Blood vessels are the lifelines of tumors, and they are increasingly the focus of cancer treatments. More than 400 clinical trials are testing ways to prevent blood vessels from supplying tumors with the nutrients and oxygen they need to grow.

The best known strategy is to inhibit the growth of new blood vessels - angiogenesis - in and around tumors. The drug bevacizumab (Avastin) was designed to do this by blocking a protein called vascular endothelial growth factor (VEGF), which is a regulator of angiogenesis.

Since 2003, studies have shown that adding bevacizumab to standard chemotherapy can increase the survival of patients with advanced lung cancer or metastatic colorectal cancer. The combination has also benefited women with metastatic breast cancer.

Some of the newer "multitargeted" drugs inactivate the receptor protein for VEGF on cells as well as other proteins involved in tumor growth. Two of these, sunitinib (Sutent) and sorafenib (Nexavar), have benefited patients with advanced kidney cancer.

With mounting evidence that antiangiogenic drugs help certain patients, researchers are trying to understand how the drugs achieve their effects. For now at least, there are more theories than answers.

The conventional view of bevacizumab is that the drug inhibits the development of tumor blood vessels, a hypothesis that originated with the pioneering research on angiogenesis by Dr. Judah Folkman in the 1970s.

But the fact that bevacizumab has demonstrated a benefit when added to chemotherapy raises the question: If bevacizumab diminishes the tumor blood supply and reduces blood flow to tumors, how does chemotherapy reach its targets?

"It's a very interesting paradox," says Dr. Rakesh Jain, a professor of tumor biology at Harvard Medical School. Five years ago he proposed an additional hypothesis to explain how certain antiangiogenic drugs work.

These drugs, he suggests, prune some blood vessels and structurally improve the remaining ones (tumor vessels are often inefficient and leaky). The result is "normalized" vessels that improve the delivery of oxygen and chemotherapy to tumors.

"This is a fascinating theory elegantly worked out in the lab, and now we need to see if it works in people," says Dr. Percy Ivy of NCI's Cancer Treatment and Evaluation Program (CTEP), who oversees clinical trials involving antiangiogenic agents.

A small trial testing bevacizumab for rectal cancer provided support for the hypothesis last year. The drug normalized the tumor blood vessels of 11 patients after 12 days, Dr. Jain and his colleagues reported.

Other trials are exploring the hypothesis, including a phase II study testing an experimental drug for recurrent glioblastoma brain tumors.

The primary goal of the study is to assess the benefits of the therapy, but the researchers will also use magnetic resonance imaging to study the effects of the drug on the function of blood vessels over time.

"We hope to use imaging tools to see if normalization is occurring in patients," says lead investigator Dr. Tracy Batchelor of Massachusetts General Hospital. The drug, AZD2171, selectively inhibits signals from the VEGF receptor.

Theories about antiangiogenic treatments are not mutually exclusive, Dr. Batchelor adds: "You may be improving blood flow to a tumor at one time and restricting blood flow at another."

Indeed, a drug may work through different mechanisms at different times, says Dr. Helen Chen of CTEP, who studies bevacizumab. Therapy is often given for months, and the mechanisms of action may vary depending on the stage of treatment, she says.

If normalization does occur, physicians might want to know so they could deliver antitumor drugs at optimal times. To learn about timing, Dr. Batchelor's team will collect images throughout treatment.

Other researchers are testing a strategy known as "metronomic" chemotherapy. This involves administering low doses of chemotherapy frequently and without continuous interruption. The goal is to achieve antiangiogenic effects by not allowing time for tumor blood vessels to repair themselves after being injured by chemotherapy, while limiting toxicity to normal tissues.

Yet another theory proposes that anti-VEGF therapies may directly affect tumor cells, an idea that emerged with the recent discovery of VEGF receptors on some tumor cells.

"It's possible that these therapies may be diminishing the survival of tumor cells and making them more sensitive to chemotherapy," says Dr. Lee Ellis of the University of Texas M.D. Anderson Cancer Center.

Dr. Ellis is also studying the role of nitric oxide in mediating changes to blood vessels after anti-VEGF therapy.

"The smart approach is to assume that there are multiple mechanisms of action," he says. "Biology is not linear, and it's not simple. If any of us thinks there's only one mechanism, then we're going to miss opportunities to understand how these drugs work."

This research is driven partly by the need to identify patients who might benefit from these drugs - a growing priority, given the expense and potentially toxic side effects of the drugs. But the results may advance the field in unexpected ways.

All of these studies contribute to our understanding of tumor biology, and the insights we gain will lead to the development of new agents," says Dr. Ivy.

By Edward R. Winstead