Large screening strategy uncovers dozens of non-oncology drugs that kill cancer cells

Attempting to invent a new drug from scratch is an incredibly risky endeavor. More than half the drug candidates that make it to the final phase of multimillion-dollar clinical trials fail just before the finish line. One increasingly attractive way to lower drug discovery’s risk and cost is to repurpose drugs that have already been shown to be safe in humans. Yet so far, successful drug repurposing has mostly been serendipitous.

In a new study, researchers report a faster, more systematic approach to repurposing (Nat. Cancer 2020, DOI: 10.1038/s43018-019-0018-6). They tested 4,518 mostly noncancer drug compounds for activity against 578 cancer cell lines. The drugs were from Broad Institute of MIT and Harvard’s Drug Repurposing Hub library. The screen revealed a surprising amount of activity, with more than 1,400 compounds slowing the growth of cancer cells, says the study’s lead author, Steven M. Corsello, a researcher at Broad Institute and an oncologist at the Dana-Farber Cancer Institute.

To run the screen, the researchers developed a high-throughput method called “profiling relative inhibition simultaneously in mixtures” (PRISM). Cell-based assays tend to be slow because of the time it takes to prepare and grow separate cell lines. The team sped up the process by inserting DNA bar codes into the various cancer cell lines, which allowed the researchers to put multiple cell lines in the same well on a plate instead of having to prepare each cell line individually.

The PRISM screen uncovered 1,448 active compounds, about half of which had not been developed for treating cancer. The team chose to further investigate 49 compounds according to two criteria: selectivity, which is the compound’s ability to kill some cancer cell lines over others, and how predictable the activity was on the basis of genetic biomarkers in the cells.

The researchers probed four of those compounds in depth. Most notably, Corsello says, they found that cancer cell lines with high levels of the MDR1 gene, which is associated with resistance to established cancer treatments, were sensitive to the drug tepoxalin, used to treat arthritic pain in dogs. Additionally, they found that tepoxalin killed the cancer cells most likely through an off-target mechanism.

These results are a promising first step to repurposing drugs, Corsello says, and they provide a starting point for new drug-discovery campaigns as researchers discover new biological targets for cancer.

Drug repurposing “is something on everyone’s mind,” says Gina Xiaojing Wang, a drug-discovery expert at Genentech. “People are always curious about whether a compound that’s already proven safe could be used in oncology settings.” Many drug companies, like Genentech, regularly screen hundreds of compounds in their existing drug libraries for new uses, Wang says. This new screening method allows compounds to be profiled quickly, which she says could be “more powerful down the road”when chemists want to screen thousands or even millions of compounds against numerous cell lines.

“The most surprising observation to me, though, was that in most cases the mechanism of anticancer action had little to do with the biological target the molecule had originally been optimized for,” says Timothy Cernak, a medicinal chemist at the University of Michigan. The finding, he adds, “illuminates that as hard as we try to invent really selective molecules, they still interact with many other proteins in the body.”

Cernak says one major question that researchers will need to address is whether the repurposed drugs have an anticancer effect at the same dose that’s already been approved as safe for humans.