Tissue plasminogen activator (tPA) was approved for therapeutic fibrinolysis in 1987, when it replaced streptokinase (SK) for the treatment of acute myocardial infarction (AMI). It remains the fibrinolytic of choice today although over the last decade fibrinolysis has been replaced by percutaneous coronary intervention (PCI) whenever possible. However, PCI is a technically demanding procedure that is time-consuming whereas for AMI “time is myocardium” and only rapid reperfusion will salvage function of ischemic myocardium. Fibrinolysis is the simplest fastest method, but it requires both plasminogen activators, the second one being urokinase plasminogen activator (uPA), the native form of which is a proenzyme, prouPA. A combination of a 5 mg bolus of tPA followed by a prouPA matches the natural paradigm and was shown to be more effective safer than tPA monotherapy. Activator gene knockout studies also showed that both activators are required, but that uPA rather than tPA was the dominant one, as well as the one neglected for the past 33 years.

Common pathologies like AMI and ischemic stroke are usually triggered by a blood clot or thrombus for which therapeutic fibrinolysis with tPA has been a standard treatment. This plasminogen activator was believed responsible for fibrinolysis due to its high fibrin clot affinity [1]. It was developed to replace streptokinase, an indirect, non-specific plasminogen activator of bacterial origin.

Since tPA was of biological origin and was a direct plasminogen activator, it was expected to be much more effective and safer. Three comparative trial were conducted with a total of 95,740 AMI patients [2-4]. In the first two trials, the 30 day mortality with tPA or SK was identical [2, 3]. Only in the third trial did tPA induce a significantly greater mortality reduction but it was only in one of the four groups in this trial [4]. A Bayesian analysis of the three trials reached the conclusion that a significant mortality difference between tPA and SK treatment had not been established [5]. In all three studies, a significantly greater incidence of symptomatic intracranial hemorrhage occurred with tPA, which was also unexpected.

An explanation for these results was never given. Instead, tPA received FDA approval for the treatment of AMI [6], and has remained the fibrinolytic of choice ever since. However, consistent with these early results, the clinical efficacy of tPA in AMI and later in ischemic stroke has been disappointing. As a result, the treatment of choice for AMI is now percutaneous intervention (PCI), despite its being significantly more time-consuming and costly. An explanation for tPA’s poor efficacy can be found in the literature.

tPA monotherapy was based on a misunderstanding of fibrinolysis
There are two biological plasminogen activators, tPA and uPA whose properties are not only different but highly complementary [9]. tPA is an enzyme that strongly promoted by fibrin [1] to which it has a high affinity [10]. By contrast, uPA is a proenzyme, prouPA, with no fibrin-affinity but nevertheless has a fibrin-specific mode of action [11]. This is was found to be due to a strong substrate affinity related to a conformational change in plasminogen [12]. Due to their complementarity, the combination of the activators induced a synergistic effect [9], especially when the combination was sequential starting with tPA . The synergistic effect enables lower, safer doses to be used.

Despite this published evidence, tPA was evidently too well-established for it to be put into question. Instead, it was believed that uPA was principally an extravascular plasminogen activator [13].

Gene knockout animal studies contradicted this concept
Two independent gene deletion studies showed that fibrinolysis requires both plasminogen activators. When the tPA gene was deleted, it had surprisingly little effect on the rate of spontaneous lysis of an intravascular clot formed in the animals. By contrast, a uPA knockout induced significant inhibition of clot lysis and caused some spontaneous fibrin deposition. Knocking out both genes arrested clot lysis completely and induced extensive fibrin deposition in the animals. The authors concluded that both activators were required for fibrinolysis to function normally, and that uPA was the dominant activator [22, 23]. These findings and conclusions are consistent with in vitro and in vivo clot lysis studies [24].

The Physiological Mechanism of Fibrinolysis
Endogenous fibrinolysis is an ongoing process as evidenced by the invariable presence of the fibrinolytic degradation product, D-dimer in blood (normal concentration: 110-250 ng/ml) Only in the presence of a potent thrombin autoantibody that inhibited fibrin formation has a D-dimer closer to zero (6-33 ng/ml) been reported [7]. Interestingly, this endogenous fibrinolysis is free of bleeding complications, in contrast to therapeutic fibrinolysis. Nevertheless, it is relatively potent since in 15% of untreated AMI patients complete patency (TIMI-3) of the infarct artery was found at the time of the initial catheterization [8]. This endogenous lysis rate compares with a 45% rate in patients treated with full dose tPA in whom the blood tPA concentration is almost 1,000-fold higher. Since the endogenous plasma concentration of tPA is only ~5 ng/ml, it is not possible that this degree of lysis is due to tPA alone.

In the endogenous system, tPA’s function is limited to the initiation of fibrinolysis, a function analogous to that of the starter in a car. This occurs when tPA is released from the endothelium of the vessel wall at the site of an intravascular thrombus. Due to its high fibrin affinity, tPA binds to the thrombus [13] at a fibrin site proximal to plasminogen bound to lysine Aα157 on the fibrin surface [14]. This tPA, plasminogen, fibrin ternary complex promotes tPA’s plasminogen activation about 1,000-fold [15] and initiates fibrinolysis.

Since tPA has no second high affinity fibrin binding site and is otherwise a weak plasminogen activator [16], this completes tPA’s function in fibrinolysis. It also explains tPA’s high dose requirements when it is used alone since to activate the remaining two fibrin-bound plasminogens high doses are necessary, but even these are not fully effective.

The initiation of fibrin degradation by tPA creates new plasminogen binding sites on fibrin [18] which are two in number [17]. These plasminogens are more efficiently activated by prouPA, explaining why the activators have complementary modes of action and are synergistic in fibrinolysis when combined [18], especially when combined sequentially starting with tPA [19].

A Therapeutic Test of the Sequential Activator Combination
The PATENT study was a multi-center trial in 101 consecutive patients with AMI treated with a mini bolus of tPA (10 mg in ten and 5 mg in the remaining 91 patients) followed by a 90 minute infusion of prouPA (40 mg/h). This fibrinolytic sequential combination induced a TIMI-3 patency at 24 h in 82% of the patients and a 30-day mortality of 1% [25]. These results compared very favorably with GUSTO, the best of the tPA mega trials in which the TIMI-3 infarct artery patency at 24 h was only 45%, and the mortality was 6% [26]. These clinical results were consistent with in vitro clot lysis data showing that together the activators have a synergistic fibrinolytic effect [9] especially when they are in a sequential combination [19], an effect also reported in an animal study [27], and cited in another review [24].

Platelets Bind prouPA and Extend its Half-Life
In the PATENT trial it was noted that about 30% of the lysis took place well after the end of the prouPA’s infusion [18]. Since the prouPA’s plasma half-life is only about 7 minutes, this required an explanation

In a study of the prouPA intrinsic to blood, about 20% of it was found bound to the outer platelet membrane [28] and evidence of a novel uPA binding protein on the membrane was found [29]. Platelets also were shown to promote fibrinolysis by prouPA [30]. Since platelets have a half-life of 2.5 days, the platelet-bound prouPA has a greatly prolonged fibrinolytic half-life. This property may also help explain the low rate of reocclusion reported with prouPA fibrinolysis [25, 31].

Fibrinolysis and Bleeding
Hemorrhagic side effects, especially intracranial hemorrhage (ICH), have limited utilization of fibrinolytic therapy. For example, in ischemic stroke a 7% incidence of symptomatic ICH has been reported with tPA treatment [32]. This bleeding risk is specific for tPA, but not with uPA since it does not bind to fibrin. Bleeding by tPA is predominantly due to the lysis of hemostatic fibrin [33] to which tPA binds as it does to all fibrin, especially if it administered as a prolonged infusion. Hemostatic fibrin can be found at vascular repair sites and seems to regularly present judging by the fibrin degradation product D-dimer that is invariably present in blood (110-250 ng/ml). Bleeding from these sites is an unpredictable risk, especially when tPA is administered as an infusion rather than a bolus.

tPA and prouPA, the two natural plasminogen activators, induce fibrinolysis sequentially by complementary modes of action. As a result, when they are administered in a sequential combination, fibrinolytic efficacy is significantly promoted without any loss of fibrin specificity or safety. This natural fibrinolytic regimen has been misunderstood and monotherapy discredited the fastest treatment of thrombotic occlusions available. In a single clinical trial, this paradigm was once used therapeutically in 101 AMI patients. The result was a six-fold reduction in mortality and an almost two-fold increase in coronary patency compared with the best of the tPA mega trials. A second trial in AMI is currently underway, and a trial in ischemic stroke is nearing completion.