The Future of Fibrinolysis Agents

Abstract

Since the 1990s, the second-generation tissue plasminogen activator alteplase has been accepted as the gold standard in the thrombolysis of acute ischemic stroke, acute myocardial infarction, and pulmonary embolism. Alteplase, however, is not without its limitations, including a short half-life precluding single-bolus administration and elevated risk of both local and systemic bleeding. Efforts have been made to create “third-generation agents” with longer half-lives, improved fibrin specificity, and safety profiles compared to alteplase. The majority of the current Federal Drug Authority (FDA) approved third-generation agents, in particular tenecteplase and reteplase, are derivatives of alteplase. This chapter intends to compare their mechanism of action, dosing, efficacy, and safety profiles based on current evidence. Beyond tissue plasminogen activators, the chapter provides an overview of other developments in the field of fibrinolysis, including advanced drug delivery mechanisms, combination therapy, direct fibrinolysis agents, and other adjuncts.

1. Introduction

1.1 Mechanism of thrombosis and fibrinolysis

Clot formation, or thrombosis, is a physiological response intended to limit the extent of bleeding. Thrombosis is typically balanced against intrinsic antithrombotic and fibrinolytic entities, which limit the extent of thrombus formation to only areas of tissue or vessel injury. Conversely, pathological thrombus formation occurs in the setting of conditions such as myocardial infarction, pulmonary embolism, or ischemic stroke, resulting in complete occlusion of an arterial lumen even in the absence of proportionate injury [1].

Thrombosis occurs through the interaction of three major entities—blood vessels, platelets, and clotting factors [2]. During primary hemostasis, vascular damage leads to vasoconstriction, platelet adhesion, activation, and aggregation culminating in the formation of a weak platelet plug [3]. The platelet plug is stabilized by a fibrin matrix, which is formed through the interaction of clotting factors, with the endpoint being the conversion of fibrinogen to fibrin [4]. At the beginning of secondary hemostasis, the extrinsic, and intrinsic coagulation pathways converge in the common pathway through the activation of factor Xa. Factor Xa combines with its co-factor factor Va and calcium ions on the phospholipid surface of cells to create a prothrombinase complex, which converts inactive prothrombin (also known as factor 2a) to active thrombin. Thrombin then converts fibrinogen to its active form, fibrin, and additionally activates factor XIII, which helps to cross-link fibrin [5].

On the other hand, Fibrinolysis is defined as the process by which the fibrin matrix of a clot is broken down at its fluid-solid interface (Figure 1). It involves the interaction of plasminogen, plasmin, and plasminogen activators, as shown in Figure 1. Plasmin is the key protease in the breakdown of the fibrin matrix, and its precursor is plasminogen [6]. Plasminogen binds to both eukaryotic cell surfaces and fibrin via plasminogen receptors [7]. Specifically, binding occurs when lysine sites within plasminogen Kringles adhere to Carboxyl-terminal lysyl residues present on cell surfaces. The process of binding to fibrin on a clotted surface, or plasminogen receptors on a cell surface, markedly enhances plasminogen’s activity [8]. This activity enhancement occurs through several mechanisms, including (1) Glu-plasminogen, following interaction with cell surfaces, undergoes conformational changes, which augment its conversion to Lys-plasminogen, a more readily-activated form of plasminogen [9]. (2) When bound to cell surfaces or fibrin, plasminogen is less readily inhibited by its major inhibitor, alpha2-antiplasmin, which competes for the Carboxyl-terminal lysyl residues on the cell surface.

Figure 1.

The process of hemostasis leading to the formation of the fibrin matrix, followed by fibrinolysis mediated by tissue plasminogen activators.

Following the binding of plasminogen to fibrin on the surface of a clot, tissue plasminogen activators cleave plasminogen to plasmin, which in turn mediates fibrin degradation [10]. The conversion of plasminogen to plasmin is mediated by several plasminogen activators in the body—namely, tissue plasminogen activators produced by the vessel endothelium and urokinase produced by the kidneys [11]. This process is irreversibly inhibited by several enzymes, namely plasminogen activator inhibitor 1 (PAI-1), which degrades free tPA, and alpha 2-antiplasmin which degrades plasmin in the circulation [12], the by-products of which are eliminated by the liver.

Fibrinolytic agents work in either an indirect or direct [13] manner to mediate the conversion of plasminogen to plasmin, which then degrades fibrin [14]. Figure 2 details some of the factors that determine the efficacy and safety of a fibrinolysis agent at different time points during the process of thrombolysis. First, the “ideal” fibrinolytic agent would be easy to give as a single bolus (as opposed to a prolonged infusion). Following administration, it would quickly reach its intended site of action without interaction with systemic circulation. The agent would have low immunogenicity with minimal host response. Upon reaching the clot, the fibrinolytic agent would have a rapid onset of action and excellent clot penetration, swiftly inducing recanalization in order to limit ischemic damage to the target organ. The agent would have a prolonged half-life with decreased inhibition by PAI-1 and antiplasmin, preventing it from being cleared too quickly from circulation. Post-thrombolysis, several adverse events may occur: (1) the release of plasmin and fibrin degradation products may trigger the coagulation cascade leading to reocclusion, (2) breakdown of the blood-brain barrier, particularly in the setting of acute ischemic stroke, may increase the risk of symptomatic intracranial hemorrhage, and (3) elsewhere in the body, the systemic action of the fibrinolytic agent may elevate the risk of major bleeding. In an ideal situation, there would be adjuncts acting together with the fibrinolytic agent to prevent such post-thrombolysis events from occurring.

Figure 2.

Pre-, intra-, and post-thrombolysis factors associated with the efficacy and safety of a fibrinolysis agent.

The first generation of fibrinolysis agents was streptokinase and urokinase. Streptokinase, a microbial plasminogen activator derived from beta-hemolytic streptococcus, forms a 1:1 stoichiometric complex with both circulating tissue plasminogen and intrinsic thrombus plasminogen, thereby mediating the conversion of plasminogen to plasmin. The GISSI-1 [15] and ISIS-2 [16] trials demonstrated significant mortality benefits in patients receiving streptokinase at 1 hour and within 24 hours of symptom onset, respectively. A major downside of first-generation agents was their lack of fibrin specificity, leading to activation of both clot-bound and circulating plasminogen and increasing systemic bleeding risk [17]. Additionally, streptokinase, being of non-human origin, led to significant immunogenicity with the inevitable development of host antibodies within a few weeks of use [18]. This impeded repeated usage within 3 weeks and led to antigenicity-related reactions, including anaphylaxis and dose-dependent hypotension. The addition of plasminogen to streptokinase, leading to the creation of an acylated plasminogen-streptokinase activator complex (APSAC, or anistreplase), failed to overcome these drawbacks [19]. Nonetheless, streptokinase continues to be used in developing countries due to its accessibility and lower cost.

1.2 Alteplase as the basis of third-generation tissue plasminogen activators

Alteplase (also known as Activase, rt-PA, or tPA) is a second-generation tissue-type plasminogen activator identical to native human tissue plasminogen activator released by vascular endothelial cells. It is synthetically produced from recombinant DNA technology in Chinese hamster ovarian (CHO) cells. Alteplase consists of 527 amino acids with a molecular weight of 65 kDa [20]. Unlike streptokinase, alteplase is highly fibrin-specific, as the presence of fibrin in a clot increases its plasminogen-activating activity. In the absence of fibrin, it does not activate plasminogen in plasma due to competitive inhibition [21]. It also does not possess the antigenicity of streptokinase.

The chemical structure of alteplase consists of several key regions (Table 1)—a fibronectin finger region and kringle domain 2, which bind to fibrin, and a serine protease domain, which converts plasminogen to plasmin. The epidermal growth factor domain and kringle domain 1 of rt-PA mediate its elimination by hepatocytes, while the protease domain is inhibited by PAI-1 [22, 23].

Component of alteplase	Purpose	Alteration in tenecteplase	Alteration in reteplase
Fibronectin finger	Facilitates binding to fibrin	Nil	Deleted
Epidermal growth factor domain	Elimination by liver	Nil	Deleted, reducing elimination by hepatocytes
Kringle 1	Elimination by liver	1. Replacement of threonine by asparagine at position 103, leading to a new glycosylation site that reduces hepatic elimination	Deleted, reducing elimination by liver endothelial cells
		2. Replacement of asparagine by glutamine in position 117, thereby removing the carbohydrate side chain that facilitates hepatic elimination
Kringle 2	Stimulates protease in the presence of fibrin	Nil	Nil
Protease domain	Splits plasminogen. Inhibited by PAI-1	Replacement of four amino acids in position 296–299 with 4 x alanine, resulting in 80-times reduction in PAI-1 inhibition compared to alteplase	Nil
Carbohydrate side chain	Elimination from plasma	Nil	Nil

Table 1.

Components of alteplase and associated modifications in tenecteplase and reteplase.

Alteplase is currently recognized as the “gold standard” in fibrinolytic therapy, having been validated in the treatment of ischemic stroke, acute myocardial infarction, and pulmonary embolism in various large-scale phase III trials. The NINDS [24], ATLANTIS [25], and ECASS [26] trials conducted in the 1990s found that alteplase achieved, on average, a 30% decrease in disability at 90 days, when compared to placebo, in eligible patients within 3–4.5 hours [27]. There was, however, a pooled 8.6% incidence of symptomatic intracranial hemorrhage. In the GUSTO-1 trial, alteplase was compared against streptokinase, achieving 90-minute TIMI grade 3 flow in 50–60% of patients, with a 14% greater reduction in mortality as compared to streptokinase. In terms of safety profile, alteplase achieved a 1% absolute risk reduction in death and/or non-fatal stroke as compared to streptokinase [28]. In patients with sub-massive pulmonary embolism, the MAPETT 3 trial demonstrated that treatment with alteplase was associated with a significantly lower rate of in-hospital death and clinical deterioration compared to heparin alone [29].

2. Third-generation tissue plasminogen activators

Despite its widespread acceptance, alteplase is not without its drawbacks. It has thus far only achieved a median 30% efficacy in improving functional outcomes when administered within a short time window of 3–4.5 hours [27]. Alteplase is strongly inhibited by PAI-1, resulting in a short half-life of less than 5 minutes, necessitating its administration as a single bolus followed by infusion(s) [30]. Furthermore, it has a relatively slow onset of action, reaching peak plasma concentrations at 60 minutes, allowing time for infarct expansion.

Various third-generation agents have been developed with the aim of achieving a longer half-life, improved fibrin specificity, faster onset of action, and greater ease of administration as single or double boluses. The first mechanism by which this is achieved is through modifications of the structure of alteplase. These include the additional, deletion, or subtraction of amino acids (such as in the case of tenecteplase/reteplase/monteplase), hybridization of the tissue plasminogen activator with other molecules (amideplase), and the use of different conformations of tPA [31]. Other third-generation agents are novel variants of tissue plasminogen activators derived from animals (such as the vampire bat in the case of desmoteplase) [32] or bacteria (such as Staphylococcus aureus in the case of staphylokinase) [33].

2.1 Tenecteplase

2.1.1 Pharmacokinetics

Tenecteplase (also known as TNKase or TNK-tPA) is a derivative of alteplase involving three point mutations in its kringle 1 and protease domains. In the kringle 1 domain, asparagine replaces threonine at position 103, leading to the creation of a new glycosylation site that reduces hepatic elimination. Additionally, glutamine replaces asparagine in position 117, resulting in the removal of a carbohydrate side chain that facilitates hepatic elimination. The third point mutation occurs in the protease domain, in which four amino acids in positions 296–299 of the protease domain are replaced by 4 x alanine. This results in an 80-fold reduction in inhibition by PAI-1 as compared to alteplase. Due to reduced hepatic elimination and increased resistance to PAI-1, tenecteplase has a decreased plasma clearance rate of 105 ml/min, a prolonged initial plasma half-life of 17–22 minutes, and a terminal half-life of around 115 minutes [34]. Studies have drawn an association between body weight and plasma clearance of tenecteplase, with a 10-kilogram increase in body weight associated with an increase in plasma clearance by 9.6 ml/min [35]. This aspect of tenecteplase’s pharmacokinetics has formed the basis of a weight-adjusted dosing system for patients with acute coronary syndrome. Tenecteplase is also more fibrin-specific than alteplase [36].

2.1.2 Dosing

Acute ischemic stroke: the TNK2SB trial of 2010 compared at 0.1 mg/kg, 0.25 mg/kg, and 0.4 mg/kg in stroke patients presenting at </=3 hours from onset. Although the trial was prematurely terminated due to slow enrollment, it found an increased incidence of symptomatic intracranial hemorrhage at a dose of 0.4 mg/kg as compared to other doses of tenecteplase and alteplase [37]. A subsequent phase IIB trial, the Australia TNK trial by Parsons et al. in 2012, found that a dose of 0.25 mg/kg of tenecteplase was superior to 0.1 mg/kg of tenecteplase and alteplase in terms of achieving its primary endpoint of clinical improvement (as defined by NIHSS) at 24 hours [38]. Phase III EXTEND-IA TNK trial by Campbell et al. in 2018 subsequently found that 0.4 mg/kg of tenecteplase was not superior to 0.25 mg/kg in achieving reperfusion of ischemic strokes with large vessel occlusion [39]. The most recent trial to demonstrate the drawbacks of administering 0.4 mg/kg of tenecteplase was the phase III NOR-TEST 2 trial in 2022, which found that this dose of tenecteplase produced inferior 90-day functional outcomes as compared to alteplase, and was associated with a higher incidence of symptomatic intracranial hemorrhage (16% for 0.4 mg/kg TNK, 5% for t-PA, p = 0.013) [40]. Since then, the accepted dose of tenecteplase has been 0.25 mg/kg in the setting of acute ischemic stroke.

Acute coronary syndrome: In the TIMI 10B trial, single boluses of 30 mg tenecteplase, 50 mg tenecteplase, and an accelerated alteplase regimen were compared, with the primary endpoint being TIMI grade 3 flow at 90 minutes post-thrombolysis [41]. These doses of tenecteplase were derived from the Phase 1 TIMI 10A trial [42]. However, shortly after the trial began, it was found that 50 mg tenecteplase was associated with an almost 3-fold increase in incidence of intracranial hemorrhage, leading to a dose reduction to 40 mg tenecteplase. Subsequent analysis of data from the TIMI 10B and in-parallel ASSENT 1 trial found that weight-adjusted dosing was strongly correlated with arterial patency and reduced risk of bleeding [43]. This led to the development of a weight-based dosing system as follows:

30 mg for <60.0 kg,

35 mg for 60.0–69.9 kg,

40 mg for 70–79.9 kg,

45 mg for 80–89.9 kg, and.

50 mg for >90.0 kg, given over 5–10 seconds.

Tenecteplase is typically given in combination with heparin (particularly in patients aged below 75 years old) and an antiplatelet.

2.1.3 Efficacy and safety profile

Acute Ischemic Stroke: Table 2 details major trials comparing tenecteplase to alteplase in the setting of both acute ischemic stroke and acute coronary syndrome.

Trial (year)	N	TNK dose	NIHSS	Window	Primary endpoint	Secondary endpoints	Safety endpoints
TRACE 2, 2023, Wang	1430	0.25 mg/kg	5–25	4.5 hours	90-day functional outcome (mRS 0–1)	90-day mortality	sICH
NOR-TEST 2 (2022), Kvistad	204	0.4 mg/kg	13	4.5 hours	90-day functional outcome (mRS 0–1)	90-day mortality	sICH
AcT (2022), Menon	1577	0.25 mg/kg	8–9	4.5 hours	90–120 day functional outcome (mRS 0–1)	90-day mortality	sICH
TASTE-A (2022), Bivard	104	0.25 mg/kg	8	4.5 hours to MSU	Volume of lesion perfused on arrival to hospital from MSU	90-day mRS 5 or 6, 90-day mortality	sICH
TRACE 1 (2021), Li	236	0.1, 0.25 and 0.32 mg/kg	7–8	3 hours	90-day functional outcome (mRS 0–1)	—	sICH
EXTEND-IA TNK (2018), Campbell	202	0.25 mg/kg	17, w LVO	4.5 hours	Reperfusion >50% of ischemic territory or absence of retrievable thrombus	90-day functional outcome by median mRS	sICH, 90-day mortality
NOR-TEST (2017), Logallo	1107	0.4 mg/kg	5.6–5.8	4.5 hours or wake-up stroke	90-day functional outcome (mRS 0–1)	90 day mortality	sICH, serious adverse events
ATTEST (2015), Huang	96	0.25 mg/kg	11–12	4.5 hours	Percentage of penumbra salvaged	—	sICH, serious adverse events
Australia TNK Trial (2012), Parsons	75	0.1, 0.25 mg/kg	14	6 hours	Clinical improvement at 24 hours based on NIHSS	Reperfusion at 24 hours	sICH, serious adverse events
TNK2SB (2010), Haley	112	0.1, 0.25, 0.4 mg/kg	8–10 for TNK, 13 for tPA	3 hours	Findings could not be convincingly interpreted in view of slow enrollment leading to premature termination.
NA/China (2009), Liang	110	0.53 mg/kg	—	6 hours	90-minute TIMI grade 3 flow	30-day mortality	ICH, major hemorrhage
NA /UAE (2004), Binbrek	266	Weight-based	—	6 hours	Time from onset of chest pain to recanalization by serial changes in CK-MM in blood	30-day mortality	—
ASSENT 2 (1999), Van Der Werf	16, 949	Weight-based	—	6 hours	30-day all-cause mortality	—	Non-cerebral bleeds, need for transfusion
TIMI 10B (1998), Cannon	886	30 mg, 50 mg > 40 mg TNK	—	12 hours	90-minute TIMI grade 3 flow	Mortality	After reduction to 40 mg TNK: ICH, moderate to severe bled

Table 2.

Randomized controlled trials comparing tenecteplase to alteplase in the setting of acute ischemic stroke and acute coronary syndrome. Legend: in comparison to alteplase, the trial found that TNK was superior (green), comparable (orange), or inferior (red) to alteplase.

The Australia TNK trial in 2012 was the first successful phase IIB trial, following the premature termination of the 2010 TNKS2B trial, to demonstrate that 0.25 mg/kg of tenecteplase was superior to alteplase in achieving its primary endpoint of clinical improvement at 24 hours based on NIHSS, with no elevated risk of symptomatic intracranial hemorrhage. The NOR-TEST trial in 2017 was the first large-scale phase III trial to compare 0.4 mg/kg of tenecteplase with alteplase, involving 1107 participants presenting within 4.5 hours of stroke onset and/or wake-up stroke. It found that tenecteplase was non-inferior to alteplase in terms of both primary and secondary endpoints of 90-day functional outcome, 90-day mortality, sICH, and serious adverse events [44]. The year after, the EXTEND-IA TNK trial involving patients with large vessel occlusion found that a lower dose of 0.25 mg/kg of tenecteplase achieved superior reperfusion of ischemic territory and 90-day functional outcome compared to alteplase. Additionally, it did not find any difference in outcomes between the 0.4 mg/kg and 0.25 mg/kg doses of tenecteplase. This finding was confirmed by TRACE 1 in 2021, which found that 0.32 mg/kg of tenecteplase was not superior to a dose of 0.25 mg/kg [45]. The NOR-TEST 2 trial in 2022, which attempted to examine the dose of 0.4 mg/kg tenecteplase in patients with moderate to severe stroke with an average NIHSS of 13, found that this dose of tenecteplase was associated with increased incidence of symptomatic ICH and mortality.

The largest-scale phase III trial to date is AcT 2022, involving 1577 participants who received either 0.25 mg/kg tenecteplase or alteplase within 4.5 hours of stroke onset [46]. Both AcT and TRACE 2 [47], which involved 1430 participants from China, demonstrated non-inferiority (but not superiority) of tenecteplase in terms of 90-day functional outcome, mortality, and incidence of sICH.

Rather than compare tenecteplase to alteplase, newer studies are now examining the application of tenecteplase in specific subgroups of patients, particularly those with late presentations from symptom onset. The phase III TWIST trial, which spanned 77 hospitals in 10 countries, focused on patients with wake-up strokes selected using non-contrast brain imaging. It found no difference in outcomes between tenecteplase and placebo, although there was also no increased incidence of sICH in thrombolysed patients [48]. The TIMELESS trial in 2024 found a modest benefit of 0.25 mg/kg tenecteplase in patients with M1 occlusion presenting between 4.5 and 24 hours from symptom onset, the majority of whom had undergone endovascular thrombectomy post-thrombolysis [49]. Tenecteplase has also been studied in other settings, such as mobile stroke units, in the 2022 TASTE-A trial.

Myocardial infarction: The TIMI 10B and in-parallel ASSENT 1 Phase II trials led to the development of a weight-based dosing system for patients with acute coronary syndrome. This was first employed in a large-scale phase III trial in 1999, the ASSENT-2 trial. In this study, 16,949 participants presenting within 6 hours of acute ST-elevation myocardial infarction (STEMI) were given either alteplase or weight-adjusted tenecteplase. While tenecteplase did not achieve decreased 30-day all-cause mortality compared to alteplase, it was associated with significantly fewer non-cerebral bleeding complications and reduced requirement for blood transfusions. Additionally, the ASSENT 2 investigators found that a subset of patients presenting beyond 4 hours of symptom onset enjoyed improved 30-day mortality in those receiving tenecteplase [50]. A subsequent study in 2004 by Binbrek et al. found that tenecteplase was associated with a shorter time interval to recanalization of 208 +/− 10 minutes compared to 237 +/− 9 minutes for alteplase, as measured by serial changes in isozyme creatine kinase-MM samples [51]. A meta-analysis of the three most recent trials found that tenecteplase was equivalent to alteplase in efficacy but had a reduced risk of major bleeding (risk ratio: 0.79, p = 0.0002) [52].

More recent studies of tenecteplase in the setting of acute coronary syndrome include the landmark STREAM trial, which compared fibrinolysis alone to primary percutaneous coronary intervention (PCI) in patients presenting within 3 hours of STEMI onset. It found that administration of weight-adjusted tenecteplase was non-inferior to primary PCI in achieving the primary endpoint, which was a composite of death, shock, congestive heart failure, or reinfarction up to 30 days, suggesting that tenecteplase alone may be a viable alternative to intervention in early-presenting patients for whom a PCI-capable institution is unavailable. However, higher rates of intracranial hemorrhage were observed in patients treated with tenecteplase compared to primary PCI (2 vs. 1%, p = 0.04), although there was no difference in rates of non-cerebral bleeding [53]. Pre-PCI treatment with tenecteplase, however, resulted in increased incidence of adverse outcomes including death, heart failure, and cardiogenic shock in the 2023 ASSENT 4 trial [54].

Tenecteplase has yet to be included in high-quality trials in the treatment of pulmonary embolism and/or venous thromboembolism.

2.1.4 Comparing tenecteplase to alteplase

Table 3 summarizes the differences between alteplase, tenecteplase, and reteplase, in terms of pharmacokinetics and molecular properties, as well as efficacy and safety in the setting of acute ischemic stroke and acute coronary syndrome.

Category	Tenecteplase	Alteplase	Reteplase
Molecule	Derivative of alteplase involving three point mutations in the kringle 1 and protease domains. 65kDA, 527 amino acids.	Identical to human tissue plasminogen activator, produced from recombinant DNA technology. 65kDA, 527 amino acids.	Single chain deletion mutant of alteplase consisting of the kringle 2 and protease domains of alteplase. 39kDA, 355 amino acids.
Half-life (biphasic)	Initial: 17–22 minutes Terminal: 115 minutes	Initial: </=5 minutes Terminal: 26.5–46 minutes	Initial: 14–20 minutes Terminal: 96–120 minutes
Onset	About 11 minutes	30 minutes	30–60 minutes
Peak plasma	At the end of 5–10 second bolus	Around 60–90 minutes	Around 30 minutes at the end of the double bolus
PAI-1 inhibition	80-fold less than alteplase	++	+
Plasma clearance	105 ml/min, four times slower than alteplase. Affected by body weight.	453 ml/min	250–450 ml/min
Fibrin specificity	++	+	++
Admin	Single bolus over 5–10 seconds	Bolus followed by infusion (s) over 60–90 minutes	Double bolus 30 minutes apart, with each bolus given over 2 minutes
Cost	$6300 for a 50 mg vial	$9200 for a 100 mg vial	About half the cost of alteplase
Acute ischemic stroke (AIS)
Dose	0.25 mg/kg single bolus over 5–10 seconds	0.9 mg/kg, given as a bolus followed by an infusion over 60 minutes	Double bolus 18 mg + 18 mg, 30 minutes apart
90-day function	Non-inferior. Exception: may be superior in large vessel occlusion (EXTEND-IA TNK)	—	Non-inferior
Mortality	Non-inferior	—	NA
Volume of reperfusion	Superior to alteplase but may not translate to better outcomes	—	NA
Symptomatic ICH	Similar incidence	—	Similar incidence
Acute Coronary Syndrome (ACS)
Dose	Weight-based system, or 0.53 mg/kg as single bolus	Either as an accelerated infusion over 60 minutes or 3-hour infusion	Double bolus of 10MU + 10 MU, 30 minutes apart.
TIMI grade 3 flow	Non-inferior	—	Superior to alteplase in RAPID 1 and RAPID 2 trials. Non-inferior in GUSTO 3.
Mortality	Non-inferior. Exception: may be superior if patients present beyond 4 hours from symptom onset (ASSENT 2)	—	Non-inferior.
Bleeding	Decreased incidence of non-cerebral bleeds or need for transfusion (ASSENT 2), or similar incidence (TIMI 10B)	—	Similar incidence. Decreased incidence of hemorrhage stroke in RAPID 1.

Table 3.

Comparison between tenecteplase, reteplase, and alteplase.

Pharmacologically, tenecteplase has a longer half-life of 17–22 minutes, a 4-fold decreased rate of plasma clearance, and 80-fold decreased PAI-1 inhibition compared to alteplase. It also has a faster onset of action at around 11 minutes and achieves peak plasma concentrations at the end of a 5–10 second single bolus with superior ease of administration. This is compared to alteplase, which needs to be given as a bolus followed by an infusion lasting from 60 minutes to 3 hours due to its short half-life of <5 minutes. Tenecteplase is, at present, also more cost-efficient than alteplase.

In the setting of acute stroke, tenecteplase is generally non-inferior rather than superior to alteplase in terms of 90-day functional outcomes and mortality, although it may be associated with a faster rate of reperfusion. There is, however, evidence to suggest that in certain subsets of patients, particularly those with large vessel occlusions, tenecteplase may have added benefit over alteplase. Safety outcomes of 0.25 mg/kg of tenecteplase are generally similar to alteplase, although 0.4 mg/kg may be associated with an increased risk of bleeding, particularly in patients with moderate to severe stroke.

For patients with acute coronary syndrome, tenecteplase is likewise associated with more rapid recanalization, but this generally does not translate to improved 90-minute TIMI grade 3 flow or mortality. However, in the subset of patients presenting late, beyond 4 hours of infarct onset, the ASSENT 2 trial found that tenecteplase improved mortality rates. Tenecteplase was also associated with a decreased incidence of cerebral and non-cerebral bleeding in the ASSENT 2 trial, although other studies generally showed a similar incidence of bleeding to alteplase.

2.2 Reteplase

2.2.1 Mechanism

Reteplase is derived from alteplase through a single-chain deletion of alteplase’s fibronectin finger region, epidermal growth factor, and kringle 1 domains [55]. It consists primarily of the kringle 2 and protease domains of alteplase and has a shorter chain of 355 amino acids. Deletion of the epidermal growth factor and kringle 1 domains reduces hepatic elimination, thereby increasing its plasma half-life to 14–18 minutes [56].

2.2.2 Dosing

For acute myocardial infarction, reteplase is typically given as two boluses of 10MU 30 minutes apart, with each bolus given over 2 hours. The RAPID 1 trial found that the 10MU + 10MU double-bolus dose of reteplase achieved higher rates of 90-minute TIMI grade 3 flow as compared to a 15MU single bolus or 10 + 5MU double bolus (62.7% for 10MU + 10MU, 40.9% for 15MU and 45.7% for 10 + 5MU) [57]. For acute ischemic stroke, the 18 mg + 18 mg double-bolus dose of reteplase is currently being studied in a phase III trial [58].

2.2.3 Efficacy and safety profile

Acute coronary syndrome (ACS): reteplase was compared to alteplase in three major trials—the RAPID 1, RAPID 2 [59], and GUSTO III trials. The phase II RAPID 1 trial (Smalling 1995) involving 606 patients found that the double-bolus 10 U + 10 U dose of reteplase was more effective in achieving 90-minute TIMI grade 3 flow than other doses of reteplase (single bolus 15MU, or double bolus 10 U + 5 U bolus) and alteplase itself, with equivalent safety outcomes in terms of 30-day mortality, rate of major bleeding, reinfarction, and requirement for additional coronary interventions. Reteplase was also associated with a lower incidence of hemorrhage stroke than alteplase (0.2% for r-PA vs. 3.9% for t-PA, p = 0.03). The phase III RAPID 2 trial likewise found that 59.9% of patients treated with reteplase achieved TIMI grade 3 flow compared to 45.2% of those receiving alteplase (p = 0.01) within 12 hours of ACS onset, with significantly reduced requirement for additional coronary intervention [60]. However, the larger-scale GUSTO III trial involving 15,059 patients presenting within 6 hours found no significant difference in its primary endpoint of 30-day mortality (p = 0.54). 1-year mortality, hemorrhage stroke, reinfarction, congestive cardiac failure, and major bleeding rates were also equivalent for patients treated with 10 U + 10 U reteplase and alteplas [61].

Acute ischemic stroke: the reteplase versus alteplase for acute ischemic stroke (RAISE) trial was the first phase 2 trial to compare intravenous reteplase at doses of 12 + 12 mg, 18 + 18 mg, and intravenous alteplase, in patients with acute ischemic stroke presenting within 4.5 hours of onset. For its safety outcome, it found no difference in the incidence of symptomatic intracranial hemorrhage among patients receiving both doses of reteplase and alteplase. It also found non-inferiority of both doses of reteplase and alteplase, with roughly equal proportions of patients (75% for 12 + 12 mg, 72.7% for 18 mg + 18 mg, and 78% for alteplase) achieving its primary endpoint based on NIHSS [62]. There are now plans for a phase 3 trial involving 1412 patients comparing 18 + 18 mg reteplase and alteplase.

2.3 Other 3rd-generation tissue plasminogen activators under study: done

Table 4 provides details on other third-generation tissue plasminogen activators that have shown promise but ultimately not achieved ubiquity in clinical practice. Of these, lanoteplase (novel or n-PA), monteplase (E6010), duteplase (met-t-PA), YM866, and amideplase are 3rd-generation derivatives of alteplase. Lanoteplase has been the most-studied derivative thus far. With a long plasma half-life of up to 45 minutes, it could be given as a single 120 IU/kg bolus. The InTIME-1 and InTIME-2 trials explored the use of lanoteplase in the treatment of early myocardial infarction and achieved a higher 90-minutes patency rate than alteplase [63]. However, there was an increased rate of hemorrhagic stroke seen in the InTIME-2 trial (1.12% n-PA, 0.64% tPA, p = 0.004). Duteplase likewise achieved a promising rate of 70% 90-minute coronary patency in the ESPRIT trial but had high rates of reocclusion and reinfarction at 7% each [64]. Monteplase has shown promise in patients with treatment-resistant deep vein thrombosis, although this has only been demonstrated in small-scale observational studies [65]. Amideplase (CGP 42935, MEN 9036, or Ktu-PA) is a hybrid of the kringle 2 domain of t-PA and the catalytic protease domain of urokinase (scu-PA). It has a longer half-life of more than 30 minutes. So far, this agent has been tested in animal models with evidence of improved clot penetration, but there have been no human trials conducted to date [66].

Agent	Structure	T 1/2	Fibrin specificity	Dose	Clinical	Downside
Lanote-plase (novel, n-PA)	Deletion of fibronectin finger and epidermal growth factor (EGF) domain of alteplase. Glutamine replaces asparagine at position 117.	45 minutes	+	120 IU/kg single bolus	InTIME-1 and InTIME-2 trials: achieved higher 90-minutes patency rate than alteplase in early MI	Increased incidence of hemorrhagic stroke in InTIME-2 (1.12% n-PA vs. 0.64% t-PA).
Dute-plase (met-t-PA)	Replacement of valine by methionine at position 245.		Less fibrin-specific than alteplase	0.6MU/kg infusion over 4 hours	ESPRIT trial: achieved 70% 90-minute coronary patency.	Moderate incidence of reocclusion, with 7% re-occluding in 36 hours.
Monte-plase (E6010)	Cys in EGF domain replaced by serine.	23 minutes	+++	0.22 mg/kg bolus	Effective in treatment-resistant deep vein thrombosis (DVT) in observational study.	No large-scale trials.
Pamite-plase (YM866, Solinase)	Deletion of position 92–173 in kringle 1, arginine replaces glutamine at position 275	27 minutes	+++	0.1 mg/kg single bolus	Japanese trial: Achieved 50% 60-minute TIMI 3 flow.	No trials with safety data.
Amide-plase	Hybrid of kringle 2 domain of alteplase and catalytic protease domain of scu-PA.	>30 minutes	Less fibrin-specific than alteplase	Not available	Better clot penetration than alteplase in animal models.	No humans trials, low fibrin specificity.
Staphylo-kinase (Staphylokinase (SAK), SAK42D, SakSTAR)	Secreted by staph aureus, similar mechanism to streptokinase.	6 minutes	++++	20-30 mg over 30 minutes	Recombinant staphylokinase (STAR), CAPTOR 1 and 2 trials: as effective as tPA in coronary recanalization and achieving TIMI-3 flow	Significant antigenicity with development of host antibodies within 2 weeks.
Desmo-teplase (Bat-PA, v0PA, DSPAalpha)	1 of 4 plasminogen activators found in vampire bats collectively called Demodus salivary plasminogen activators (DSPAS).	2.8 hours	+++	125 μg/ kg as single bolus	DIAS Phase 2 trial: when given at 3-9 hours of stroke onset, had better reperfusion rates and clinical outcomes than placebo.	No difference in 90-day functional outcomes in the DIAS 2 and DIAS 3 trials compared to placebo.

Table 4.

Non-FDA-approved third-generation tissue plasminogen activators.

Non-alteplase derivatives include staphylokinase (SAK, or SAK42D) and desmoteplase. Staphylokinase has a similar mechanism of action to streptokinase but additionally requires other plasminogen activators to form an active complex in order to function. As such, it was thought to be more fibrin-specific than streptokinase. In the setting of acute coronary syndrome, the STAR trial in 1995 [67] and CAPTOR 1 and 2 trials in the 2000s [68, 69] found that staphylokinase achieved non-inferior rates of coronary recanalization and 90-minute TIMI grade 3 flow compared to alteplase, with no significant increase in allergic reactions. However, its use has been limited by its significant immunogenicity, with nearly all patients found to have developed host-mediated antibodies within the first two weeks of treatment.

Desmoteplase (Bat-PA, v0PA, DSPAalpha1) is another unique plasminogen activator selected out of four plasminogen activators found in vampire bats collectively known as the Demodus salivary plasminogen activators (DSPAS). Desmoteplase is the DSPAS with the longest half-life at 2.8 hours. Additionally, it is inactive in the absence of fibrin and can be inhibited by tranexamic acid. The Phase II DIAS trial, which administered a single bolus of 125 microgram/kg of desmoteplase to acute ischemic stroke patients at 3–9 hours of onset, initially had promising results with improved reperfusion compared to placebo [70]; however, this did not translate to improved 90-day functional outcomes in DIAS 2 and DIAS 3 [71, 72].

3. Beyond tissue plasminogen activators

Table 5 provides details of adjuncts and alternatives to tissue plasminogen activators that may address their limitations pre-, intra-, and post-thrombolysis. These include advanced drug delivery systems like liposomes, combination therapy with anti-thrombotics, direct fibrinolytic agents like plasmin, direct thrombin inhibitors, and adjuncts shown to protect the blood-brain barrier.

Category	Examples	Benefits	Drawbacks
Advanced drug delivery systems (ADDSs)	Conjugation of plasminogen activators to PEG, liposomes, polymeric carriers. Ultrasound-sensitive ELIPs, hydrostatic pressure-sensitive SA-NTs. red blood cell (RBC) carriers.	More targeted delivery of drug to site of action, prolonged dwell time, decreased interaction with systemic circulation, decreased immunogenicity.	Limited human trials. Requirement for localized ultrasound. RBC carriers only suitable for thrombo-prophylaxis.
Combination therapy with anti-thrombotics	tPA/TNK/r-PA + abciximab, eptifibatide, aspirin, tirofiban	Ameliorate the procoagulant effect of plasmin post-thrombolysis, preventing reocclusion. Reduced-dose tPA with abciximab may reduce risk of bleeding.	Mixed findings apart from abciximab. Aspirin, eptifibatide may be associated with increased risk of bleeding.
Direct fibrinolysis agents	Plasmin, micro-plasmin, delta-plasmin, alfimeprase	In preclinical models, found to have superior clot lysis and decreased rate of local bleeding compared to urokinase.	No evidence of clinical benefit apart from more rapid recanalization. Requires catheter-based delivery system. Rapidly inhibited by systemic antiplasmin.
Blood-Brain Barrier (BBB) protective agents	Drug: ascorbic acid, minocycline, GCSF, imatinib, batimastat (matrix metalloproteinase (MMP) inhibitor), bryostatin (protein kinase C (PKC) modulator) and atorvastatin	Protects BBB from breakdown and bleed beyond In preclinical models to reduce area of infarct, extent of hemorrhagic transformation, and improve neurological recovery in preclinical models	No clear evidence of clinical benefit due to limited human studies.
	Non-drug: neural stem cells, normobaric oxygen

Table 5.

Novel adjuncts and alternatives to tissue plasminogen activators currently under study.

3.1 Advanced drug delivery systems (ADDSs)

Ensuring that fibrinolytic agents concentrate their effect on the target site of action alone without interacting with systemic circulation is a persistent problem encountered in their use, despite the increased fibrin specificity of third-generation tissue plasminogen activators (tPAs). Advanced drug delivery systems (ADDSs) attempt to address this issue through several approaches [73].

Firstly, conjugation of plasminogen activators with polyethylene glycol (PEG), liposomes, and other polymeric carriers (dendrimers) may protect them from degradation, reduce immunogenicity, and increase dwell time. A second approach would be the use of locally-applied ultrasound to trigger the release of plasminogen activators contained within ultrasound-sensitive carriers such as echogenic liposomes (ELIPs) and PEG-gelatin nanocarriers [74]. In addition to accelerating tPA release, the ultrasound disintegrates clots directly, allowing for improved drug permeation. Where a locally-applied ultrasound is not feasible, shear-activated nanotherapeutics (SA-NTs) have been developed which disintegrate to release their constituents at areas of arterial narrowing with increased shear stress [75].

Finally, erythrocytes have been considered as vehicles for the transport of encapsulated drugs given via transfusions. As clots are impermeable to erythrocytes once established, RBC carriers have been suggested for transient targeted thromboprophylaxis rather than therapeutic thrombolysis. In-vitro studies found that RBCs conjugated to plasminogen activators remained in circulation for hours to days and also had fibrin affinity without being inactivated by PAI-1 [76].

3.2 Combination therapy

A major drawback of any tissue plasminogen activator is that their administration produces a paradoxical procoagulant response through the generation of excess plasmin and hemostatic reaction products [77]. These activate the coagulation cascade and lead to thrombin formation, platelet aggregation, and subsequent reocclusion with failure of thrombolysis [78]. The co-administration of heparin was not shown to effectively eliminate this response [79]. Various studies have examined if giving anti-thrombotics across thrombolysis could prevent reocclusion.

GP IIb/IIIa inhibitors act by binding to glycoprotein receptors on the platelet membrane responsible for platelet aggregation, competitively inhibiting fibrinogen and von Willebrand factor (vWF) from binding to the same receptors [80]. As explained in the prior paragraph, it is postulated that this would inhibit a key step in re-thrombosis after the administration of fibrinolytic agents. Examples of GP IIb/IIIa inhibitors include intravenous eptifibatide [81], which is given as boluses followed by a continuous infusion over 12 hours, and abciximab [82] derived from Fab fragments of GP IIb/IIIa receptor-targeting immunoglobulins.

The most successful clinical trials to date have involved combining alteplase (or its derivatives) with abciximab. The first large-scale phase 3 trial to do so was TIMI 14, involving 888 patients presenting within 12 hours of infarct onset. It found that combination therapy with reduced-dose 50 mg alteplase and abciximab achieved superior 60 and 90-minute TIMI grade 3 flow, with a reduced risk of major hemorrhage, as compared to full-dose alteplase monotherapy. This finding was replicated with half-dose reteplase and abciximab in the year 2000 SPEED trial. However, a larger-scale trial in 2001, GUSTO 5, did not find an associated improvement in 30-day mortality with combination therapy and moreover found that the risk of intracranial hemorrhage was doubled with combination therapy in patients above the age of 75 years old [83]. The ASSENT 3 trial in 2001, which combined half-dose tenecteplase with abciximab, found a decreased incidence of its composite endpoint of 30-day mortality, in-hospital reinfarction, or refractory ischemia in its combination therapy arm, although the best outcomes in this trial were achieved with full-dose tenecteplase combined with enoxaparin for 7 days (rather than abciximab) [84].

Various smaller Phase II and IIb trials have examined tissue plasminogen activators in combination with eptifibatide (CLEAR, CLEAR-ER, CLEAR-FDR, and INTEGRITI) [85, 86, 87, 88] and the direct thrombin inhibitor argatroban (ARTSS, ARTSS-2) [89]. The Phase II CLEAR-ER trial found that reduced-dose alteplase of 0.6 mg/kg combined with eptifibatide 225 micrograms/kg achieved comparable functional outcomes in patients with acute ischemic stroke. Moreover, there was a significantly lower incidence of sICH in patients receiving combination therapy at 2% compared to 12% in the alteplase monotherapy arm. The INTEGRITI phase 2 trial in 2003 compared full-dose tenecteplase monotherapy against half-dose tenecteplase combined with eptifibatide in the setting of acute coronary syndrome. However, no difference was found in the primary endpoint of 60-minute TIMI grade 3 flow between both groups, although there was a trend toward improved ST-segment resolution and arterial patency in the combination arm. There was a greater incidence of major hemorrhage and major bleeding requiring transfusion in the group of patients receiving combination therapy.

The more recent phase 3 ARTIS trial in 2010 examined the combination of 300 mg aspirin with full-dose alteplase. It was prematurely terminated after finding that the addition of aspirin was associated with a greater risk of symptomatic intracranial haemorrhage (ICH) (4.3% for rt-PA/aspirin vs. 1.6% rt-PA alone) without clear clinical benefit [90].

3.3 Direct fibrinolysis agents

Prior to the development of tissue plasminogen activators, another approach that received significant attention were direct fibrinolysis agents, i.e., plasmin, micro-plasmin, delta-plasmin, and alfimeprase [13]. Plasmin was found to have greater lytic activity than t-PA in an in-vitro model of catheter delivery [91]. Additionally, it was demonstrated in a rabbit ear-puncture bleeding model that there were no events of rebleeding after administration of plasmin as compared to alteplase [92]. Alfimeprase was also found to be superior to urokinase in clot lysis, with reduced myocardial reperfusion injury and myocardial infarct sizes in animal models [93]. However, tissue plasminogen activators were ultimately favored due to concerns that plasmin would be rapidly neutralized by systemic antiplasmin [94]. Additionally, plasmin and its variants required catheter-based delivery for which the required expertise and equipment were not readily available.

Among the agents under study, only plasmin, micro-plasmin, and alfimeprase have been examined in human trials. Plasmin was found in various phase 1 and phase 2 trials to be safe and effective in the treatment of critical limb ischemia due to peripheral arterial or graft occlusion [95]. Micro-plasmin was studied in large vessel occlusion, peripheral arterial, and graft occlusion, with general findings of increased rates of recanalization, which did not translate into clinical benefit. Majority of the trials were prematurely terminated [13]. Micro-plasmin, however, was found to be used in vitreoretinal disease, where the local injection of micro-plasmin was found to decrease the need for vitrectomy [96]. Alfimeprase has also achieved encouraging results in the treatment of occlusive peripheral arterial disease with superior rates of recanalization but was associated with hypotension in 18% cases [97].

3.4 Overcoming the neurotoxicity of tissue plasminogen activators

tPA is ultimately found to be effective in only 30% of patients [98]. Besides the risk of re-thrombosis and hemorrhage, another explanation for treatment failure could be the neurotoxicity of tPA [99]. tPA in both endogenous and exogenous forms has been found to hasten the ischemic death of brain cells, thereby increasing infarct size and cerebral edema. The mechanism behind this is not fully understood, but several preclinical models have demonstrated potential explanations including (1) increased permeability of the blood-brain barrier through the activation of metalloproteases that lead to matrix degradation [100], (2) cleavage of the NR1 subunit of the N-methyl-D-aspartate (NMDA) receptor [101] with consequent release of intracellular calcium thereby inducing apoptosis, and (3) the induction of cerebral vasodilatation in non-ischemic areas leading to decreased blood flow to the penumbra [102].

NMDA receptor antagonists have been of particular clinical interest, given the role of NMDA receptor activation in driving tPA-induced neurotoxicity. In particular, recombinant ADAMTS 13, NMDA receptor antagonist M-801, and NR2B-specific NMDA receptor antagonist ifendopril have been tested in preclinical settings [103].

Various adjuncts have also been proposed with the intent of stabilizing the blood-brain barrier and reducing ischemic cell death following tPA administration. In animal models, agents such as ascorbic acid, minocycline, granulocyte colony-stimulating factor (GCSF), imatinib, batimastat (MMP inhibitor), bryostatin (PKC modulator), and atorvastatin have shown promise. Non-drug approaches, such as the administration of normobaric oxygen and neural stem cells, have also been explored. These agents have variously been proven to reduce infarct volume, the incidence of hemorrhagic transformation, and improve neurological function in mouse models. However, such findings have yet to be translated into actual clinical benefits or have been approved for human studies [104].

4. Conclusion

There has been heated debate about whether it is time for tenecteplase to replace alteplase as the gold standard in thrombolysis. While the majority of clinical trials comparing tenecteplase to alteplase have proven non-inferiority but failed to prove superiority, tenecteplase may nonetheless be preferred in subsets of patients, including those presenting beyond four hours of symptom onset, and stroke patients with large vessel occlusions. Moreover, its ease of administration as a single bolus and currently lower cost may place it in good standing as a viable alternative to alteplase. Further areas of study would include head-to-head comparisons of tenecteplase and reteplase, novel delivery systems for fibrinolysis agents, including ultrasound-guided and liposomal modules, administration of fibrinolytic agents in varying clinical settings E.g. in mobile stroke units/pre-endovascular thrombectomy, as well as adjunct therapies with the potential to ameliorate the procoagulant and/or bleeding side effects of tissue plasminogen activators.

Conflict of interest

The author declares no conflict of interest.