Process of Angiogenesis Plays a Role in Response of Cancers to Radiation Therapy

DENVER-Research on angiogenesis is revealing the role this phenomenon plays in the response of cancers to radiation and, in the process, providing some important lessons for clinicians, according to a keynote address given by Judah Folkman, MD, professor of cell biology, Children's Hospital and Harvard Medical School, at the 47th Annual Meeting of the American Society for Therapeutic Radiology and Oncology.

"In 1971, I published this hypothesis that tumors are angiogenesis dependent, and I thought that the field would explode after this paper. But nothing happened," Dr. Folkman said. "In fact, nothing happened for 10 years . . . until the early 1980s, when we isolated basic fibroblast growth factor (bFGF) from tumors; then critics converted to competitors and the field has taken off. There are now 70 papers a week with angiogenesis in the title."

Experts now regard angiogenesis as a critical phenomenon in cancer (as well as other diseases), view angiogenesis inhibitors as the fourth modality in cancer therapy, and credit them with most of the recent gains in survival of colon, lung, and breast cancer, Dr. Folkman said. "Therefore, it is timely to ask about the role that the process of angiogenesis plays in the radiation response and to ask what lessons have been learned from angiogenesis research."

Radiation Sensitizers?

Turning to tumor biology, Dr. Folk-man addressed a series of questions about angiogenesis and radiation response, the first among them, can antiangiogenic therapy increase blood flow and oxygen in a tumor-changes in the tumor microenvironment that may increase the effectiveness of radiation.

"Conventional wisdom was that antiangiogenic therapy would always reduce oxygen in a tumor," he noted. But a 1995 study showed that administration of an angiogenesis inhibitor (the investigational agent TNP-470, an analog of fumagillin) to mice after tumor implantation led to a 17% increase in intratumoral oxygen, as well as a 2.2-fold increase in tumor growth delay after radiation therapy-findings that were counterintuitive.

This, in turn, leads to a second question, he said: How does antiangiogenic therapy increase tumor blood flow and oxygen? A series of studies showed that one mechanism was a reduction in vascular leakage, an action that has been shown for caplostatin, endostatin, thalidomide (Thalomid), and etoposide.

Another mechanism appears to be downregulation of expression of hypoxia-inducible factor 1--α (HIF1--α), a factor produced by hypoxic cells that increases VEGF expression as well as tumor invasiveness; this mechanism has been shown for 2-methoxyestradiol, endo-statin, rapa-mycin, and tumstatin.

Some agents alter expression of a variety of other genes as well; for example, endostatin downregulates expression of about 20 genes (including those for HIF1--α and VEGF) in endothelial cells and upregulates expression of a dozen or so others (including the gene for thrombospondin, a powerful endogenous inhibitor of angiogenesis).

Based on thse findings, Dr. Folkman addressed the question of whether angiogenesis inhibitors can radiosensitize tumors. Experiments with thalidomide, for example, have indeed shown that this agent increases intratumoral blood flow and oxygen levels, he noted. And experiments with other inhibitors indicate that after the first week or two of treatment, assuming the right dose, a new phenomenon-capillary dropout-takes over, mediated by the therapy itself or by added radiation therapy. Eventually, because each capillary supports a cylinder of tumor cells, the tumor begins to regress.

These findings give rise to the clinical question: Can angiogenesis inhibitors be combined with radiation therapy to increase efficacy? "That is still debatable," Dr. Folkman said, citing several studies that have shown increased inhibition of tumor growth when various angiogenesis inhibitors (angiostatin, SU5416, or SU11657) were combined with radiation therapy. He stressed the need for clinical trials assessing the effects of adding angiogenesis inhibitors (alone or in combination) to radiation therapy.

Loss of Endogenous Inhibitors

A possibly odd-seeming question, Dr. Folkman acknowledged, is whether radiation-induced regression of a primary tumor can lead to loss of endogenous angiogenesis inhibitors and increased growth of remote metastases.

Preclinical studies show that this phenomenon occurs for at least certain cancers, he noted. For example, irradiating a primary Lewis lung carcinoma growing subcutaneously in mice leads to regression of the primary tumor but also to a 10-fold increase in the number of lung metastases and a 3-fold increase in lung weight, relative to nonirradiated controls. Although the mice with nonirradiated primaries also have lung metastases, these remain microscopic and dormant.

The same phenomenon of enhanced metastasis may be seen clinically, he commented, when some non-small-cell lung cancers are irradiated and when certain primary tumors are removed surgically.

Subsequent experiments have indicated that some primary tumors produce enzymes that generate angiogenesis inhibitors. "The concept is that, maybe for certain tumors, you can measure the endogenous angiogenesis inhibitors the tumor is making by getting a blood test before you do the radiation therapy," Dr. Folkman said, and then add back the lost inhibitor or a similar one.

"I don’t think many people believe this concept, but it has been confirmed," he said. "It has not gotten into practice, but it is something to think about."

In fact, he added, 28 endogenous angiogenesis inhibitors have now been identified, some of which have been confirmed to keep tumor growth in check. Moreover, a variety of orally administered low-molecular-weight drugs used to treat various conditions have been found to increase levels of these endogenous angiogenesis inhibitors; for example, celecoxib (Celebrex) increases levels of endostatin, while doxycycline increases levels of thrombospondin.

Serologic research further substantiates the existence of endogenous inhibitors, Dr. Folkman commented. "All serum from all animals virtually always stimulates the growth of endothelial cells in culture, except for serum in tumor-bearing animals if the tumor is liberating angiostatin, endostatin, or one of the other inhibitors," he said. Experiments have shown, for example, that serum from animals with tumors has a markedly reduced stimulatory effect on endothelial proliferation relative to control serum, but with increasing time after tumor removal, the stimulatory effect of the serum returns to control levels.

In the preclinical setting, this angiogenic phenomenon explains what has come to be known as concomitant re-sistance, Dr. Folkman pointed out. Specifically, for certain cancers, if the same number of tumor cells are implanted bilat-erally in mice, only the cells on one side grow into a tumor. But if that tumor is removed, the contralateral cells suddenly begin to grow into a tumor.

In the clinical setting, a related phenomenon is the occurrence of so-called parallel growth, in which a patient's tumor has lost its angiogenesis inhibitor through mutation. "All oncologists know when patients progress from the stage of having one tumor and years later, a tumor recurrence, and later, the stage where tumors are recurring very rapidly-parallel tumor growth-that is a fearsome kind of problem," he said.

Thus, when it comes to treating primary tumors, he advised, "Certain angiogenesis inhibitors are generated by primary tumors . . . and if you treat the primary, it would be a good idea to replace these inhibitors, either with the inhibitor that the primary has lost or with another inhibitor."