BOSTON (The Boston Globe) - Houston researchers began second-stage trials on the highly touted cancer drug Endostatin, the first earnest evaluation of the tumor-shrinking potential of Harvard University researcher Judah Folkman's signature treatment.

In previous trials, the drug was found to be safe. It also showed subtle signs of working as Folkman intended. Two patients displayed temporary tumor shrinkage - although most of the 81 participants got worse.

In the new tests, run by the University of Texas' MD Anderson Cancer Center, the patients are slightly healthier than those in previous trials. And they will get the drug from a tiny infusion pump instead of from daily injections.

"Pre-clinical data indicate that the continuous infusion of Endostatin to the patient throughout the day may significantly improve the effectiveness of the drug," said Dr. Edward Gubish, president of EntreMed, the Maryland-based company that licenses Folkman's discoveries.

Endostatin attacks the blood vessels that nourish cancer tumors. Over the course of three decades, Folkman showed exactly how tumors hijack these vessels, a process called angiogenesis. Then he created drugs designed to inhibit it.

More than 40 other anti-angiogenesis drugs are further along in clinical trials than Endostatin. But Endostatin, which caused tumors in mice to virtually disappear, has become the focus of much attention and hope around the world.

The MD Anderson trial will involve about 30 patients, the largest Endostatin test to date. Thursday, the first patients were offically enrolled. Results are not expected for a year.

Researchers there said the trials would be open to all cancer patients, but they would search for sarcoma and melanoma patients because earlier trials found some evidence that Endostatin was particularly effective on these types of cancer.

Since 1999, 81 patients have received daily Endostatin injections at hospitals in Amsterdam and Wisconsin, at MD Anderson, and at the Dana Farber Cancer Institute.

The new MD Anderson trial uses infusion, similar to the method used to deliver insulin to diabetics. Patients will wear belts equipped with pumps about half the size of a paperback book. They will continuously pump Endostatin in tiny amounts through a tube running into a chest vein. The patients must return to the hospital every two days for a refill.

This new delivery method is a result of recent research that found mouse tumors could be shrunk with lower Endostatin doses if infusion pumps were used instead of injections.

Dana Farber already has four patients being treated with Endostatin infusion pumps. These patients were carried over from the injection-based trial there that ended in June.

The Phase I trials found Endostatin to be safer than any cancer treatment in history. The new tests, however, have different criteria for success.

"Phase II means you're beginning to look for efficacy. What is the effect on the tumor?" said Folkman.

The trial will also seek to answer three questions: What is the most effective dose level for Endostatin? Do infusion pumps work better than injections in humans? Does the drug work better on less advanced cancers?

The Phase I patients were close to death in many cases. Most showed no signs of tumor shrinkage, though the only goal of the test was to determine if the drug was safe. Nonetheless, many cancer patients and investors were disappointed.

But there were signs of hope. A jaw cancer patient in Houston had 50 percent tumor shrinkage and a Boston neuro-endocrine cancer patient had 20 percent shrinkage. Tumors remained stable in five patients for more than four months. But all these tumors eventually grew again.

Folkman made clear that such results would not be satisfactory in the new trial.

"They're looking for measurable tumor regression," he said.

Copyright 2001 The Boston Globe. All rights reserved.