Sepsis kills more than 140,000 people every week. It occurs when the immune system overreacts to injury or infection: it’s difficult to treat—even with antibiotics—and it’s also hard to diagnose . Patients often don’t receive treatment until hours or days after their initial injury.

At the Center for Engineering in Medicine at Massachusetts General Hospital, a team led by Biju Parekkadan has found a promising lead for treating this global scourge. The team’s study, published in Science Translational Medicine in August 2014, looked at the potential of using fibroblastic reticular cells (FRCs), a type of cell found in lymph nodes that is instrumental in maintaining and repairing tissue.

Sepsis begins when the immune system recruits cytokines (a kind of signaling molecule), which in turn attract additional immune cells to the site of an infection or injury. In a vicious cycle, those extra immune cells stimulate still more cytokines, leading to an inflammatory cascade called a cytokine storm.

Researchers have tried various drugs to interrupt this chemical cascade without success. They’ve also attempted to thwart sepsis by improving early diagnosis, using tools such as a blood test that would detect endotoxin, a molecule released from dying bacteria that can trigger a septic response. And a few researchers, including Parekkadan, are taking a somewhat counterintuitive approach. Rather than simply trying to stifle the body’s immune response to the bacteria, they aim to influence it in a strategic way.

Parekkadan sought to address this runaway inflammation with a drug made of whole, living cells. Compared with treatments based on single molecules, which typically affect just one target, whole-cell therapies can influence multiple sites in the body and respond to their environments. The team chose FRCs because they regulate several aspects of the immune response: If a treatment could leverage the natural function of FRCs, it might help the body modulate the overreaction that sepsis brings about.

The initial research on mice looked at how FRCs worked on sepsis induced in two ways. Some mice were given an endotoxin, and other mice had their intestines punctured—an injury that triggers a major septic response. In both cases, researchers infused the therapy into the mice’s abdominal cavity.

The researchers also looked at how well the treatment worked at a later stage of infection, to address the reality that sepsis can be difficult to diagnose quickly. Some mice were treated four hours after they were injured, and others after 16 hours. Because human sepsis patients receive antibiotics and fluids, the mice got those as well. And the research team used both young and elderly mice, because sepsis occurs in elderly humans far more often than in younger people.

FRCs significantly improved survival in mice infected in both ways. Even by the 16-hour mark, the FRCs still worked well, with a 44% survival rate. And both young and elderly mice benefited.

The next puzzle is to figure out exactly how the treatment works. Parekkadan’s team found that the FRCs prevented damage to the spleen and the death of immune cells that are found there. A separate finding suggests that the FRC treatment had a widespread rather than a localized effect on the immune system.

Translating this promising cell-based therapy to humans is still a ways away, but Parekkadan’s team is moving forward with additional studies and research.