ABOVE: © ISTOCK.COM, ANTONIOKHR

The failure of drugs that seemed promising in laboratory tests to prove themselves in human testing is an all-too-common occurrence, dashing hopes and costing billions of dollars. The reasons for a particular drug’s flop are often mysterious, but a study published yesterday (September 11) in Science Translational Medicine finds that one cause may be misidentifying the protein targeted by a therapy.

“It’s hard enough to develop drugs when you know their mechanism of action,” cancer researcher William Kaelin of the Dana-Farber Cancer Institute in Boston tells Nature. “It’s really difficult when you don’t know the mechanism of action.” Kaelin was not involved in the study.

The project began when researchers, led by Jason Sheltzer of Cold Spring Harbor Laboratory, used the gene-editing technique CRISPR to disable the gene for a protein called MELK in cultured cancer cells. MELK was thought to be essential to...

The team went on to perform similar experiments with six other proteins targeted by 10 drugs currently in the development pipeline for cancer. In all cases, the drugs did kill cells, even when the supposed targeted protein wasn’t present. 

Based on these and other experiments, the researchers concluded that the drugs don’t target the proteins they were thought to home in on—which could lead to unanticipated side effects and the selection of cohorts of clinical trial participants who aren’t the most likely patients to benefit from the drugs. 

“A lot of the drug targets that are in clinical trials today were discovered with the best technology from five or 10 years ago,” Sheltzer tells The New York Times. That is, they were identified with RNA interference (RNAi), meaning researchers used targeted strands of RNA to block protein production from a given gene in cancer cells grown in the lab. If that interference killed the cells, then scientists would look for a drug compound that could target the same protein and block its action. But the paper’s authors suggest RNAi blocked the production of more than one type of protein, leading drug developers to misidentify the true targets of the drug compounds used in the current study.

“There clearly exists a legacy of RNAi-guided bad targets that needs to be purged from the drug development pipeline,” MD Anderson Cancer Center cancer biologist Traver Hart, who was not involved in the study, tells the Times.

The new findings come with caveats. In an interview with Nature, coauthor Ann Lin of Stanford University notes the study was carried out in cultured cells, leaving open the possibility “that these drug targets are essential in human patients.” And the developer of one of the drugs tested, Paul Hergenrother of the University of Illinois Urbana-Champaign, tells Science that in addition to the target the researchers knocked out with CRISPR, the drug also activates another protein with a similar function. So Sheltzer’s team would have needed to knock out the genes for both proteins in order to eliminate the drug’s effects, he says.

Sheltzer tells Nature that the study results open up fresh possibilities. “There is an unexplored world of cancer targets out there,” he says. “By using CRISPR and other technologies to examine these drugs, we might unlock new targets.”

Shawna Williams is an associate editor at The Scientist. Email her at swilliams@the-scientist.com or follow her on Twitter @coloradan.

Interested in reading more?

an illustration of dividing cancer cells

The Scientist ARCHIVES

Become a Member of

Receive full access to more than 35 years of archives, as well as TS Digest, digital editions of The Scientist, feature stories, and much more!
Already a member?