Current issues of ACP Journal Club are published in Annals of Internal Medicine


Therapeutics

Human monoclonal antibody HA-1A did not reduce mortality from septic shock

ACP J Club. 1994 Nov-Dec;121:62. doi:10.7326/ACPJC-1994-121-3-062


Source Citation

McCloskey RV, Straube RC, Sanders C, Smith SM, Smith CR, the CHESS Trial Study Group. Treatment of septic shock with human monoclonal antibody HA-1A. A randomized, double-blind, placebo-controlled trial. Ann Intern Med. 1994 Jul 1;121:1-5.


Abstract

Objective

To determine the effectiveness of human monoclonal antibody HA-1A in patients with septic shock.

Design

14-day randomized, double-blind, placebo-controlled trial (Centocor: HA-1A Efficacy in Septic Shock [CHESS] trial).

Setting

513 community and university-affiliated hospitals.

Patients

Patients were in shock (systolic blood pressure < 90 mm Hg after adequate resuscitation) within 6 hours before enrollment and went into shock within 24 hours before enrollment, had a presumptive diagnosis of gram-negative infection judged to have caused the shock episode, and received full supportive care. Exclusion criteria included age < 18 years, pregnancy, life expectancy < 3 months, organ or bone marrow transplantation within the previous 6 months, total leukocyte count < 500 cells/mm3, > 10% body surface burns, and "do-not-resuscitate" orders. 2199 patients were randomly assigned. Follow-up was complete.

Intervention

Patients were randomly allocated to HA-1A, 100 mg (n = 1113), or placebo (n = 1086). Of 621 patients with gram-negative bacteremia, 328 received HA-1A and 293 received placebo.

Main Outcome Measures

14-day all-cause mortality and occurrence of spontaneously reported adverse events.

Main Results

At the first interim analysis, the trial was stopped because the all-cause mortality rate for patients without gram-negative bacteremia receiving HA-1A exceeded that of patients receiving placebo by an amount greater than that specified by the stopping rule. When the additional patients without gram-negative bacteremia who enrolled between the interim and final analyses were included, the mortality rate was 41% for the HA-1A group and 37% for the placebo group {95% CI for the 4% difference -1% to 8.5%}*. For patients with gram-negative bacteremia, the all-cause mortality rate was 33% for the HA-1A group and 32% for the placebo group {CI for the 1% difference -7% to 8%}*. The groups did not differ for spontaneously reported adverse events.

Conclusion

Compared with placebo, human monoclonal antibody HA-1A was not effective in reducing mortality in patients with septic shock.

Source of funding: Centocor, Inc.

For article reprint: Dr. R.V. McCloskey, Centocor Inc., 200 Great Valley Parkway, Malvern, PA 19355. FAX 610-651-6125.

*Numbers calculated from data in article.


Commentary

The CHESS study found that the monoclonal antibody HA-1A did not reduce mortality in patients with septic shock. In fact, there was a trend toward higher overall mortality with HA-1A compared with placebo. The study's large size and simple design support the validity of its findings.

Evidence from both the HA-1A pivotal trial (1) and the CHESS trial suggests that HA-1A may harm some patients with septic shock. The finding that spontaneously reported, acute adverse events (2) were similar in the HA-1A and placebo groups is only marginally reassuring. The high background rate of adverse events in septic shock makes detection of drug-related events difficult; also, untoward effects of HA-1A may occur late. Therefore, mortality may be the best measure of harm.

In the CHESS trial, excess mortality was found in patients without gram-negative sepsis who were treated with HA-1A (14-day mortality, 318 of 785 [41%] vs. 292 of 793 [37%]; P = 0.07), invoking an early stopping rule set up by the investigators to protect patients. Similar results were found in the HA-1A pivotal trial (1) (28-day mortality, 68 of 150 [45%] vs. 73 of 184 [40%]; P = 0.18). When data from these studies are combined, mortality on HA-1A compared with placebo is 42% vs. 37% (Fisher 1-tailed exact test; P = 0.04). HA-1A appears to increase mortality in a substantial number of patients with septic shock, and it should not be used for this syndrome.

Zenaide M.N. Quezado, MD
Robert L. Danner, MD National Institutes of Health Bethesda, Maryland, USA


References

1. Ziegler EJ, Fisher CJ Jr, Sprung CL, et al. Treatment of gram-negative bacteremia and septic shock with HA-1A human monoclonal antibody against endotoxin. A randomized, double-blind, placebo-controlled trial. N Engl J Med. 1991;324:429-36.

2. Cross AS. Antiendotoxin antibodies: a dead end? [editorial]. Ann Intern Med. 1994;121:58-60.


Author's Response

The conclusion of Quezado and Danner differs from ours, which was that HA-1A produced neither benefit nor excess harm in any study group.

Richard V. McCloskey, MD Centocor, Inc.
Malvern, Pennsylvania, USA
Elizabeth Ziegler, MD
University of California Medical Center San Diego,California, USA