Ischemic stroke (IS) is one of the leading causes of mortality and disability worldwide [1]. Thrombolytic therapy (TLT) is aimed on reperfusion of cerebral blood flow in the first 4.5 hours of IS [2]. The main drug for blood flow reperfusion is a recombinant tissue plasminogen activator (alteplase, A) [3]. Despite its high efficacy, A has a number of some negative features: the need to select a dose according to the patient’s weight, intravenous (IV) infusion for 1 hour, relatively low fibrin selectivity and the associated systemic effect on hemostasis with an increased risk of hemorrhagic complications [4]. In addition, experimental studies have found that high concentrations of exogenous tissue plasminogen activator during TLT may contribute to the damage of neurons and glia [5].

All these emphasize the importance of creating more advanced thrombolytic drugs, the key criteria of which are bolus administration, the use of one standard dose regardless of the patient’s weight, high fibrin selectivity and, correspondingly, low/absent systemic effect on hemostasis and low risk of hemorrhagic complications [6].

One of these drugs is staphylokinase, which belongs to the group of exogenous plasminogen activators [7]. Staphylokinase was isolated from Staphylococcus aureus, and its fibrinolytic effect was established in 1948 [8]. The pronounced fibrin-selective activity of staphylokinase is due to its interaction only with plasminogen gamma, which is located in the thrombus [9]. An additional contribution to staphylokinase fibrin selectivity is made by a significantly higher rate of neutralization of the plasmin-staphylokinase complex by α2-antiplasmin in plasma than on the thrombus surface [10].

After cloning the gene [11, 12], the effect of recombinant staphylokinase was studied in experimental and clinical studies. The thrombolytic agent was found to be no less effective than A in reducing the foci of ischemia in experimental IS [13] and restoring blood flow in patients with ST-elevated myocardial infarction (STEMI) [14]. In STEMI, a higher frequency of reperfusion of the infarct-related coronary artery was observed in most patients treated with recombinant staphylokinase compared with the A group. However, the patients developed neutralizing antistaphylokinase antibodies after the drug administration [14], which prevented its widespread use.

In 2012, a new form of recombinant staphylokinase — non-immunogenic staphylokinase (NS) — was patented (patent RU 2448158 C1, 20.04.12). Three amino acids were replaced in the immunodominant epitope of recombinant staphylokinase, which significantly lowered its immunogenicity by reducing by more than 200 times the titers of neutralizing antistaphylokinase antibodies. The latter were determined using an indirect solid-phase enzyme immunoassay test system developed by the Institute of theoretical and experimental biophysics of the Russian academy of sciences [15].

In addition, modification of the three-dimensional structure of staphylokinase led to an increase in the formation rate of staphylokinase-plasminogen complex compared to the same interaction of plasminogen with recombinant staphylokinase [16].

Non-immunogenic staphylokinase was registered in Russia in 2012 with the trade name Fortelyzin® (INN/chemical name — recombinant protein containing the amino acid sequence of staphylokinase).

A multicenter randomized comparative clinical trial of NS vs tenecteplase (T) in patients with STEMI showed that a single IV bolus injection of NS at a dose of 15 mg regardless of body weight, had a similar frequency of reperfusion as in the T group and, at the same time, less frequent minor bleeding events, as well as the absence of hemorrhagic stroke and neutralizing antistaphylokinase antibodies [17].

The results obtained made it possible to study the efficacy and safety of NS in comparison with A in patients with ischemic stroke within 4—5 h after symptom onset.

Material and methods

The protocol of an open-label, randomized, comparative, parallel-group, non-inferiority study of NS and A in patients with IS within 4-5 hours after the onset of symptoms — “FRIDA” (Fortelyzin® in a randomized trial compared with Actilyse®) and amendments were approved by the Russian Ministry of Health and the ethics committee of the Russian Ministry of Health, local ethics committees of the clinical centers and fully complied with current Russian legislation, the principles of Good Clinical Practice (ICH GCP) and the Declaration of Helsinki (resolution of the Russian Ministry of Health No. 498 dated 15.07.2016). The study was registered with the Russian Ministry of Health (www.rosminzdarv.ru/documents) and on the website https://clinicaltrials.gov/ (NCT03151993). Randomization was carried using a blocked randomization scheme (84 blocks, block sizes of 4). Random distribution of drugs in different order was carried out within the block — 2 NS and 2 A. The clinical centers were provided with envelopes with randomization numbers which were opened in numerical order.

The FRIDA trial was planned as a non-inferiority study. The sample size calculation to determine the margin of non-inferiority was described in detail earlier [Gusev E.I., 2021] [18]. 336 patients were randomized (168 — group of A and 168 — group of NS). In the initial version of the protocol, inclusion in the study was carried out according to the indications for TLT in patient with of NIHSS >5. In the updated version of the protocol, patients with a diagnosis of mild stroke (NIHSS ≤ 4) were also enrolled in the study. These changes were made in accordance with the international clinical guidelines [2]. NS (10 mg) was administered as a single intravenous bolus for 10 s, regardless of bodyweight, and A was administered according to the instructions for medical use.

The first patient was randomized on 18.03.2017 and the last patient on 23.03.2019. Before randomization, all participants or their legal representatives signed an informed consent form to participate in the study. In cases where the patient’s condition did not allow him/her to express his/her will and there was no legal representative, and the need for treatment was urgent, the question of medical intervention in the interests of the patient and his/her inclusion in the study was resolved by a council of physicians, and, if it was impossible to convene, by the treating clinician (duty medical officer) with following notification of the Principal Investigator, officials of the clinical center and the legal representative.

The primary efficacy endpoint was a favorable outcome, defined by the modified Rankin scale (mRS) score of 0–1 on day 90 after drug administration. The secondary efficacy endpoint was defined as combined outcomes of mRS score 0–1, NIHSS score 0–1, and Barthel index score of 95 or more on day 90. Additional secondary efficacy endpoints were NIHSS score after 24 hours and on day 90. An mRS score of 0-2 on day 90 was included in a retrospective, post-hoc analysis of outcome. Safety endpoints included mortality on day 90, all cases of intracranial hemorrhages and other serious adverse events (SAE). Symptomatic intracranial hemorrhage was defined as any hemorrhage with neurological deterioration (an increase in the NIHSS score by ≥4 points from baseline or the lowest value in the first 7 days) or any hemorrhage leading to death. Additionally, hemorrhage (according to ECASS III criteria [19]) must have been identified as the predominant cause of neurological deterioration.

The statistical analysis software R (version 3.5.1) was used for data analysis. Continuous variables were presented as the mean and standard deviation if Pearson’s nonparametric skewness coefficient was less than 0.2. In other cases, median and interquartile range (25%-75% percentiles) were given for continuous variables. Qualitative variables were described by absolute and relative frequencies. When comparing two independent groups, we used a nonparametric analogue of the Student’s unpaired t-test — the Mann-Whitney test. For repeated measurement analysis, we used a nonparametric analogue of the analysis of variance (ANOVA) — the WTS criterion based on the calculation of Wald statistics [20]. In this analysis, the Holm-Bonferroni correction was taken into account when conducting intergroup multiple comparisons of parameters. To identify the relationship between categorical features in the 2×2 contingency tables, we used the two-sided Fisher’s exact test. Comparing the indicators in the two treatment groups, the odds ratio (OR) was calculated with 95% confidence intervals (95% CI); OR was considered statistically significant if the 95% CI did not include 1. The non-inferiority hypothesis was tested using the Welch t-test. Kaplan-Meyer survival curves were constructed for each of the drugs. Comparison of the survival curves for the two studied drugs was carried out using the Log-rank test. The level of statistical significance p was assumed as ≤ 0.05.

Results

Both groups were completely comparable in terms of demographic characteristics and comorbidities (Table. 1), stroke severity on admission, stroke localization and TOAST stroke classification and the time from the onset of symptoms to TLT (Table 2). The IS subtype distribution according to the TOAST classification corresponded to previous studies in Europe and in the USA [21].

Table 1. Demographic and clinical characteristics of patients at admission

Table 2. Distribution of patients by severity, localization and subtype of IS and time before the start of TLT

In the first 3 hours of IS, thrombolytic therapy in the NS group was performed in 85 (51%) patients, in the A group — 84 (50%) patients. The number of days in hospital did not differ statistically significantly between the two groups and was 16.0 ± 7.5 and 15.5 ± 5.8, respectively (p = 0.67 for the Mann-Whitney criterion).

Primary and secondary endpoints and additional efficacy criteria

On day 90 after drug administration, 84 (50%) patients in the NS group reached the primary endpoint (mRS 0-1), in the A group — 68 (41%) patients (p=0.10, OR = 1.47, 95% CI = 0.93-2.32) (Table 3).

Table 3. Results of evaluation of the effectiveness and safety of treatment

Note. * — tendency to significant differences at p<0.10.

There were no differences in the frequency of reaching the secondary combined endpoint (90 days) and additional efficacy criterion (NIHSS scores after 24 hours and after 90 days). There were no differences in survival by day 90 between the groups (Fig. 1).

Fig. 1. Survival at 90 days after stroke.

When analyzing the relationship between the subtype of stroke according to TOAST classification and the efficacy of thrombolytic therapy, there was a trend towards a higher efficacy in the frequency of reaching the primary endpoint (mRS = 0-1) in patients with non-atherothrombotic stroke (only patients with cardioembolic, lacunar and undetermined stroke were included in the analysis) in the NS group compared with the A group. The primary endpoint in the NS group was met in 63 out of 124 patients, while in the A group– in 46 out of 120 (p = 0.054, OR = 1.66, 95% CI = 0.97-2.86) (Table 4).

Table 4. Results of treatment effectiveness depending on the subtype of IS, n (%)

The difference in the rate of favorable outcomes between the study groups according to the distribution of mRS score on day 90 was 9.5% (95% CI = -1.7-20.7). The lower limit of the 95% CI did not cross the margin of “non-inferiority” (Pnon-inferiority<0.0001; Fig. 2), not reaching only 1.7% to the margin of greater efficacy (superiority) of NS compared with A.

Fig. 2. The hypothesis of «non-inferiority» of NS in comparison with A.

Safety criteria

When assessing the safety criteria, significant differences between the treatment groups were noted in the frequency of severe adverse events SAE: 22 (13%), in the NS group, 37 (22%) in the A group (p = 0.044, OR = 0.53, 95% CI = 0.28-0.98). There was also a trend towards significant differences in the rates of symptomatic intracranial hemorrhage: in the NS group — 5 (3%) patients, in the A group — 13 (8%) patients (p = 0.087, OR = 0.37, 95% CI = 0.1-1.13). In the first 3 hours, the percentage of symptomatic intracranial hemorrhage in the NS group was 4% (3 cases per 85 patients), in the A group — 7% (5 cases per 84 patients). In the interval from 3 to 4.5 hours, symptomatic intracranial hemorrhage in the NS group was observed in 2% of patients (2 out of 83), in the A group — in 10% (8 out of 84 patients); (p = 0.099; OR = 0.24, 95% CI = 0.02-1.24). There were no differences in the rates of major and minor bleeding according to TIMI classification and in the rates of deaths between the study groups. Cerebral edema with the shift of midline structures was observed less frequently in the NS group than in the A group (7 vs 14 patients, respectively, p = 0.17).

Fibrin selectivity

An important feature of NS was its pronounced fibrin-selective effect, which was confirmed by the difference in the dynamics of changes in fibrinogen level between the two study groups. The concentration of fibrinogen in blood before TLT in the NS group and in the A group averaged 3.24 g/l and 3.22 g/l, respectively, and did not differ significantly (p = 0.36). However, starting from the 6^th hour and almost until the 28^th hour after TLT onset, at each moment of observation, a statistically significant decrease in the fibrinogen concentration was observed in the A group compared with the NS group (Fig. 3). At the same time, on the 7^th day after TLT (p = 0.38) and at discharge (p = 0.68), there was no difference in fibrinogen concentration between both studied groups.

Fig. 3. Dynamics of fibrinogen content in blood after TLT.

Discussion and conclusion

The study outcomes are the world’s first data on the use of a staphylokinase-based drug in patients with IS. Staphylokinase belongs to a group of external plasminogen activators, and the main feature of these thrombolytic agents is immunogenicity. In most cases, it is difficult to overcome immunogenicity, since the reduction/elimination of the immunodominant epitope can lead to the instability of the molecule and a decrease in its fibrinolytic activity [22]. In this regard, NS has fundamental differences: replacement of 3 amino acids in the immunodominant epitope allowed to reduce the titers of neutralizing antistaphylokinase antibodies compared with other variants of recombinant staphylokinase and at the same time increase the formation rate of its complex with plasminogen. Thus, the titer of anti-staphylokinase antibodies to NS on day 30 after its administration was less than 0.3 mcg/ml and was 25-135 times lower compared with the staphylokinase variants SakSTAR.M38, SakSTAR.M89 and SakSTAR [22, 23].

Evaluation of the efficacy of NS based on primary and secondary endpoints and an additional criterion showed that it was not inferior in efficacy to A. Furthermore, when analyzing based on the patient’s weight and the subtype of stroke according to the TOAST classification, a trend towards its greater efficacy was observed. NS exceeded by 9.5% the margin of “non-inferiority” at the primary endpoint (mRS 0-1 score) — 50% and 40.5%, not reaching only 1.7% to the margin of greater efficacy (superiority) compared with A.

Another important feature of NS was the safety of its use. In the NS group, the frequency of SAE was significantly lower than in the A group. Fewer deaths were observed in the NS group compared with the A group (by day 90) — 10% and 14% and symptomatic hemorrhagic transformation — 3% and 8%.

An extremely important positive feature of NS is its pronounced fibrin-selective effect, which was manifested by a significantly higher level of fibrinogen during the first 28 hours after TLT. These results correspond to the results of the multicenter study, in which the fibrin-selective effect of NS in STEMI patients was also shown. Administration of NS led to a minimal decrease in blood fibrinogen level compared with other thrombolytic drugs, including T [17], which currently is the most fibrin-selective drug from the group of r-tPA derivatives. It is important to note that the selectivity of staphylokinase action concerns both fibrinogen and other ways of hemostasis. According to Okada K. et al. [24], administration of recombinant staphylokinase Sak42D was accompanied by a significantly less pronounced increase in the fibrinopeptide level and the thrombin-antithrombin III complex compared with alteplase in patients with myocardial infarction. Thus, the available data confirm an extremely high selectivity of staphylokinase action and the absence of systemic activation of hemostasis during its application.

It is also necessary to note the efficacy of NS in patients with non-atherothrombotic stroke. This may be due to the activation of staphylokinase on the thrombus surface [9, 25] or thrombus structure [26]. According to Khismatullin RR. et al. [27], blood clots in the cardioembolic IS subtype have a higher fibrin content compared to those in the atherothrombotic stroke subtype — 23% and 13% (p = 0.034), respectively, which may lead to greater efficacy of NS in the non-atherothrombotic variants of IS, as it was shown in the FRIDA study.

Administration and dosing features of NS can be attributed to its undeniable advantages. A single bolus injection of NS at a standard dose of 10 mg, regardless of bodyweight, allows to perform a rapid thrombolytic therapy, especially when approaching the upper time limit of 4.5 hours and avoiding possible errors when adjusting the dosage to bodyweight.

In the future, NS can also be used at the prehospital stage in ambulances equipped with CT scanners, which may shift the TLT time limits to the left and increase the frequency of TLT in patients with IS, as had already been done in the treatment of patients with STEMI.

Thus, further clinical studies of NS will be focused on expanding the therapeutic window of 4.5 h after the onset of symptoms in patients with IS.

The authors declare no conflicts of interest.