committee of NIH and extramural investigators. Vascular biology programs in each IC together would constitute a virtual institute. Some favored the creation of an NIH Office. Ongoing coordination in the field should include a Web site and possibly a print newsletter to foster communications and disseminate findings. There should be a physical annual meeting, teleconferenced to major sites, also educational and administrative workshops, and sponsored sessions at meetings to reach a larger audience.

  3. A core facility would centralize storage and distribution of data and reagents. These would include: drugs, antibodies, probes, cDNA libraries, gene and siRNA expression systems; transgenic, knockout and other animal models. Bench scientists especially need access to clinical data, and tissues from responders vs. nonresponders. The core or other sites would perform services such as validation of precursor phenotypes and training in key technologies.

  4. Sites (centers) of various sizes should be encouraged. To attack problems like molecular mechanisms, some "descriptive" research and support for technology are necessary. These kinds of projects will likely need a special review. If such an initiative and its follow-on are to succeed, the structure of the peer review process is critical. A cross-cutting and trans-IC study section should be considered for assessing scientific merit.

  5. New biomarkers are needed to identify the phenotype of the lesion, assess the angiogenic potential of the patient, and measure the effectiveness of drug action. We need bioassays for endothelial and precursor cells, cell purification methods, and markers for cell surface epitopes, target molecules and circulating factors. Current assays require a large number of cells. Biological effectiveness as measured by such markers may be separate from clinical outcome. With markers in hand, we can conduct large-scale screening to study how phenotypes change with activation status, across tissues and microenvironments, in disease states, and over time.

  6. In order to judge whether therapies should destroy or "normalize" pathological blood vessels (i.e. complete the program), we must understand the properties and mechanisms of their formation. Research should aim at distinguishing types of normal and pathological angiogenesis to deduce the determinants of EC heterogeneity and dysfunction. We should study differences due to EPC homing, recruitment, and differentiation; and vascular branching. We should investigate lipid mediators and their cellular sources, and understand the role of proteinases. More work is needed to understand the role of lymphangiogenesis in pathologies, using markers of lymphatic endothelium, cytokines and factors involved, and the role of edema.

  7. Novel technologies, in addition to assays, should be developed. The field needs systems and integrative analysis tools encompassing the complexity of genetic polymorphisms, interacting effector molecules, and microenvironments. The underlying tissue modulates vessels, but vessels also affect the tissue (and each other).

  8. Existing 2-D culture methods do not reflect complexity and shear. We need in vitro models that reflect in vivo conditions but permit mechanistic studies. 3-D in vivo testing can address cell-cell interactions and identify potential drugs and regulators, but cost is a barrier to, e.g. multiphoton microscopy. Familiar animals are, in some ways, poor models for the human; we should look at others.

  9. New in vivo and molecular imaging techniques will use the anticipated biomarkers. Improved in situ hybridization and other staining techniques are needed. Attention should be given to permeability and to collateral vs. nutrient vessels.

  10. Clinical trials could be conducted by a multi-site network under the guidance of a TARP advisory board, and would feed the central data bank. Incentives are needed for combination drug trials, as regulatory hurdles make them difficult for industry to conduct. A paradigm that would allow treatment through disease progression was encouraged. JDRF, SBIR, and RAID offer models for accelerating connections from the bench to Phase I and Phase II trials.

  11. At the institutions, Vascular Biology organizations should provide seed money to promote collaboration and retention.

* "Endothelial precursor cell" is used, as there is no conclusive evidence that true stem cells are involved

National Heart, Lung, and Blood Institute

National Cancer Institute

National Institute of Diabetes & Digestive & Kidney Diseases

National Institute of Child Health and Human Development

  Home   |   Contact   |   Privacy Notice   |   Accessibility National Institutes of Health   Department of Health and Human Services