Sepsis is widely believed to be due to an out-of-control inflammatory response, and efforts to develop drugs to treat that condition have focused on reining in the immune system.

Those approaches, to put it mildly, haven't worked. More than 25 drugs have failed in clinical trials in as many years. In fact, so have approved drugs: In October, Eli Lilly and Co. said it was withdrawing Xigris [drotrecogin alfa (activated)] after a study failed to demonstrate clinical benefit. Xigris, which had been approved by the FDA in November 2001, was the only product approved specifically for patients at high risk of death from severe sepsis.

Now, new human data published in the Dec. 21, 2011, issue of the Journal of the American Medical Association may explain why sepsis drug development is such a bloody battlefield. In trying to suppress immune activation, drug developers and doctors very often may be barking up exactly the wrong tree.

'A significant number of sepsis patients are dying from immune suppression,' Richard Hotchkiss told BioWorld Today. Hotchkiss is at Washington University in St. Louis and the paper's senior author.

His team's conclusion flies in the face of conventional wisdom, which holds that sepsis patients die from hyperimmunity. 'They can die from that, early in the disease,' Hotchkiss said. But as the disease progresses, the immune system switches from hyperactive to suppressed.

Hotchkiss, in fact, compared later-stage sepsis patients to AIDS patients, because his team found that they showed a marked loss of killer T cells. 'But they also lose additional components of the immune system, including both innate and adaptive immunity,' he added.

Hotchkiss, co-author Jonathan Green, also at Washington University, and other colleagues decided to test whether sepsis patients become immunosuppressed as their illness progresses for several reasons.

Some clinical studies had found that sepsis patients can suffer from a reactivation of latent infections, which would be more consistent with immune suppression than hyperactivation. Animal studies had shown such immune suppression. But no direct clinical evidence existed either way.

In their studies, Hotchkiss, Green and their colleagues compared the postmortem spleen and lung tissues of patients who had died from sepsis caused by infections in the ICU to patients who had died from other causes, such as trauma.

Hotchkiss cautioned that, although some people define sepsis as a hyperinflammatory response due to any cause, his team restricted itself to sepsis that was secondary to infections, and the results were not valid for hyperinflammation that was due to other causes.

They found that, a few hours after the autopsy, spleen cells taken from deceased sepsis patients produced less than 10 percent of the amount of cytokines that spleen cells from controls did. Immune cells also expressed fewer surface molecules that would allow their activation, and there was a buildup of inhibitory immune system cells.

Green and members of his lab also looked at lung tissues, because one of the controversies in sepsis research is whether sepsis affects different organs in the same way. What they found was that receptors on the lungs appeared to contribute to the patients' immunosuppressed state.

'The lung can turn the T cells off when they get in there,' Hotchkiss explained. 'Maybe that limits damage to the lung, but it also further compromises immune function.'

The results suggested that sepsis patients who die late in the course of their illness are in some ways the victim of the progress that has enabled doctors to keep them alive longer. Death from sepsis used to be extremely rapid – and sometimes still is. But one of the patients whose tissues Hotchkiss' team analyzed had survived for 40 days before dying.

In order to better treat them, Hotchkiss said, 'what you need is some type of test, or a series of tests, to tell you whether the patient is in the hyper- or the hypo-immune phase of sepsis.'

For drug developers, the results may mean that, for a significant fraction of sepsis patients, drugs aimed at damping down the immune system may make things worse rather than better. An accompanying editorial suggested that treatments with interleukins might restore T cells to immunosuppressed sepsis patients and that, more generally, the study 'underscores the desperate need for a better understanding of sepsis and the urgent need for new therapeutic strategies.'

Hotchkiss himself said that one solution may be to look at whether cancer drugs that aim to stimulate the immune system might be useful for treating those sepsis patients who have progressed to an immunosuppressed state.

'Some of the drugs that look good in cancer also look good in animal models of sepsis,' Hotchkiss said. 'And we are anxious to move them into clinical trials.'