Two bacteria, injected into the bloodstream of mice, wiped out tumors without help from the immune system. A new study shows how they do it, and the mechanism is not what most people would expect. Researchers at the Japan Advanced Institute of Science and Technology, with the University of Tsukuba and Daiichi Sankyo, tested a pair of naturally occurring bacteria called AUN. One is Proteus mirabilis (called A-gyo), isolated from inside a tumor. The other is Rhodopseudomonas palustris (called UN-gyo), a purple photosynthetic bacterium. No genetic engineering was used. The idea of using bacteria against cancer is old. In the 1890s, William Coley noticed that some patients improved after infections and began injecting bacteria on purpose. Most modern versions still depend on the immune system to finish the job. That is a problem. Chemotherapy and radiation suppress immunity, and solid tumors have a low-oxygen core (hypoxia) that immune cells struggle to enter. Checkpoint inhibitors and CAR T cells, the newer immunotherapies, also rely on those immune cells. This study asked whether bacteria could kill tumors on their own. Key findings: • Tumors disappeared completely in the treated mice, and the mice lived longer. This held in mice with normal immune systems, mice with weakened immune systems, and mice carrying human tumors • The mice showed no visible signs of toxicity. There was also no cytokine storm, a dangerous overreaction where the immune system floods the body with inflammation, which is a known risk with some cancer immunotherapies • The two bacteria were mixed at about 3 parts A-gyo to 97 parts UN-gyo. In the lab, they settled back into that same balance within a week, even when researchers started them at very different ratios • Inside tumors, the balance reportedly flips to about 99 parts A-gyo to 1 part UN-gyo, according to the institute's press release • A-gyo was missing the genes for the sticky surface structures bacteria normally use to attach to tissue and cause infection. That likely makes it less harmful than its commercial cousin • Treated tumors were full of blood clots and showed widespread tissue death So this is what seems to be happening from mechanistic angle: First, the bacteria reach the tumor. Tumors are a hospitable place for bacteria. They have low oxygen, leaky blood vessels, and a suppressed local immune response. The bacteria grow there, form biofilms (sticky bacterial communities), and also kill cancer cells directly. Second, they trigger clotting inside the tumor. Cytokines (inflammatory signaling proteins) rise, platelets aggregate, and fibrin (the protein mesh in blood clots) builds up. This is thrombosis, the formation of clots within vessels. Third, the vessels collapse. Think of cutting the water supply to a building. A tumor needs a dense network of new blood vessels to bring in oxygen and nutrients. Clot them off, and the tissue behind them starves and dies. The partnership also appears to act as a safety system. UN-gyo dampens A-gyo's pathogenicity, reduces its ability to rupture red blood cells (hemolysis), and helps push A-gyo into a long, filament-like form in the presence of cancer metabolites. In other words, the second bacterium keeps the first one in check outside the tumor. Big caveats, though. This was all done in mice. Mouse tumors, mouse blood clotting, and mouse immune responses differ from ours, and complete remission in mice has a poor track record of translating to people. Injecting live bacteria into the blood, in people who may already be sick, carries infection risks that mouse studies can't fully capture. The same clotting mechanism that destroys tumors could be dangerous if it occurred outside them. And the study was led by a group that is preparing to launch a startup, which is worth keeping in mind when reading the safety claims. This is, however, a new way to think about cancer, one of the leading causes of death in older adults. The rate of cancer rises steeply after 50, and most current treatments depend on an immune system that also weakens with age and with treatment. A therapy that works without immune cells would matter most for exactly the people who need it most. The broader point is that the tumor itself is an ecosystem, and bacteria that live there may be turned into a tool. Whether that tool is safe enough for people is the question that remains.