OUP user menu

The effect of ITF-1697 on reperfusion in patients undergoing primary angioplasty

Maurits T Dirksen, GertJan Laarman, Arnoud W.J van ‘t Hof, Giulio Guagliumi, Wilhelmus A.L Tonino, Luigi Tavazzi, Dirk J.G.M Duncker, Maarten L Simoons
DOI: http://dx.doi.org/10.1016/j.ehj.2003.12.018 392-400 First published online: 1 March 2004


Aim ITF-1697 is a C-reactive protein-derived tetrapeptide that, based on pre-clinical studies, is thought to reduce reperfusion injury. We performed a dose-finding study to assess safety, preliminary efficacy and clinical outcome of prolonged i.v. infusion of ITF-1697 in patients with an acute myocardial infarction (AMI) who were eligible for percutaneous coronary intervention (PCI).

Methods and results This was a multicentre dose-finding study that was randomised, double blind, and placebo-controlled. Four hundred and two patients were enrolled. Intravenous infusion of four dosages of ITF-1697 (0.1, 0.5, 1.0 or 2.0 μg/kg/min) or placebo was started before PCI and continued for 24 h. After interim analysis of data from 242 patients the study continued with the 0.1 and 1.0 μg/kg/min ITF-1697 regimes. Analysis did not raise any safety concerns. Post-procedure perfusion, assessed by TIMI flow, corrected TIMI frame count, blushgrade and ST-segment resolution, was similar for the placebo, 0.1 and 1.0 μg/kg/min regimes. Furthermore, the results showed no differences between the treatment regimes in enzymatic infarct size or clinical outcome up to 30 days.

Conclusion ITF-1697 was well tolerated. However, neither a dose-relation nor improvement of perfusion, clinical outcome or reduction of myocardial damage could be demonstrated with ITF-1697 during and after primary PCI for AMI.

  • Reperfusion injury
  • Acute myocardial infarction
  • Primary percutaneous intervention
  • Inflammation
  • Drug therapy
  • Coronary microcirculation
  • Randomisation


Acute myocardial ischaemia caused by thrombotic coronary occlusion leads to time-dependent loss of myocytes. Percutaneous coronary intervention (PCI) and thrombolytic therapy are effective and commonly used therapies to obtain early reperfusion in acute myocardial infarction (AMI).1–3 Early reperfusion conveys tissue salvage, i.e., reduction in infarct size and preservation of left ventricular function, thereby improving short- and long-term outcome after AMI.4 However, it has been suggested that reperfusion itself can precipitate further myocardial injury, known as ‘reperfusion injury’.5–7 If this ‘reperfusion injury’ could be treated and eliminated, the outcome for patients with AMI might improve further.8 Reperfusion injury is thought to be mainly mediated by an inflammatory response (polymorphonuclear neutrophil, complement and platelet activation), intracellular Ca+ overload and reactive oxygen species. Several pharmacological agents or adjunctive therapies have been used in the clinical setting to attempt to reduce the injury caused by these pathways.9–13

This report presents the first study investigating the effects of ITF-1697 on reperfusion injury in humans. ITF-1697 is a proprietary tetrapeptide Gly–(Et)Lys–Pro–Arg that corresponds to sequence 113–116 of C-Reactive Protein (CRP) and has been shown to exert anti-inflammatory actions and marked anti-anaphylactive properties.14 ITF-1697 is a new substrate that showed promising results in the pre-clinical phase to prevent reperfusion injury in AMI, mainly by inhibiting the inflammatory response. Experimental data, obtained in models of ischaemia and reperfusion, have shown that ITF-1697 prevents PMN adhesion and extravasation, preserves vascular endothelial phenotype and limits the increase in vascular permeability and loss of microcirculatory blood flow (unpublished data on file at Italfarmaco). For example, in the hamster cheek pouch model of microcirculation subjected to 30 min of ischaemia followed by reperfusion, ITF-1697 (1.0 μg/kg/min) virtually abolished microvascular damage (capillary leak, fluid extravasation, PMN adhesion and capillary plugging) and resulted in preservation of microcirculatory flow. Also, in a study in the in vivo dog heart, 0.83 μg/kg/min ITF-1697 (administered i.v. just prior to reperfusion after 90 min of coronary artery occlusion) reduced myeloperoxidase activity (reflecting reduced neutrophil activity) and decreased myocardial infarct size by 50% resulting in preservation of ejection fraction (unpublished data on file at Italfarmaco). Other in vivo studies in rat, rabbit, and dog showed improved survival with ITF-1697 with doses as low as 0.05 μg/kg/min, administered either before coronary artery ligation or just prior to reperfusion after coronary occlusion resulting in myocardial infarction.

Accordingly, it was hypothesised that ITF-1697 may provide benefit on top of optimal drug and interventional therapy in patients with acute ST elevation myocardial infarction. The present study was designed to assess safety, and to compare preliminary efficacy and clinical outcome of prolonged intravenous infusion of 4 doses of ITF-1697 and placebo during and after a primary PCI procedure in patients with acute myocardial infarction.


The Protect Against Reperfusion Injury with ITF-1697 in acute Myocardial Infarction (PARI-MI) was a multicentre, randomised, double blind, placebo-controlled, dose-finding study conducted in 11 centres in Italy and The Netherlands between January and December 2001. The study was started comparing 4 doses of ITF-1697 (0.1, 0.5, 1.0, 2.0 μg/kg/min) and placebo with about 50 patients in each group. After interim analysis the study was designed to continue with another 50 patients in each of 2 doses of ITF-1697 and placebo. Thus, finally 3 groups of 100 patients each (2 doses of ITF-1697 and placebo) and 2 groups of 50 patients each (ITF-1697) were to be enrolled.

Study population

Patients with typical symptoms of AMI and electrocardiographic (ECG) signs of a large or medium size infarction, who were eligible to undergo invasive revascularisation through primary PCI, were included in the trial. Written informed consent was obtained from all patients and the study design was approved by the institutional ethics committees.

Patients of both sexes had to be at least 18 years of age and time from symptom onset to enrolment had to be within 12 h. Chest pain had to last for at least 20 min and thrombolytic treatment prior to PCI was not allowed. In order to enrol patients with large and medium size AMIs ST segment elevation had to be ⩾0.2 mV in two or more contiguous leads with the cumulative ST deviation ⩾10 mV on a 12-lead ECG. Patients were excluded having a history of bleeding diathesis; major bleeding ⩽30 days; major surgery ⩽30 days; history of known hemorrhagic stroke at any time or any stroke ⩽30 days; uncontrolled hypertension (Math200/110 mmHg); participation in another trial; inability to perform 30 day follow up; pregnancy or women of childbearing potential; concomitant disease that interferes with prognosis; contra-indications to standard drugs for coronary intervention and coronary heart disease; weight Math105 kg. Study medication was started in eligible patients after informed consent without further delay. When subsequent platelet count Math60,000/mm3 (<Math109/L), hematocrit Math30% or creatinine ⩾2.9 mg/dL (⩾177 μmol/L) indicated that the patient should have been excluded, the infusion of study medication was stopped. Nevertheless, the patient remained in the study and all measurements were performed according to the intention to treat principle.


Patients were randomised by an Interactive Voice Response System to an 1–2 min i.v. loading dose followed by a continuous i.v. infusion of ITF-1697 in different doses (0.1, 0.5, 1.0, 2.0 μg/kg/min) or placebo. The intravenous loading dose was 11, 55, 110, and 220 μg/kg in the 0.1, 0.5, 1.0 and 2.0 μg/kg/min ITF-1697 treatment groups, respectively. The rationale for the dose regimen employed was based on (unpublished) animal data. In an in vivo dog heart study, ITF-1697 was injected intravenously starting with a slow bolus of 100 μg/kg (starting 10 min prior to reperfusion) followed by 0.83 μg/kg/min throughout the 4.5 h reperfusion protocol and markedly reduced myocardial infarct size (unpublished data on file at Italfarmaco). Study drug was started after enrolment prior to angiography and continued up to 24 h. PCI of the infarct related coronary artery was performed at the investigator’s discretion. Aspirin was given daily starting on hospital admission with a loading dose of 300–500 mg. In the case of stenting, clopidogrel was given at a loading dose of 300 mg followed by 75 mg/day for four weeks. Concomitant treatment with (low molecular weight) heparin, glyco-protein IIb/IIIa (GPIIb/IIIa) inhibitors and all other therapies was at the investigator’s discretion.


The safety endpoints were death, recurrent AMI, bleeding and other Serious Adverse Events (SAE) from hospital admission until 30 days. This was assessed by laboratory blood samples, routine ECG and an outpatient clinic visit at 30 days (23–37). All deaths were included (cardiac and non-cardiac) and classified whenever possible on the basis of an autopsy. Bleeding, including intracranial haemorrhage, was classified according to the TIMI criteria.15 SAEs were defined as any medical occurrence that resulted in death, life-threatening situations, hospitalisation or prolongation of the existing hospitalisation, or persisting significant disability.


The efficacy endpoints were myocardial perfusion and myocardial infarct size. Perfusion was assessed by TIMI flow grade,16 corrected TIMI frame count (CTFC, corrected for 30 frames/second (f/s): LAD),17 myocardial blush grade,18 ST-segment resolution and recurrent ischaemia or recurrent AMI. With regard to the CTFC the distribution plot of TIMI frame counts was drawn for placebo and ITF 1697-treated patients. Values below 28 f/s were taken into account for defining normal reperfusion. Leftward shift of the mean frame count of at least 6 f/s was considered clinically significant. In case of an occluded vessel the maximum frame count measurement was scored. Two independent observers performed ST-segment analysis. The sum of ST segment elevation and depression was measured with lens-intensified callipers to the nearest 0.025 mV, 60 ms after the end of the QRS complex (J-point) of all 12 leads. Measurements were done at 1 and 3 h after intervention.

Myocardial infarct size

Myocardial damage was estimated by the area under the curve of serial α-hydroxybutyrate dehydrogenase (HBDH).19–21 Haemolysed samples were discarded and all other measurements were performed centrally by independent, blinded, and qualified personnel in the core laboratory. The angiographic measurements, the ST-segment measurements, as well as infarct size analysis and calculation were all performed centrally by independent and qualified personnel in the core laboratory, blinded for treatment assignment.

Clinical outcome

Re-infarction was defined as a second rise in CK-MB or total CK ⩾2 times the upper limit of normal and the onset of recurrent angina or new significant Q-waves of ⩾0.04 s in duration or having a depthMathone fourth of the corresponding R-wave amplitude in two or more contiguous leads.

Recurrent ischaemia was defined as (1) an episode of chest pain at rest, associated with new ST segment shift of ⩾0.1 mV or T-wave inversion/pseudo-normalisation in at least two contiguous leads, (2) chest pain resulting in an invasive cardiac intervention (including diagnostic catheterisation, intra-aortic balloon pump counterpulsation, PCI or Coronary Artery Bypass Graft (CABG)) within the same hospitalisation, or (3) readmission for unstable angina.

Statistical analysis

Two interim analyses were performed to evaluate safety and dose–response relationships in order to select two dosages for continuation and final analysis. With respect to preliminary efficacy assessment the following information regarding clinically significant changes in the relevant parameters applied, based on the control group data versus the ITF-1697 treatment group: a TIMI flow grade 3 of 85 versus Math90%2, myocardial blush grade 3 of 25 versus Math30%18 and enzymatic infarct size of 4.4 versus Math3.9 g eq.4 In addition, if any dose relationship in the incidence of adverse events seemed to be present, logistic regression analysis was used for trend analysis.

The first interim analysis was performed after 125 patients and the second interim analysis after about 250 patients had received ITF-1697 infusion or placebo. The following criteria were applied: (a) If dose–response was detected, two safe and effective dosages were to be selected. (b) If a dose–response was not evident, the number of 50 patients per group were allowed to have 90% statistical power to select the best dosage with respect to the aforementioned parameters of TIMI flow grade, blush grade and infarct size.22 The final sample size of 100 patients for the 2 selected dosages and placebo allowed an estimate of two-sided 90% C.I. equal to ±10% for differences between the response rate of active doses and placebo.

Data were presented as means±SD and for non-parametric measurements as median with percentiles. All analyses were performed comparing each active ITF-1697 dosage group separately with placebo. For discrete variables a Math test was used. Binomial confidence intervals were constructed and a Fisher’s Exact Test was performed to compare the incidence of adverse events of ITF-1697 with placebo. No adjustments were made for multiple testing. A Student’s t test was performed on normally distributed continuous data or a Wilcoxon rank test in case of skewed data distribution. Significance was defined as Math.

Subgroup analyses. Additional subgroups analysis was performed for gender; blush grade before procedure; ST-segment resolution; age (above or below median, Math or ⩾70 years); use of GPIIb/IIIa antagonist; time from onset of symptoms and intervention (⩾or Math2 h); infarct location; LAD or RCA as culprit lesion; infarct size (above or below median); and an event free history prior to the AMI.


Patient groups

A total of 402 patients were enrolled. Ten patients were excluded from the analysis: seven patients were randomised but did not receive study medication for various reasons, one other patient died shortly after randomisation before study medication was given, one patient refused written informed consent although he had given oral consent in the presence of a witness, and one patient withdrew the informed consent before discharge. Finally, 392 patients were enrolled and received study medication. After the first interim analysis the study was continued without modification. After the second interim analysis, with 242 patients, the study was continued with the placebo, 0.1 and 1.0 μg/kg/min dose regimes. The 1.0 μg/kg/min ITF-1697 regime was chosen as it was closest to the 10% absolute difference in TIMI flow grade 3 indicated by the protocol. The 0.1 μg/kg/min ITF-1697 was chosen to maintain reasonable pharmacological interval. Ninety two patients received placebo, 93 were treated with ITF-1697 0.1 μg/kg/min and 94 with 1.0 μg/kg/min. The treatment groups ITF-1697 0.5 and 2.0 μg/kg/min contained 55 and 58 patients, respectively. Baseline characteristics were similar between the groups (Table 1). In total 18 patients did not undergo a PCI because of various reasons, and did not continue on study medication after the initial loading dose and infusion, hence were expelled from further analysis. Three of these patients underwent emergency coronary bypass surgery. Pre-procedurally there were no significant differences apparent, except for TIMI flow: the ITF-1697 1.0 treatment group contained a higher percentage of patients with an occluded vessel (TIMI flow grade Math) at baseline. Because of this unbalanced distribution it was decided post hoc to perform a subgroup analysis in 208 patients with occluded vessels treated with placebo, 0.1 or 1.0 μg/kg/min ITF-1697.

View this table:
Table 1

PARI-MI: baseline demographics

Placebo, Embedded Image (%)0.1 ITF-1697, Embedded Image (%)1.0 ITF-1697, Embedded Image (%)
Mean age61±1362±1160±12
Male76 (83)70 (75)79 (84)
Non-smokers29 (32)33 (36)29 (31)
Diabetes (N)IDD11 (12)9 (10)17 (18)
Previous MI17 (19)11 (12)14 (15)
Previous angina20 (22)15 (16)14 (15)
Previous PCI8 (9)6 (6)11 (12)
Previous CABG2 (2)1 (1)2 (2)
Previous stroke
Non-hemorrhagic1 (1)3 (3)3 (3)
TIA2 (2)4 (4)3 (3)
Heart failure (Killip classification)
Class I82 (89)92 (98)84 (89)
Class II–IV10 (11)2 (2)10 (11)
  • No significant differences between the groups.

    (N)IDDM, (non) insulin dependent diabetes mellitus; MI, myocardial infarction; PCI, percutaneous coronary intervention; CABG, coronary artery bypass grafting; TIA, transient ischaemic attack.

Angiographic results

A successful PCI procedure, indicated by a post-procedural epicardial TIMI flow grade 3, was achieved in the placebo, ITF-1697 0.1 and ITF-1697 1.0 group in 82%, 84% and 88%, respectively (Table 2). Coronary stenting was performed in 84 (92%) of the placebo cases compared with 80 (85%) and 83 (89%) in the ITF-1697 0.1 and 1.0 group, respectively. GPIIb/IIIa antagonists were used in 122 patients, 44%. The median CTFC in the total population was 24 f/s. The results of the angiographic measurements (TIMI flow, CTFC and blush grade) used as markers of reperfusion efficacy did not show any significant difference between the treatment groups.

View this table:
Table 2

PARI-MI: angiographic and angioplasty results

TIMI flow grade pre-procedurePlacebo, Embedded Image (%)0.1 ITF-1697, Embedded Image (%)1.0 ITF-1697, Embedded Image (%)
TIMI flow 0–163 (69)69 (74)76 (83)*
TIMI flow 213 (14)14 (15)13 (14)
TIMI flow 315 (17)10 (11)3 (3)
p Value0.510.013*
TIMI flow grade post-procedurePlacebo, Embedded Image (%)0.1 ITF-1697, Embedded Image (%)1.0 ITF-1697, Embedded Image (%)
TIMI flow 0–14 (4)1 (1)3 (3)
TIMI flow 212 (13)13 (14)8 (9)
TIMI flow 374 (82)76 (84)77 (88)
p Value0.630.34
CTFC pre-procedurePlacebo, Embedded Image0.1 ITF-1697, Embedded Image1.0 ITF-1697, Embedded Image
Median f/s (P25-P75)173 (173–173)173 (173–173)173 (173–173)
p Value0.920.16
CTFC post-procedurePlacebo, Embedded Image0.1 ITF-1697, Embedded Image1.0 ITF-1697, Embedded Image
Median f/s (P25-P75)24 (17–36)26 (17–41)24 (14–41)
p Value0.520.99
Blushgrade post-procedurePlacebo, Embedded Image (%)0.1 ITF-1697, Embedded Image (%)1.0 ITF-1697, Embedded Image (%)
Blush grade 0–132 (41)38 (49)43 (53)
Blush grade 220 (26)27 (35)23 (28)
Blush grade 325 (33)13 (17)15 (19)
p Value0.160.13
Subgroup TIMI flow grade Embedded Image pre-intervention
TIMI flow grade post-procedurePlacebo, Embedded Image (%)0.1 ITF-1697, Embedded Image (%)1.0 ITF-1697, Embedded Image (%)
TIMI flow 0–14 (6)1 (2)3 (4)
TIMI flow 212 (19)11 (17)7 (10)
TIMI flow 346 (74)54 (82)63 (86)
p Value0.250.09
CTFC post-procedurePlacebo, Embedded Image0.1 ITF-1697, Embedded Image1.0 ITF-1697, Embedded Image
Median f/s (P25-P75)24 (19–46)26 (17–43)24 (14–42)
p Value0.970.62
Blushgrade post-procedurePlacebo, Embedded Image (%)0.1 ITF-1697, Embedded Image (%)1.0 ITF-1697, Embedded Image (%)
Blush grade 0–120 (38)29 (43)36 (55)
Blush grade 214 (26)18 (32)18 (27)
Blush grade 319 (36)9 (16)12 (18)
p Value0.060.05
  • p Value: ITF-1697 dosage compared with placebo.

    TIMI, thrombolysis in myocardial infarction; CTFC, corrected TIMI frame count.

  • * Statistically significant.

Electrocardiographic results

ECGs at enrolment as well as at 1 versus 3 h post-PCI were available in 372 patients (of the total of 392). There was no effect of either one of the ITF-1697 treatment regimes on the extent of ST-segment resolution (Table 3).

View this table:
Table 3

PARI-MI: ECG resolution of ST-segment elevations at 1 and 3 h after intervention

1 h ST-resolutionPlacebo, Embedded Image (%)0.1 ITF-1697, Embedded Image (%)1.0 ITF-1697, Embedded Image (%)
Complete res (⩾70%)22 (31)20 (29)20 (24)
Partial res (30–70%)22 (31)30 (44)36 (43)
No res (Embedded Image30%)28 (39)19 (28)27 (33)
p Value0.450.96
3 h ST-resolutionPlacebo, Embedded Image (%)0.1 ITF-1697, Embedded Image (%)1.0 ITF-1697, Embedded Image (%)
Complete res (⩾70%)29 (37)30 (39)23 (29)
Partial res (30–70%)32 (41)36 (46)44 (55)
No res (Embedded Image30%)17 (22)12 (15)13 (16)
p Value0.560.71
Subgroup TIMI flow grade Embedded Image pre-intervention
1 h ST-resolutionPlacebo, Embedded Image (%)0.1 ITF-1697, Embedded Image (%)1.0 ITF-1697, Embedded Image (%)
Complete res (⩾70%)11 (23)15 (30)15 (23)
Partial res (30–70%)16 (33)20 (40)31 (47)
No res (Embedded Image30%)21 (44)15 (30)20 (30)
p Value0.190.32
3 h ST-resolutionPlacebo, Embedded Image (%)0.1 ITF-1697, Embedded Image (%)1.0 ITF-1697, Embedded Image (%)
Complete res (⩾70%)14 (26)21 (38)19 (30)
Partial res (30–70%)27 (51)27 (48)35 (55)
No res (Embedded Image30%)12 (23)8 (14)10 (16)
p Value0.150.43
  • p Value: ITF-1697 dosage compared with placebo.

    res, resolution; TIMI, thrombolysis in myocardial infarction.

Infarct size

The median enzymatic infarct size, as measured by the area under the HBDH curve in the total group, was 5.2 g-eq. The infarct size for placebo, ITF-1697 0.1 and ITF-1697 1.0 and in the subgroup TIMI Math pre-intervention as shown in Table 4 was not significantly different. Larger infarct size was found in the total population in the TIMI Math pre-intervention subgroup compared to TIMI 2, 3 pre-intervention.

View this table:
Table 4

PARI-MI: enzymatic infarct size

Median enzymatic infarct sizePlacebo, Embedded Image0.1 ITF-1697, Embedded Image1.0 ITF-1697, Embedded Image
g-eq/L (P25-P75)4.5 (3.2–9.1)5.0 (3.2–7.3)5.3 (3.6–9.8)
p Value0.840.36
Subgroup TIMI flow grade Embedded Image pre-intervention
Median enzymatic infarct sizePlacebo, Embedded Image0.1 ITF-1697, Embedded Image1.0 ITF-1697, Embedded Image
g-eq/L (P25–P75)5.8 (3.7–9.1)5.6 (3.7–9.4)6.0 (3.9–11.2)
p Value0.980.24
  • p Value: ITF-1697 dosage compared with placebo.

    g eq, gram equivalents; TIMI, thrombolysis in myocardial infarction.

Clinical outcome

Clinical endpoints did not raise any safety concerns in any group. Ten patients died during the initial hospitalisation. At 30 days there were 16 deaths reported (Table 5). The occurrence of adverse events such as recurrent MI, CABG, or bleeding did not show significant differences between the regimes. After 30 days 10 (3.6%) patients had suffered from recurrent ischaemia. A combination of the endpoints of mortality, recurrent ischaemia or recurrent AMI did not reveal a treatment effect of ITF-1697 (Table 5). Fifteen of the total of 16 reported deaths had TIMI flow grade Math pre-intervention. Recurrent myocardial infarction or recurrent ischaemia was not evidently higher in this subpopulation compared with those having TIMI 2, 3 flow pre-intervention.

View this table:
Table 5

PARI-MI: endpoints

Placebo, Embedded Image (%)0.1 ITF-1697, Embedded Image (%) [p value]0.5 ITF-1697, Embedded Image (%) [p value]1.0 ITF-1697, Embedded Image (%) [p value]2.0 ITF-1697, Embedded Image (%) [p value]ITF total, Embedded Image (%) [p value]
Events at 30 days
Mortality3 (3.3)3 (3.2)4 (7.3)2 (2.1)4 (6.9)13 (4.3)
Recurrent MI001 (1.8)2 (2.1)2 (3.4)5 (1.7)
CABG2 (2.2)3 (3.2)1 (1.8)2 (2.1)3 (5.2)9 (3.0)
Bleeding13 (14.1)7 (7.5)7 (12.7)14 (14.9)3 (5.2)31 (10.3)
Mortality or MI or re-Ischemia6 (6.5)4 (4.3)5 (9.1)10 (10.6)7 (12.1)26 (8.7)
Subgroup TIMI Embedded Image pre-intervention
Placebo, Embedded Image (%) [p value]0.1 ITF-1697, Embedded Image (%) [p value]0.5 ITF-1697, Embedded Image (%) [p value]1.0 ITF-1697, Embedded Image (%) [p value]2.0 ITF-1697, Embedded Image (%) [p value]ITF total, Embedded Image (%) [p value]
Events at 30 days
Mortality3 (4.8)2 (2.9)4 (8.9)2 (2.6)4 (10.0)12 (5.2)
Recurrent MI001 (2.2)2 (2.6)2 (5.0)5 (2.2)
Recurrent-Ischemia1 (1.6)1 (1.5)05 (6.6)2 (5.0)8 (3.5)
  • p Value: ITF-1697 dosage compared with placebo.

    MI, myocardial infarction; CABG, coronary artery bypass grafting; TIMI, thrombolysis in myocardial infarction.

Subgroup analysis

Analysis of the before mentioned subgroups on safety and efficacy between either one of the treatment groups compared with placebo did not reveal a dose–response relation. Also, the angiographic outcome in the TIMI flow grade Math pre-intervention subgroup showed no significant differences, albeit a trend was observed for post-procedural TIMI flow in favour of both ITF-1697 treatment groups. The CTFC and blush grade did not differ significantly in this subgroup. In fact, the latter showed a possible adverse trend (Table 2).


The principal finding of the present study was that prolonged i.v. administration of ITF-1697 as an adjunct to primary angioplasty was well tolerated and without safety concerns. In fact, the mortality after a large AMI in the total study population was low (4.1% at 1 month), which compares favourably with the overall 8.4% one-month mortality after myocardial infarction in the European Heart Survey.23 Overall, ITF-1697 did not appear to exert a significant effect on indices of reperfusion (TIMI flow grade, CTFC, or blush grade), ST-segment resolution, myocardial infarct size, or clinical outcome, and no dose–response was apparent. However, a trend towards better TIMI flow post-procedure was observed in the ITF-1697 1.0 μg/kg/min dose group. Furthermore, there appeared to be an imbalance in the number of patients with occluded infarct related vessels at angiography (TIMI flow Math). These patients showed a larger infarct size and a higher mortality rate. In a separate analysis of patients with an initially occluded vessel, a possible dose–effect relation was observed for post-procedural TIMI flow in favour of both ITF-1697 treatment groups. However, no such trends were apparent for reperfusion, infarct size or clinical outcome, while blush grade even showed a possible adverse trend (Table 2). Future studies might investigate whether the latter observations in patients with an initially occluded vessel are true, or a chance finding.

Reperfusion injury: purported mechanisms

Reperfusion is the only intervention that has been proven to halt the process of myocardial infarction after coronary artery occlusion. However, benefits of reperfusion may be limited by untoward effects, termed “reperfusion injury”.5–8 The mechanism of reperfusion injury is not completely understood, but may include (i) cytosolic and mitochondrial Ca+-overload, (ii) massive release of reactive oxygen species, (iii) damage to the extracellular matrix, (iv) impaired cellular energetics, and (v) an acute inflammatory response.5,8,24,25 The latter involves recruitment of polymorphonuclear neutrophils (PMN), complement and platelet activation, and disturbed endothelial function, which leads to increased permeability and results in more necrosis and myocardial damage. Recruitment and activation of PMN results in capillary plugging (thereby mechanically blocking flow), capillary leakage and enhancement of generation of reactive oxygen species, together culminating in secondary impairment of microcirculatory flow (“no-reflow” phenomenon) despite initial restoration of epicardial coronary patency and blood flow.24–27 The occurrence of no-reflow correlates with increased final infarct size and a poorer long-term clinical outcome.28,29 Consequently, targeting one or more of the above-mentioned mechanisms could further improve the benefit of early reperfusion.

Apparent lack of effect of ITF-1697 in patients with myocardial infarction

Results from different animal models of reperfusion injury indicated that with ITF-1697 the infarct size, occurrence of ventricular tachycardias and fibrillation as well as mortality was reduced. These beneficial effects were observed in models in which the substrate was administered before and during ischaemia as well as when administered at the time of reperfusion (unpublished data on file at Italfarmaco). In contrast to these findings the present study showed neither consistent benefit nor a dose–effect relation of ITF-1697 on indices of perfusion after direct PCI. To detect a dose–response in this study population an extensive prospectively prescribed statistical analysis was developed, unfortunately this analysis could not demonstrate a dose–response. Several factors may contribute to the discrepancy between animal experiments and the present study in humans.

Collateral blood flow. Collateral blood flow is an important determinant of infarct size.30 In the AMISTAD trial the authors speculated that the benefit of adenosine (as an adjunct to thrombolysis), which occurred selectively in patients with an anterior AMI, could be attributed to collateral flow.10 In contrast, the present study failed to reveal any evidence of a differential outcome in patients with culprit lesions in either RCA or LAD.

Timing of administration. It could be argued that ITF-1697, administered prior to, during and after PCI, did not reach the jeopardised myocardium in sufficient concentrations prior to the onset of reperfusion, except perhaps via microvascular collateral vessels that are undetectable by coronary angiography.31,32 Yet, it should be appreciated that ITF-1697 was infused intravenously several minutes before PCI was performed. This should give ample time to achieve adequate plasma levels of ITF-1697, so that at the time of onset of reperfusion most neutrophils should have been exposed to ITF-1697. The apparent lack of benefit in patients with an open IRV at angiography may be related to the timing of drug administration, since in those patients spontaneous reperfusion had already occurred, implying that ITF-1697 was administered at a considerable time after (partial) reperfusion.

Duration of ischaemia. In a previous study in dogs, ITF-1697 limited myocardial infarct size when administered just prior to reperfusion after up to 90 min coronary artery occlusion. In contrast, in the present study coronary artery occlusions lasted more than 2 h. It is possible that ischaemia-related damage predominates in patients with such prolonged coronary occlusion (which is typical for clinical studies), with few cardiomyocytes left to be affected by reperfusion injury.31 Consequently, little effect might then be expected of reperfusion injury limiting strategies. This may, at least in part, explain why over the past few years several pharmacological agents that have shown considerable promise in the experimental setting have failed to show any benefit in the clinical setting of ischaemia and reperfusion.9,11–13 However, even when we analysed the subgroup of patients that had re-established flow within 2 h of symptom onset, no benefit was observed.

Presence of pre-infarct angina. Another important difference between the clinical and experimental setting is that the myocytes used in animal studies are, in contrast to human coronary arteries, generally not subjected to the process of atherosclerosis, endothelial dysfunction and/or myocardial ischaemia prior to the sustained coronary artery occlusion.27,31 It is thus possible that patients, particularly those that experience angina prior to infarction, are protected via ‘ischaemic-preconditioning’,33,34 so that less additional benefit of ITF-1697 might be expected. However, also in patients with an event free history prior to the AMI no benefit of ITF-1697 was observed in the present study.

Micro-embolisation. Another consequence of the process of atherosclerosis in the human heart versus coronary artery ligation in experimental animals,26,31 is that impaired post-procedural perfusion or the no-reflow phenomenon could be, at least in part, the result of embolisation of plaque debris into the distal microvasculature rather than of reperfusion injury.35–38

Co-medication. It may be hypothesized that treatment with heparin and GPIIb/IIIa blockers may influence the effect of ITF-1697 on adherence and extravasation of PMNs and possibly on endothelial permeability.39–41 Subgroup analysis did not support such a hypothesis, as we did not observe any effect of ITF-1697 in patients who did or did not receive heparin and GPIIb/IIIa blockers.

Doses of ITF-1697. The highest dose used in the current trial (2.0 μg/kg/min) was 2.4-fold higher than the dose of the limited infarct up to size 50% in the dog. Nevertheless, it is possible that the dose regimes that were employed in this study may not have been optimal in humans.

Sample size. The sample size used was large enough for safety assessment and dose–effect detection, but not large enough to conclude definitely whether ITF-1697 administration, as an adjunct to primary PCI, can modulate reperfusion. We do realise that an effect of ITF-1697 could have been missed in this relatively small study. Furthermore, it should be appreciated that this study was rather complex, concerning many different measurements in patients admitted with myocardial infarction. Because of this complexity measurements were incomplete in part of the patients. Yet, the primary angiographic endpoints were complete in 84% (blushgrade) to 96% (TIMI flow grade) of the patients.

Reperfusion injury in clinical practice

The concept of reperfusion injury has been clearly demonstrated in many experimental studies, although even in those studies data are not fully consistent.8,24,25,31,42 In contrast, in clinical myocardial infarction, the importance of reperfusion injury has not been established. In fact, clinical studies with anti-inflammatory drugs,11,13 Na+/H+ exchange inhibitors,9 glucose–insulin–potassium,12,43,44 and adenosine10,45 so far led to inconclusive results. With this in mind, one might raise the question that possibly treatment in clinical practice is too late to reduce reperfusion injury or that there is no (relevant) reperfusion injury in humans, and thus no clinical benefit of decreasing such injury. Similarly, the PARI-MI study did not reveal a clear benefit, and no dose-relation, even though higher dosages were studied than in the pre-clinical studies, in which ITF-1697 was shown to be beneficial. The main reason for discrepancy between animal and human findings would be the different pathophysiologic contexts in which the drug was tested, leaving open the hypothesis that the drug might have beneficial effects in different pathophysiological settings, more appropriate for the drug activity.

Appendix A

Trial organisation Sponsor: Italfarmaco S.p.A., Milano, Italy. Executive Committee: M.L. Simoons (chairman), L. Tavazzi, A.J. van ‘t Hof, M. Sardina, E. Bonizzoni, A.J. van Boven, J.A. Kleijne. Data Safety Monitoring: J.G.P. Tijssen, W. Wijns, F.W.A. Verheugt. Clinical Event Committee: L.G.P.M. van Zeijl, J.W. Deckers, C. Vrints. Data management/Corelab: Cardialysis, Rotterdam, NL. Central Blood analysis: Carim, Maastricht, NL. Participating centres with principle and co-investigators in order of enrolment (number of patients): Amsterdam Department of Interventional Cardiology, Onze Lieve Vrouwe Gasthuis, Amsterdam, The Netherlands (NL). Principle-investigator: G.J. Laarman (101). Isala Klinieken, De Weezenlanden, Zwolle, NL. Principle investigator: A.W.J. van ‘t Hof (66); J.C.C. van der Horst. Ospedale Riuniti di Bergamo, Italy (I). Principle-investigator: G. Guagliumi (59); A. Costalunga, O. Valsecchi, M. Tespili. Catharina Ziekenhuis, Eindhoven, NL. Principle-investigator: H.R. Michels (57); P.C. Rademaker, W.A.L. Tonino, W.H. Aarnoudse. IRCCS Policlinico San Matteo, Pavia, Italy. Principle-investigator: L. Tavazzi (48); E. Bramucci, L. Angoli, M. Ferrario, A. Repetto. Thoraxcenter Erasmus University Medical Center, Rotterdam, NL. Principle-investigator: M.L. Simoons (23); J. Vos Academisch Ziekenhuis Groningen, NL. Principle-investigator: A.J. van Boven (21); L.J. Wagenaar. Sint Antonius Ziekenhuis, Nieuwegein, NL. Principle-investigator: M.J. Suttorp (9); J.G.M. Jol, J.W.M. van Eck. Thoraxcentrum Ignatius, Breda, NL. Principle-investigator: J.A.M. te Riele; (8); G. Uytdehaag Academisch Ziekenhuis Maastricht, NL. Principle-investigator: F.W.H.M. Bär (7); H. de Swart, C. de Zwaan. Ospedale Sante Croce, Cuneo, I. Principle-investigator: G. Steffenino (3); P. Russo, A. Dellavalle. Total number of patients included 402.


The authors thank Jeroen Kleijne for the data-management.


  • This study was supported by a grant from Italfarmaco S.p.A., Milano, Italy.


  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
  29. [29]
  30. [30]
  31. [31]
  32. [32]
  33. [33]
  34. [34]
  35. [35]
  36. [36]
  37. [37]
  38. [38]
  39. [39]
  40. [40]
  41. [41]
  42. [42]
  43. [43]
  44. [44]
  45. [45]
View Abstract