Mice fought off cancer — using bacteria, according to a new study published today in Science Translational Medicine. Basically, the bacteria narced on the cancer, giving the mice’s immune system the heads up to come fight it.

Scientists injected mice with a strain of engineered salmonella that infiltrated the tumors. The salmonella had been tweaked to release a protein named flagellin B, or FlaB, which is found in a deadly flesh-eating pathogen people sometimes contract after eating raw oysters. The researchers chose this protein because they knew it would trigger the immune system — which is what the FlaB-producing salmonella did. That immune response shrank the tumors to the point that they were no longer detectable in 11 out of 20 mice, and even prevented the cancer from spreading. The bacteria also didn’t invade healthy tissues in the mice, suggesting it’s got limited side-effects. “We believe that our bacterial strain is extraordinarily safe to be used as a therapeutic,” study co-author Joon Haeng Rhee wrote in an email to The Verge.

Much more research needs to be done in other animals, such as dogs and our fellow primates, before the bacteria described in today’s study can be tried on people. But the results seem promising, and add to the body of knowledge on what could be the next big cancer therapy. “This paper peels back a little bit of a window into the mechanisms perhaps by which bacterial-based therapies may be stimulating the immune system,” says Saurabh Saha, a cancer researcher who’s studied the cancer-fighting properties of Clostridium novyi bacteria and was not involved in the study.

The idea to use bacteria to fight cancer has been around since at least the 1800s. Back then, an American bone surgeon and cancer researcher named William Coley observed something weird: some cancer patients who developed infections after tumor-removal surgeries were more likely to recover. So he began injecting patients with Streptococcus bacteria and other bacterial products, which caused tumors to shrink. The so-called Coley’s Toxins were injected into more than 1,000 cancer patients, but criticism of his methods — as well as the development of radiation therapy and chemotherapy — caused his treatment to pretty much disappear from research labs.

Now, a hundred years later, scientists have revisited the idea, trying different types of bacteria to encourage the immune system to fight tumors. The work is part of an emerging field called immunotherapy, and it’s growing fast. That’s because unlike radiation therapy and chemotherapy, which hurt healthy cells alongside the cancerous ones, immunotherapy promises to target only cancer cells. In 2014, scientists led by Saha injected innocuous strains of C. novyi bacteria into 16 dogs who had soft-tissue sarcomas and even one person with a rare type of cancer. The bacteria shrank the size of the tumors; in a few of the animals, the tumors completely disappeared. In 2015, scientists modified a strain of salmonella to attack tumors in mice.

The authors of today’s study also used salmonella — a kind of bacteria you may know from food poisoning. Salmonella thrives in environments with no oxygen, and tumors have no oxygen and a lot of dead cells the bacteria can feed on. The tumor is “a very favorable home for salmonella,” says Roy Curtiss III, a professor at the University of Florida’s College of Veterinary Medicine, who’s studied the use of salmonella as a cancer therapy and did not take part in today’s study. “It’s a good food supply.”

The researchers engineered Salmonella typhimurium to make it less infectious and to produce the FlaB protein. When the bacteria were then injected into the mice with human colon cancer, the salmonella stimulated the activity of different types of white blood cells. These cells attacked the tumors, shrinking their mass and even keeping the cancer from spreading to other parts of the body. The salmonella was also shown to be fairly safe: the number of bacteria in the tumor cells were 10,000-fold higher than in other vital organs in the mice.

The results matter because they show how these bacteria are stimulating the mice’s immune system, says Saha, who’s a venture partner at Atlas Venture. Before it can be used in humans, however, it will have to be tried on other animals like dogs and monkeys. “Mice are nice but they’re not humans,” Curtiss says. “They can mislead sometimes, so having other animal models is advantageous.”

In the future, this type of bacterial therapy might be combined with other types of anticancer immunotherapy, boosting the body’s immune response to tumors. A lot of research still needs to be done, of course, but this finding is promising. “It shows that what was done 120 years ago with Coley’s Toxins deserves to be revisited again today, using bacteria as an adjuvant to stimulate the immune system to fight cancer,” Saha says. “I think it’s a very important modality, and one that we should continue pressing forward on to learn more about.”