By David N. Leff

FT. LAUDERDALE, Fla. * Half a loaf is better than none, but half an antibody is looking better than the whole molecule.

Molecular biologist Ira Pastan reported to the closing session of the Winter symposium here Wednesday on the "Design of recombinant immunotoxins for the treatment of cancer."

Pastan heads the laboratory of molecular biology at the National Cancer Institute, in Bethesda, Md. His group is treating 43 patients with solid tumors, mostly colorectal, in an ongoing modified Phase I clinical trial that is still enrolling subjects.

"The drug is given intravenously by infusion over 30 minutes," Pastan told his audience, "usually three times every other day."

He describes the open-label Phase I study as "modified," because it appraises responses to the cancer, as well as determining optimum dosage of the therapeutic conjugate.

"A Phase I trial formally is designed only to determine toxicity and not efficacy," Pastan explained to BioWorld Today. "For ethical reasons, we need to select patients who have antigen on their tumors; and who theoretically therefore could respond. We only treat patients who could potentially benefit, because their tumors react with the antibody."

Pastan added, "Since the trial is not finished, of course we don't yet know what the outcome will be. In one patient," he observed, "we did see a clear response * signs of some tumor regression * but we're still escalating the dose, and we hope to see more responses."

The antigenic target of his new antibody consists at this point of "a carbohydrate that's present on almost all colon cancers, more than half of breast cancers and many other adenocarcinomas."

His team constructed an antibody that recognizes that carbohydrate. It combined the Fv fragment of the original antibody with the toxin to make what they call the single-chain immunotoxin.

"This is the first single-chain immunotoxin that's been developed for the clinic," Pastan pointed out. He added that "Bristol-Myers Squibb is making a similar immunotoxin, using technology that they licensed from the government, from our NCI work."

The single-chain antibody's smaller size enhances its facility in entering a tumor cell.

The tumor-killing drug that the antibody delivers to the target, Pastan said, "is made by the pseudomonas aeruginosa bacterium, and is called pseudomonas exotoxin A. It kills cells by entering them and arresting their protein synthesis. This somehow induces the programmed cell death pathway in the cell, and they die rapidly.

Pastan pointed to the trial's preliminary positive responses as "the first results that have shown tumor regressions in patients. We think that's in part due to our selection of the single-chain antibody. Our antibody is somewhat more specific. It may also be that the toxin is better tolerated than some of the drugs that we have used previously. We're hoping that these trials will verify that."

He recalled that "a lot of other agents were developed either preclinically or clinically, which had severe side effects when they got into the clinic and failed for that reason.

As the solid-tumor studies proceed, the NCI team has begun a new Phase I trial, featuring patients with cancers of the blood, such as leukemias and lymphomas.

"We have designed an immunotoxin also to treat hematological malignancies," he said, "which over-express an interleukin-2 receptor. That agent's only been in clinical trial for a few months, so we're still at fairly low dose levels.

It uses the same Pseudomonas endotoxin, but a different antibody, which reacts with a subunit of the IL-2 receptor.

And a third clinical study is planned to start, which has the stabilized form of the antibody. We hope to start "some time this summer." It will employ a structurally modified version of the IL-2 receptor-targeting single-chain antibody.

"Depending on the results of the three trials we're now doing," Pastan said, "we'll formulate future plans. The most important thing to us is that not only do these things work well in animal models but they do show some antitumor responses in patients, particularly patients with solid tumors. To us that's a very impressive result, and, we think, means that if we can design better molecules, we'll see more responses."

One way in which he and his associates have bettered their antibody molecule, Pastan told the symposium, is by enhancing its stability in vivo.

"One of the problems with these Fv antibodies, he pointed out, "is that they are unstable, and we have had to increase their stability." The way we developed to do that was to engineer a disulfide bond into a conserved region of the Fv fragment, so it will hold that fragment together for days or weeks or even months. This is a generally applicable method of stabilizing Fv fragments.

Pastan's presentation was invited by the symposium organizers on the occasion of conferring on him its annual Special Achievement Award. *