News Release

Successful Cancer Therapy May Doom Later Treatments To Failure

Peer-Reviewed Publication

Harvard Medical School

Drugs that successfully carve away at cancers of the lung, brain, breast, and other organs could be acting as double-edged swords, according to a new study by Massachusetts General Hospital and Harvard Medical School researchers.

The study, which appears in the April 14 Proceedings of the National Academy of Sciences, is the first to measure the size of the pores that line the blood vessels surrounding a variety of solid tumors. As solid tumors shrink, so do the microscopic pores in the blood vessels surrounding the tumors -- so much so that some tumor-killing agents may no longer be able to squeeze through vessel walls to reach their targets. In addition, first-time therapies may fail because therapeutic drug molecules are too bulky to pass through the blood vessel pores in the first place.

The findings suggest a fundamental change in the approach to designing chemotherapy agents. They may need to be smaller and more agile. "The public, investors, and the government are so enamored of the sophisticated new gene therapies. They're looking at Mercedes and Lamborghinis. But they don't think "can these new therapies get through the doorway?" says Rakesh Jain, the A. Werk Cook Professor of Radiation Oncology at Massachusetts General Hospital and Harvard Medical School.

Jain and his colleagues grew six different cancers in immune-deficient mice and measured pore sizes of surrounding blood vessels. The one human and five mouse cancers were grown at various places inside the mouse's body to represent what happens during metastasis. Working with hormone-dependent tumors, the researchers induced the tumors to shrink by cutting off their supply of hormones.

The researchers found that pore size varied among the six tumor types -- for example, mouse breast tumors had much bigger openings than mouse liver cancers -- and among sites within the same tumor. Solid tumors often contained a mixture of bigger and smaller openings. Pore size also varied depending on where the cells grew. Breast tumors grown in the brain, for example, had blood vessels with much smaller pores than did tumors grown under the skin. Pore size was also greatly reduced in the shrunken hormone-dependent tumors. In fact, withdrawal of testosterone resulted in a reduction of pore size from 200 nanometers (nm) to less than 7 nm within 48 hours in a testosterone-dependent tumor.

"That tells you that you just can't make a drug and expect it to work on all tumors," says Jain. Sue Hobbs and Wayne Monsky, an MD-PhD student and postdoctoral fellow, respectively, in Jain's lab, are lead authors on the paper.

The study casts new light on some old cancer puzzles. For example, some anticancer agents that work well in the petri dish fail to kill cancer cells in the body. The new findings suggest that such candidate drugs may never get a chance to prove their cancer-killing powers because they have been engineered too large for the microscopic doors they have to pass through.

Equally perplexing, some chemotherapies lose their efficacy after only a few rounds. Many have chalked up these failures to the development of resistance -- when successive rounds of chemotherapeutic agents allow impervious mutants to gain a foothold. "You can't talk about multidrug resistance if you can't get the drugs into the tumor," says Jain. He believes there is another explanation for such disappointing results: chemotherapeutic agents may, by shrinking tumors, be shrinking blood vessel pores -- in effect, closing the door behind them.

The implications are that drug makers must not only design agents that get through a variety of microscopic doors, they must also think about aiming their drugs at a moving target -- blood vessel pores that change in size over the course of treatment. Yet drug companies have been taking a one-size-fits-all approach, Jain says. In the area of gene therapy, the trend has been to design bigger and more impressive molecules.

Shimmying through shrinking blood vessel pores is just one of many challenges facing drug agents. Ten years ago, Jain showed that just getting into the blood supply surrounding the tumors is a difficult task. "Nobody had measured resistance to blood flow in tumor vessels before. It was like walking into a wide open field," he says. Jain and his colleagues showed that the capillaries, arteries, and veins surrounding tumors are a jumbled mass -- thick in some areas and pinched at others -- rather than a neat and orderly array. And to make navigation even more difficult, the blood flowing through them is highly viscous -- more like sludge than water.

Blood vessels surrounding tumors are also known to be very leaky in some areas but not in others. It was only recently with the development of his immune-deficient mouse model that Jain was able to grow enough kinds of tumors to tackle the question of blood vessel pores systematically. He and his colleagues injected fluorescently labeled particles of increasing size in to the blood vessels of the six tumors at various locations and watched to see which ones got out. They used the same approach to see the effects of tumor shrinkage.

Even if drugs are designed that can get through pores, they will face an uphill battle once they get inside a tumor. The pressure exerted by the material surrounding tumor cells, called the interstitium, is enormous, as Jain and his colleagues have recently shown. Jain believes that there are ways around such obstacles. For example, it may be possible to lower pressure by draining the interstitium inside tumors. In fact, Jain and his colleagues are currently working on such a tumor drain. "We are trying to understand and modify the barriers more and more," he says.

The research was funded in part by the National Cancer Institute and the National Institutes of Health.

Editors: A two-color graphic that helps illustrate this research is available from Harvard Medical School.

###



Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.