A failing grade
Results of preclinical drug testing with animals may not offer the assurances of safety and effectiveness that we often think they do, according to a study published in PLOS Biology.
After testing a new drug on animals, researchers usually write up their findings in an investigative brochure to submit to regulatory agencies and review boards that must approve human trials. But, according to the study, the brochures often lack crucial information that the agencies and review boards need to adequately evaluate the animal trials.
The researchers looked at 109 brochures, which reported on 708 animal studies that institutional review boards approved between 2010 and 2016. They found that less than 5 percent of the brochures described any measures the researchers took to assure the validity of the study, and 89 percent did not offer documentation that the researchers secured peer review or independent evaluation. Only 6 percent of investigational brochures reported results showing that the tested drug proved ineffective, causing the researchers to raise the possibility of bias in data reporting.
According to a recent study appearing in Biostatistics, less than 10 to 15 percent of human clinical trials prove successful. The new study helps explain why so many results in animal tests don’t hold up in human trials, Malcolm Macleod, a neurologist at the University of Edinburgh, told Science Magazine. —J.B.
A simple shot to cure cancer?
Stanford University researchers will soon begin the first human trial of a cancer treatment that cured up to 97 percent of tumors in mice. The treatment consists of two immune-stimulating agents injected directly into solid tumors.
The therapy is rapid, relatively inexpensive, and less likely to cause the adverse side effects often produced by methods that stimulate the entire immune system. In the study, published in the peer-reviewed journal Science Translational Medicine, the researchers transplanted two lymphoma tumors at two different sites in 90 mice. In 87 of the mice, both the treated tumor and the untreated cancer regressed. The tumors returned in three of the mice but regressed after a second treatment. The researchers obtained similar results in mice with breast, colon, and melanoma tumors.
Although approval from the Food and Drug Administration remains a long way off, Ronald Levy, the lead researcher, believes if approved the therapy could eradicate many tumor types. “I don’t think there’s a limit to the type of tumor we could potentially treat, as long as it has been infiltrated by the immune system,” he said in a statement. —J.B.
Experts condemn AI weapons research
Nearly 60 artificial intelligence experts from more than 30 countries recently sent an open letter to KAIST, a public university in Daejeon, South Korea, warning against its artificial intelligence program. The experts fear the KAIST Research Center for the Convergence of National Defense and Artificial Intelligence, in collaboration with Hanwha Systems, South Korea’s leading arms company, is working to develop weapons with artificial intelligence that could act independent from human control.
The letter writers threatened to boycott all collaboration with any part of the university until its president affirms that the center will not develop autonomous weapons. The authors indicated they previously sought such assurance with no success.
Autonomous weapons “will permit war to be fought faster and at a scale greater than ever before” and will allow terrorists to use them against innocent populations, the letter stated. Urging KAIST to ban this technology, the experts wrote, “This Pandora’s box will be hard to close if it is opened.”
The United Nations is currently discussing ways to contain the threat posed by the development of autonomous arms. —J.B.