In January 2011 the highly anticipated drug eritoran, made by Tokyo-based Eisai, was withdrawn after clinical trials showed no benefit in comparison with a placebo in preventing sepsis deaths. By October of that year the drug that eritoran had been intended to compete with—Xigris, made by Eli Lilly—was withdrawn from sale 10 years after being licensed by the U.S. Food and Drug Administration because mandatory postmarketing studies showed no benefit to patients. In February 2012 Agennix in Munich, Germany, halted studies that applied its existing cancer drug, talactoferrin, to sepsis after the number of deaths among patients receiving it in rigorous clinical trials turned out to be greater than the number of deaths among placebo recipients. Later that year, in August, research into the new drug CytoFab was canceled by partners AstraZeneca and BTG after an early trial showed no effect greater than a placebo.
All those drugs were predicted to be billion-dollar blockbusters if they had proved effective against sepsis, but instead they cost their creators millions in foregone research and canceled trials. As a result, “I think the big companies have given up,” says Richard P. Wenzel of Virginia Commonwealth University, a longtime sepsis researcher.
How did such a diverse array of compounds, from so many different companies, fail? Investigators have suggested several reasons. A 10-year study published online in February revealed serious flaws in the animal research on which sepsis drug trials were based. Severe inflammatory responses in mice, it turns out, do not accurately mimic sepsis in people—so a drug that helps mice could make things worse for a person. In addition, patients in sepsis trials may have been too sick to respond well to any drug. Because sepsis patients are seldom diagnosed early, by the time they are enrolled in a trial they are likely to be more challenging to treat. Furthermore, people with sepsis may not all be sick in exactly the same way, so individuals may respond quite differently to the same compound.
“Where we are currently is similar to diagnosing people with ‘cancer’ without giving them any further information—not that they have leukemia, not what cell type, not what molecular abnormalities are present,” O'Brien explains. “‘Sepsis’ is a bucket that contains different groups of patients. We have not done the groundwork of separating out their underlying pathophysiology.”
Despite the many disappointments, researchers continue to search for treatments that could halt sepsis before it becomes deadly. Spectral Diagnostics in Toronto, for example, pairs a diagnostic blood test with a therapeutic device. The test looks for endotoxin, a molecule that is released from dying bacteria and can trigger the start of sepsis. About 50 percent of sepsis patients, often the most critically ill, have high levels of endotoxin in their blood. Blood is drawn from these patients and pumped over a filter infused with an antibiotic that binds to endotoxin, removing the molecule before the blood is returned to patients. The FDA has approved the test (the filter already existed, and Spectral has licensed it), and the combination of diagnostic and device is currently being tested in 14 U.S. states and in Canada in a large trial.
Spectral acknowledges that its treatment, even if successful, would not help patients who do not have high levels of endotoxin. But “if our trial turns out positive, we could potentially save 50,000 lives per year,” says Paul Walker, Spectral's chief executive officer and a critical care specialist.
Richard Hotchkiss, who is a professor of anesthesiology, medicine and surgery at the Washington University School of Medicine in St. Louis, has proposed a very different approach—not just to treatment but to the whole way of thinking about sepsis as well. He sums up his view of the current research landscape as: “Just because you're traveling down a well-worn path doesn't mean you're heading in the right direction.”