Researchers Find Possible Causes for Toxics Leaking from Tumors

A Duke research collaboration has identified a likely route for "leakage" of therapeutic gene-bearing viruses out of tumors in experimental anti-cancer gene therapy experiments in laboratory animals. The group also found this toxic leakage can be avoided by using a chemical extracted from common brown algae.

Their work was described in a presentation Sept. 11 at the American Chemical Society's national meeting in New York, as well as in a research paper accepted for publication in the journal Molecular Cancer Therapeutics.

Investigators from the biomedical engineering department at the Pratt School of Engineering and the radiation oncology department at the Duke Medical Center collaborated to trace why high concentrations of the protein produced by the therapeutic genes were present in the wrong places during animal experiments directed against tumors.

The experiments involved transplanting tumors into the legs of mice and then injecting those tumors with adenoviruses genetically altered to carry the cancer-fighting gene. About 24 hours after those adenoviruses infected the tumor cells, the virus-carried genes could then begin manufacturing a known anti-cancer protein called mouse interleukin-12 (IL-12).

When the researchers first tried the experiment using concentrations of IL-12 genes in the viruses they judged high enough to treat the cancer, "the animals died within 10 minutes," said Fan Yuan, an associate professor of biomedical engineering, in an interview.

Exploring the reasons for the sudden deaths, the group used lower gene amounts that the animals could tolerate to trace what happened in their bodies during the extended infection and gene expression process.

They found that the virus preparations did not stay in the tumors as planned but also moved elsewhere in significant concentrations, principally to the liver.

The reason for that unanticipated migration was tumor blood vessel damage by the injection needle, Yuan said. After entering those vessels through the tiny wounds, viruses could quickly migrate through the entire interconnected bloodstream. Identifying the problem, the researchers also discovered an answer when they mixed the virus preparation with alginate, a major constituent of the cell walls of brown algae.

Injecting the combination of alginate and virus into the mouse tumors reduced by eight-fold the concentration of IL-12 in the animals' livers compared to injecting the gene-bearing virus alone, the investigators found.

"The alginate solution is simple and straightforward," said Yuan, who added that this algae preparation is a nontoxic biocompatible polysaccharide used in tissue engineering.

The researchers suspect that the alginate solution's high viscosity, about 1,000 times higher than water's, may block most of the viruses from leaking out of the tumor tissue through injection wounds.

Such an action, Yuan acknowledged, would resemble how certain automotive products can stop leaks in radiators and power steering or oil circulation systems. "That's the same analogy," he said. "It's like Jello."

But that high viscosity also makes injecting the viruses into tumors more difficult. "You really have to push very hard," Yuan said. So the researchers are now investigating whether other kinds of long-chained molecular polymers will block viral leaks as effectively as the alginate. With those, "you can use a very small force to push them through the needle," Yuan said.

Besides Yuan, others in group include Yong Wang, Yuan's graduate student; Ava Krol, Yuan's research associate; Chuan-Yuan Li, an associate research professor in radiation oncology; and Jim Kang Hu and Yong-Ping Li, research associates of Chuan-Yuan Li.

The work was supported by the National Science Foundation and National Institutes of Health.