Getting More from Drugs

A biotech company called CombinatoRx has found that at the right doses, thousands of counterintuitive drug pairs are synergistic. The Cambridge, MA, company has eight drug combinations in clinical trials and several more in preclinical development. In a few years, diabetics, instead of injecting insulin, might be prescribed a cholesterol drug and a pain medication to help control their blood sugar. People suffering from chronic pain might find relief through a combination of a steroid and an antidepressant, with fewer side effects than they experience with current therapies.

Alexis Borisy, founder and CEO of the company, says his researchers take a brute-force approach to finding fruitful drug combinations. In the lab, they test combinations of several thousand drugs at several different doses on cellular models of diseases including cancer and arthritis–regardless of what diseases the drugs are currently approved for, if any. Then they feed the data into software that looks for synergies.

“We started the company based on theoretical ideas,” says Borisy: that cellular pathways in the body are intricate and redundant, and that it should be possible to find powerful combinations of drugs that attack the same disease pathways in different ways. Other drug companies are testing combinations of drugs, but they usually try a pain drug with a pain drug, or a cancer drug with a cancer drug.

Borisy’s company has found that drugs on the market for very different applications can have synergy, presumably because they are acting on the same networks of genes, proteins, and signaling molecules in cells. “We have found thousands of synergistic compounds we wouldn’t have expected,” says Borisy, Technology Review’s 2003 Young Innovator. (See “TR35.”)

For example, CombinatoRx has found several of what Borisy calls selective steroid amplifiers–drugs that, when given with steroids, amplify the latter’s good effects and dampen their side-effects. The company has two drugs in phase II clinical trials that include the steroid prednisolone, which can cause weight gain and insomnia and destabilize the blood sugar, among other side effects. In one formulation, for the treatment of rheumatoid arthritis, a low dose of prednisolone is combined with a drug called dipyrimadole, which is normally used to prevent blood clots in patients with cardiovascular disease. In the second formulation, for the treatment of chronic pain, a low dose of prednisolone is combined with a low dose of an antidepressant (which also has side effects at higher doses).

The company has a library of thousands of compounds–drugs approved in the United States and other countries, food additives, natural products, drugs under development by partner companies–that can be combined in millions of ways. Add to that the need to test multiple doses, and there are tens of millions of experiments to perform, says Borisy.

For each pair of compounds, CombinatoRx tests 36 dosages on several cell types. To test whether a combination has potential as a therapy for inflammatory diseases like arthritis, for example, researchers try the combination on white blood cells. They first stimulate the cells to mimic inflammatory disease, then administer the different dosages to the cells, and finally test for indicators of inflammation (such as signaling molecules that keep the immune system on alert). Data from these experiments are fed into a computer featuring software that determines whether the drugs have synergy and, if so, at which doses. If a combination shows promise, the company tests it in animals and then starts clinical trials.

In its search for therapies for cancer, metabolic disorders like diabetes, and neurodegenerative disorders like Huntington’s disease, the company performs similar experiments on fat, muscle, nerve, and prostate-cancer cells, among others.

CombinatoRx’s rheumatoid-arthritis and chronic-pain drugs will be entering another round of large phase II clinical trials later this year. When Borisy founded the company in 2000, he says, its approach was considered provocative. “Now we have shown that it really works in huma