Double-Blind, Randomized Trial of an Anti-CD18 Antibody in Conjunction With Recombinant Tissue Plasminogen Activator for Acute Myocardial Infarction

Methods

The Limitation of Myocardial Injury following Thrombolysis in Acute Myocardial Infarction (LIMIT AMI) study was a randomized, double-blind, placebo-controlled, multicenter, 3-group study conducted in 60 centers in the United States and Canada between September 1998 and March 2000. It was approved by Institutional Review Boards at each institution. All subjects provided written, informed consent.

Results

A total of 394 subjects were randomized and treated with rhuMAb CD18 or placebo. An additional 22 subjects were withdrawn before study drug administration. Patient characteristics are shown in Table 1.

There were no differences between placebo and active rhuMAb CD18 groups in any efficacy end point (Figure 1 and Table 2). The incidence of TIMI grade 3 flow in the placebo group was unexpectedly high (66%). Neither the proportion of subjects with TIMI myocardial perfusion grade 2 or 3 or infarct size or ECG ST-segment resolution end points exhibited a treatment effect (Table 3). No treatment effect was observed in the following subgroups (data not shown): treatment ≤3 hours, men versus women, or use of a glycoprotein IIb/IIIa inhibitor in the first 2 days.

• Download figure
• Open in new tab
• Download powerpoint

Figure 1. Primary end point: corrected TIMI frame count 90 minutes after rtPA administration.

There was no excess of any specific adverse event in the rhuMAb CD18 groups. However, when adverse event terms were grouped as those consistent with a “likely bacterial infection” (defined as sepsis, nonaspiration pneumonia, urinary tract infection, pyuria, abscess, furunculosis, nonviral infection, and abnormal healing) a trend toward a consequence of treatment was noted (Table 4; P=0.33 for trend). One self-limited, severe, febrile reaction was reported 30 minutes after the administration of a 2.0 mg/kg dose of rhuMAb CD18. There was no effect of rhuMAb CD18 on “serious and life-threatening” bleeding events, although the incidence of overall bleeding events was slightly higher in the rhuMAb CD18 groups (Table 4). White cell count increased from a baseline of ≈11 000 cells/mL in all groups to a peak of ≈16 000 to 17 000 at 24 hours in both rhuMAb CD18 groups, which returned toward normal more slowly in the highest dose group (Figure 2). No treatment effects were seen with respect to hemoglobin, platelet count, white cell differential count, fibrinogen, d-dimer, liver function, creatinine, or ferritin. No antibody formation to the active binding sites of rhuMAb CD18 was observed.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2. Time course of mean peripheral white blood cell (WBC) counts by treatment.

The mortality in the study population as a whole was 4.8% at day 30, and 5.1% at day 90. There were no significant differences between groups in the incidence of recurrent AMI, heart failure, or a composite end point (death, recurrent AMI, and serious/severe heart failure) at day 30 (Table 4).

Discussion

Our results demonstrated no effects of rhuMAb CD18 on multiple cardiac end points. The failure of anti-CD18 therapy in human subjects, despite positive findings in most experimental animal studies,^1,2,4,16 was surprising. However, results in canine models were varied, with no effect on infarct size, despite an apparent increase in post-reperfusion blood flow in one study⁶ and efficacy in another with 2 of 3 anti-CD18 antibodies tested. In a third study, efficacy was not seen when myocardial ischemia was severe.¹⁷

One possible explanation for the failure of anti-CD18 therapy in humans is that the duration of ischemia is so long that endothelial cell barrier function fails. If so, blocking the Mac-1–intracellular adhesion molecule interaction would not prevent neutrophils from migrating into the subendothelial tissues and contributing to tissue damage. The median time of presentation (2.7 hours) was longer than the ≤90 minutes of experimental ischemia studied in animals. In rabbits with myocardial infarction, anti-CD18 therapy was effective after coronary occlusion lasting 30 but not 45 minutes.¹⁸ By contrast, in dogs with neutrophil depletion, myocardial necrosis was reduced with ischemia of either 90 or 240 minutes.⁵ Sharar et al³ found, using a rabbit ear preparation and 6 hours of occlusion, that administration of anti-CD18 therapy immediately on reperfusion was as effective as that initiated 1 hour but not 4 hours later. No effect was observed when the antibody was administered after 12 hours.

At least 2 in vitro studies have reported CD18-independent migration across human endothelial cell layers in response to selected stimuli.^19,20 The mechanism of CD18-independent migration is not understood. It may be organ-dependent, with blocking antibodies consistently showing complete inhibition of neutrophil migration into skin but not into lung or liver.²¹ Although other families of adhesion receptors are well known, it is the β2 family, all of which contain CD18 subunits, that is thought to enable firm neutrophil attachment as a prelude to migration.²² Leukocytes and/or acute inflammation may not be critical to the amount of myocardial damage after coronary occlusion in humans. This possibility is consistent with the failure of large doses of methylprednisolone to improve AMI outcome.²³

Medications with possible activity against leukocyte adhesion used in this study included unfractionated heparin and abciximab. Either could have influenced the effects of rhuMAb CD18.^24–26 Doses of unfractionated heparin low enough not to affect the activated partial thromboplastin time bind to Mac-1 and interfere with n-formyl-methionine-leucine-phenylalanine (fMLP)-induced leukocyte-endothelial adhesion in rats.²⁷ The time course of the leukocytosis after anti-CD18 therapy is consistent with the prevention of margination of leukocytes, and yet this anti-adhesion phenomenon has not been reported with either heparin or abciximab. Mac-1 (CD18/11b, occurring mainly on neutrophils and monocytes) is also a receptor for C3bi and fibrinogen. Interference with its functions is seen in vivo and in vitro at clinically relevant concentrations of heparin.²⁴ Thus, heparin may have interfered with anti-CD18 effects in this study.

Despite an unexpectedly high proportion of subjects with TIMI grade 3 flow at 90 minutes after rtPA (66%) in the placebo group, the study had adequate power to detect treatment effects. The power of this study to demonstrate the expected 15% absolute increase in TIMI grade 3 flow rate (from the observed rate of 66% to 81%) was ≈70%.

The protocol design was sufficiently sensitive to detect clinically relevant effects of anti-CD18 treatment in view of the multiple cardiac end points assessed. Coronary flow and 90-minute ECG end points were obtained before mechanical intervention to avoid confounding effects of the intervention itself. Blinded core reference laboratories were used to assess efficacy end points independently. The levels of missing data were acceptable, and the findings were robust in regard to analysis methods. The doses studied resulted in the expected plasma concentrations and were sufficient to fully saturate CD18 receptors within the first few hours. Biological activity was evident, as judged from the 2-fold rise in mean peripheral white cell counts in both active treatment groups. Thus, it seems that dosage was adequate.

Our results indicate that rhuMAb CD18 was well tolerated but lacked any beneficial cardiac effects in patients with ST-segment elevation AMI.

Appendix