Ten targeted cancer drugs currently being tested in clinical trials involving 1,000 patients do not reach the targets at which they are aimed, according to research published this week.
Researchers at Cold Spring Harbor Laboratories showed that the 10 compounds are cytotoxic even when cancer cells lack the proteins that they are designed to inhibit.
Each of the compounds is aimed at the products of genes that have been described as being essential for cancer cell survival on the basis of RNA interference knockdown phenotypes. The researchers suspect the lack of specificity of that technique is the cause of the targeting discrepancies.
The scientists used more precise CRISPR gene editing techniques to examine the mechanisms of the 10 cancer drugs, which inhibit one of six proteins that have been reported as important for the survival of cancer cells in more than 180 publications.
The drugs have been tested in at least 29 different clinical trials and include prominent candidates such as the HDAC inhibitors citarinostat and ricolinostat, which are being tested against multiple myeloma.
"I think that there are some fundamental issues our work raises: in particular, RNAi promiscuity is pretty common, and proving that a drug interacts with a protein in vitro doesn't mean that interaction is relevant in cells or organisms," Jason Sheltzer, fellow at Cold Spring Harbor, told BioWorld.
As Sheltzer notes, 97% of all cancer drugs tested in humans fail to get FDA approval. Lack of efficacy and dose-limiting toxicity are the most common causes, but said Sheltzer, "I think this extremely high failure rate suggests there are some fundamental issues in how any new drug targets are studied and how new drugs are characterized."
Those issues threaten to undermine the precision medicine premise of finding a particular gene that is mutated or overexpressed, identifying that biomarker in a patient's tumor and treating with a drug thought to inhibit the protein for which the gene encodes.
The team focused on drugs for which no resistance mutations have been described. Resistance mutations are the gold standard for establishing a drug's on-target action.
"When you have a drug, but you have no known resistance granting mutation, then I think in those instances our results suggest that off-target toxicity is likely to be a common phenomenon that you observe," Sheltzer told listeners of a teleconference held to discuss the research, which is published in the Sept. 11, 2019, issue of Science Translational Medicine. "And what we would suggest is that finding these resistance granting mutations can really help clinicians and scientists figure what a drug is really doing."
The researchers used CRISPR to remove the supposed targets of the 10 drugs from a variety of cancer cell lines. "We were very surprised to find that when we eliminated these proteins that the cancer cells continued to grow just fine, in spite of what had previously been published," he said.
The second surprise was that tumor cells totally lacking the target died when they were treated with drugs that are supposed to be specific for the absent target.
"They must be killing the cells through some other, off-target effect," Sheltzer said. "We have 10 potent anticancer drugs, but our data suggest the mechanism of action is incorrect."
To investigate further, the researchers studied OTS-964, which is said to target PBK, the gene which codes for lymphokine-activated killer T-cell originated protein kinase (TOPK). They found the drug remained effective in PBK knockout cancer cells.
By systematically knocking out other genes, the researchers showed OTS-964 actually targets cyclin-dependent kinase 11 (CDK11). As Sheltzer noted, while other members of the cyclin-dependent kinase family have been targeted and translated through to approved therapies, OTS-964 is the first known drug to target CDK11.
Recent data on OTS-964 in combination with temozolomide in treating glioblastoma, published in March, refers to OTS-964 as a TOPK inhibitor. Meanwhile, overexpression of CDK11 has been implicated in a number of other cancers.
It seems clear from these findings that poor understanding of the actual target of a drug may have led to failures in clinical trials. Ten years ago, RNAi was really the only method to study a gene in a human cancer cell line. Now CRISPR has come along, and although far from flawless, it is much better.
But, said Sheltzer, "I think that the more orthogonal approaches you take towards drug and target discovery, the more likely it is to be successful."
Sheltzer told BioWorld that to some extent, these issues may arise in other disease conditions apart from cancer, particularly if a lead target was discovered with RNAi. "Our results may extend beyond cancer. I don't have any specific examples of that at the moment, though," he said.
And although it would be more complicated, Sheltzer said using CRISPR gene editing could be applied to look at the precision of antibody targeting. "I think the fundamental idea of studying drug and antibody targets is sound and people should consider these approaches," he said.
The researchers admit there are caveats to their work, and that drug targets could be essential in humans and not in vitro. It also could be the case that other proteins take over the role of one that is knocked down by CRISPR.
But, they noted, it is important to look at all possible ways of increasing the probability a drug will be successful in clinical trials.
Sheltzer said he had tried to get safety and efficacy data relating to the 10 drugs in clinical trials from the FDA and from the companies concerned, but they have not been prepared to disclose the information.
Sheltzer's project started out with the entirely different goal of identifying genes tied to low survival rates among cancer patients. That led to the discovery that MELK (maternal embryonic leucine zipper kinase), which dozens of academic papers previously identified as absolutely essential to the survival of cancer cells, has absolutely no influence on cancer growth.
The intention of the latest work was to find out if MELK is an aberration. At the end of their project, the researchers ended up with a list of potent cancer drugs where the putative mechanism of action has been invalidated. They now plan to find out what the actual targets are.
They also are taking forward their discovery that OTS-964 is a CDK11 inhibitor and will screen cancer cells to find those where inhibiting CDK11 is particularly effective.