A syringe full of noxious bacteria sounds like the last thing a cancer patient needs. But a new study of dogs with tumors, and even one human cancer patient, reveals that injecting certain bacteria directly into the growths can shrink or even eliminate them. The results strengthen the case that using bacteria to treat cancer, an approach that performed poorly in some clinical trials, will work.
Doctors first noticed that bacterial infections sometimes diminished or even eradicated tumors more than 200 years ago. William Coley, a surgeon in New York City, was the first to run with this idea. In the 1890s, he began injecting cancer patients with live Streptococcus bacteria to combat their tumors. After two recipients died from infections, he switched to administering dead bacteria and ended up treating more than 1000 patients with his so-called Coley toxins. Coley sometimes injected the bacteria into tumors and sometimes into the bloodstream, and many of his patients survived. But treatments such as radiation, chemotherapy, and surgery soon pushed Coley’s approach into the history books. Still, a 1999 reanalysis of some of his cases suggested his success rate was about the same as for modern cancer therapies.
Recent attempts to revive bacterial cancer treatment have run into obstacles. For example, a clinical trial in which patients received intravenous doses of weakened Salmonella bacteria found that the treatment was safe but had little impact on tumors. For more than a decade, cancer geneticist Bert Vogelstein of Johns Hopkins University in Baltimore, Maryland, and colleagues have investigated a different bacterium, the soil -dweller Clostridium novyi, a relative of the microbe responsible for botulism. Oxygen is scarce inside tumors, and these bacteria “love areas of low oxygen,” says Saurabh Saha, a cancer researcher at BioMed Valley Discoveries Inc. in Kansas City, Missouri, and a co-author on the new study, which appears online today in Science Translational Medicine. “They grow and divide and kill the cancer cells,” Saha says. The researchers hypothesize that the bacteria release enzymes that destroy the tumor cells, and then they feast on the debris.
Injecting spores from the bacterium into the brain tumors of rats extended the animals’ survival, the researchers found. But treatments that work in lab rodents have a bad habit of failing in people, so the researchers wanted to test the bacteria in animals that more closely resemble human cancer patients. They chose dogs. Like humans, dogs are more genetically diverse than lab rodents are. And like human tumors, dog tumors sprout spontaneously, in contrast to the researcher-induced tumors of lab rodents.
Saha and colleagues injected C. novyi spores into the tumors of 16 pet dogs whose owners had run out of options for treating them. In six of the dogs, the tumors shrank or disappeared, and the tumors stopped growing in another five animals. Several dogs needed surgery to clear the wounds as the tumors disintegrated.
Bolstered by results of the animal studies, the researchers began a safety trial of the treatment in people. The first person who received the bacteria was a woman whose abdominal tumor had metastasized to several parts of her body, including her right shoulder. Although the researchers injected less than 1% of the bacterial dose the dogs had received into the shoulder metastasis, the growth began to dwindle. However, the treatment spurred an unusual side effect. The tumor had barged into the humerus, the bone in the upper arm, and was apparently providing physical support. Destruction of the cancer cells led the bone to break, which required surgery to repair. Eventually, the patient died from her other metastasized tumors.
The bacteria not only destroy tumor cells, but they also spur immune cells to attack the cancer, the researchers previously showed. And because the microbes survive only in the oxygen-poor milieu of a tumor, the treatment is specific, Saha says. “It distinguished the tumor from normal cells.” The researchers plan to continue their safety trial and want to determine which types of tumors will respond to bacterial therapy, he says.
Saha and colleagues altered their bacteria to be less dangerous to people, and the microbes die when they contact oxygen, which limits their ability to spread. Nonetheless, some of the dogs and the human patient in the trial received antibiotics, and doctors and other caregivers used standard anti-infection measures, such as wearing protective gowns and gloves.
The study is significant because it provides “proof of the concept that this particular approach can have antitumor activity in ‘real tumors’ ” rather than just in the induced tumors of lab rodents, says Douglas Thamm, a cancer biologist and veterinary oncology researcher at Colorado State University’s Flint Animal Cancer Center in Fort Collins. Doctors might need to pair the treatment with other therapies, such as radiation, to mop up any tumor cells that escape the bacteria, he says.
“It’s a very good and very important paper,” adds cancer biologist Robert Hoffman of AntiCancer Inc., a San Diego, California–based biotech firm. He and his colleagues have shown that a different strain of Salmonella bacteria than the one used in previous clinical trials could eradicate various types of tumors from mice, but they have not performed any studies on human patients.
One concern about the new approach is that most cancer patients aren’t killed by the original tumor, but by metastases. Thamm and Hoffman worry that injecting C. novyi directly into tumors will leave these deadly metastases untouched. “If bacterial therapy is going to be widely available and efficacious,” Hoffman says, “it has to target metastatic disease.”
*Correction, 14 August, 2:45 p.m.: The researchers did not genetically alter bacteria to attenuate them; they used heat.