WASHINGTON -- The FDA's Anti-Infective Drugs AdvisoryCommittee last week took on the difficult challenge of settingparameters for conducting clinical trials on drugs designed tomediate or reverse the cascade of immune responses that leadto sepsis.

Based on the meeting, FDA's Division of Anti-Infective DrugProducts will upgrade information in the Points to Considerdocument that provides guidance for companies conductingclinical trials, and then present it to the advisory committee forfine tuning.

Early on, the committee tackled the issue of definitions. RogerBone, vice president of medical affairs and dean of RushMedical College at Rush-Presbyterian-St. Luke Medical Centerin Chicago, suggested his appellation systemic inflammatoryresponse syndrome (SIRS) and said, "Let us make it so sensitivethat it includes anyone a doctor says has sepsis."

Bone defined SIRS as a combination of any two of theseindicators: a temperature change, a heart-rate change, a changein the number of white blood cells, and tachypnea (excessivelyrapid breathing).

Some of the most important points of discussion concerned endpoints for studies. "We need to look not only at mortality ratebut at reversal of multiorgan dysfunction," said Bone.

The issue of other end points has been controversial because adrug can improve patients' organ function without saving theirlives.

William Knaus, director of intensive care unit (ICU) research atGeorge Washington University Medical Center, suggestedsurvival time as a middle ground. "You can increase survivaltime of patients with these conditions," he said. "That is yourstatistical test. ... At the same time, there has to be a clinicallyimportant difference in 28-day mortality."

Jay Seigel of the FDA's Center for Drug Evaluation and Researchexplained that adjuvants for sepsis can sometimes harm thepatient. Another complicating factor, said Seigel, is that "certainclasses of drugs may work better in certain populations."

On the other hand, Bone explained, many patients with sepsisdie from the underlying condition, such as trauma or headinjury, rather than from sepsis. Therefore, even a true magicbullet -- a purely hypothetical concept -- might thereforereduce mortality by only 20 percent. "We don't want to get onthe slippery slope of not approving drugs that do work becausewe have unreasonable expectations," said Bone.

"Because of our inability to define the patient population," hesaid, "we are potentially rejecting an agent ... that may beeffective."

But Apache Medical Systems of Washington, D.C., which wasfounded by Knaus, has developed an algorithm that can beused to define a patient population prospectively. Thealgorithm uses 17 physiological parameters to predictindividual risk of death. These include blood pH, creatinine,age, white blood cell count and chronic conditions the patientmay have; these data can be obtained from a patient inintensive care within minutes to several hours (instead of indays, as with other methods).

In tests, the algorithm has quite accurately predictedindividual risk of mortality, said Knaus. He is using a data baseof recently treated ICU patients as a placebo group. "We cangain understanding of what is important in sepsis before welook at clinical trials," he said.

It can also help researchers and clinicians decide who willbenefit the most from a drug. In Synergen Inc.'s recent PhaseIII trial, patients with less than a 24 percent risk of mortalitywere not helped by the company's IL-1 receptor antagonist,Knaus said.

This was true when all 11 predefined subsets of sepsis werebroken out during tests of the algorithm, including differentcausal bacteria and failure of different organs.

"Further investigation of the relationship between baseline riskand efficacy could lead to insights regarding the mechanism ofaction and quantitative estimates of benefit to individualpatients from new therapeutic approaches to sepsis," saidKnaus.

Synergen announced last March that it will be using theAPACHE algorithm to conduct a second phase III trial on its IL-1 receptor antagonist, said Robert Thompson, vice presidentand director of research and clinical affairs for Synergen.

-- David C. Holzman Washington Editor

(c) 1997 American Health Consultants. All rights reserved.