By age 38, the patient’s breast cancer had spread to her bones, a typically fatal turn of events. She became an alcoholic, and her doctors stopped all cancer treatment, instead giving her a drug to discourage her drinking. She died 10 years later, after an inebriated fall from a window. But an autopsy revealed something unexpected: Her bone tumors had melted away, leaving only a few cancer cells in her marrow.
That 1971 case report, along with numerous lab studies, have suggested that the 6-decade-old drug disulfiram (commercially known as Antabuse), which makes people feel sick from drinking small amounts of alcohol, might also be a cancer fighter. Now, researchers have finally figured out how—by blocking a molecule that is part of a process that gets rid of cellular waste.
“This paper solves a puzzle that has persisted in cancer research for decades,” says cancer biologist Michele Pagano of New York University School of Medicine in New York City, who was not involved in the study.
Starting in the 1970s, scientists found that disulfiram killed cancer cells and slowed tumor growth in animals. It increased survival in women who had breast tumors removed in a small clinical trial published in 1993. But since then, disulfiram hasn’t gotten much attention for treating cancer, partly because scientists disagreed about how it worked.
In the new study, a Danish-Czech-U.S. team first firmed up the drug’s anticancer effects by combing through Denmark’s unique cancer registry—more than 240,000 cases diagnosed between 2000 and 2013, along with data on the medications each patient took. Of the more than 3000 patients taking Antabuse, the cancer death rate was 34% lower for the 1177 who stayed on the drug compared with those who stopped taking it, the researchers report today in Nature. The drug was an equal opportunity anticancer weapon; its benefits held for prostate, breast, and colon cancer, as well as cancer overall.
The researchers also confirmed that disulfiram slows the growth of breast cancer tumors in mice, particularly if combined with a copper supplement, which was already known to enhance its effects. They then showed that when the mice broke down disulfiram, its main metabolite, ditiocarb, forms a complex with copper that blocks the machinery that cells use to dispose of misfolded and unneeded proteins. “Everything is frozen,” says cancer biologist Jiri Bartek of the Danish Cancer Society Research Center in Copenhagen, a co-leader of the study. Partly because of the resulting protein buildup, the cancer cells become stressed and die.
Although some approved cancer drugs and others in development interfere with the same protein cleanup process, known as the ubiquitin-proteasome system, disulfiram targets only a specific molecular complex within this machinery. That could explain why it is so effective, Pagano says. Bartek’s team also solved another puzzle—why normal cells aren’t harmed by disulfiram, even when patients take it for years. For unclear reasons, the copper metabolite is 10 times more abundant in tumor tissue compared with other tissues, the group found.
Despite the compelling 1971 anecdote, disulfiram probably “is not a cure” for most cancer patients, cautions cancer biologist Thomas Helleday of the Karolinska Institute in Stockholm. However, the drug could help extend the lives of patients with metastatic cancer—it’s already shown evidence of doing so when combined with chemotherapy in a small lung cancer trial. Bartek and collaborators are now launching trials to test a disulfiram-copper combo as a treatment for metastatic breast and colon cancers and for glioblastoma, a type of brain cancer.
Finding a new use for an approved drug is appealing because the compound has already passed safety testing. However, “big pharma probably won’t be interested” in developing disulfiram for cancer because there’s no patent protection on the drug, Bartek says. Still, if the pending clinical trials provide convincing evidence, Halleday points out, oncologists could go ahead and prescribe it anyway as an inexpensive treatment.