Our trial showed no significant difference between different dosages of rhPro-UK regarding clinical outcomes in the 4.5–6 h therapeutic time window. Without an increase in SAEs and sICH risk, the proportion of patients with slight or no neurological disability at 90 days exceeded 60% with rhPro-UK treatment at either dose. Thus, IV rhPro-UK was effective and safe for AIS patients within 4.5–6 h after stroke onset.

To date, rapid administration of rt-PA to well-selected patients remains the only pharmacotherapy approved by the Food and Drug Administration for early treatment of AIS [8]. However, only about 5–20% of patients received rt-PA treatment in clinical settings [9]. Hacke et al. [10] demonstrated that the proportion of patients with full recovery decreased with increased time gap between stroke onset and treatment initiation. Beyond 4.5 h, there was increased mortality with no significant clinical benefit.

In the third international stroke trial (IST-3) [11], AIS patients treated within 6 h of onset were enrolled, and 37% of patients in the rt-PA group versus 35% in the control group were alive and independent at 6 months. Based on Virtual International Stroke Trials Archive (VISTA) dataset, Fulton et al. [12] found that usage of simple clinical variables and plain CT alone failed to identify a population in whom the alteplase effect would be safe and effective with treatment initiation between 4.5 and 6 h. A meta-analysis of nine randomized controlled trials (RCTs), of which six RCTs could randomized patients beyond 4.5 h and up to 6 h after symptom onset, showed no evidence of significant benefit of alteplase compared with placebo after 4.5 h of stroke onset or when last seen well [13]. According to the European Stroke Organization, AIS patients of 4.5–9 h duration (known onset time), without advanced imaging, were recommended no IV thrombolysis [14]. Thus, other thrombolytic drugs with equal or superior efficacy and safety are urgently needed for AIS exceeding 4.5 h.

As a specific plasminogen activator that mainly acts on fibrin at the thrombus site, prourokinase (Pro-UK) does not form covalent complexes with protease inhibitors in plasma. Minimal deactivation by circulating inhibitors and clot-specific plasminogen activation are two advantages of Pro-UK [15]. In contrast to rt-PA, Pro-UK induces minimal (< 5%) or no re-thrombosis and exerts no procoagulant effect in blood.

In a rat model with thromboembolic stroke, Hao [16] found that intravenous infusion of Pro-UK in the acute-early stage (1.5 h after embolism) reduced the neurological deficits and decreased the lesion and infarct volume, without increasing the ICH as compared with urokinase and rt-PA. In the 4.5 h or 6 h time windows, the use of Pro-UK thrombolysis was slightly better than rt-PA without hemorrhage. In a rabbit model with cerebral ischemia at 3, 4.5, and 6 h therapeutic time windows, IV Pro-UK improved recanalization with reduced risk of cerebral hemorrhage, exerting therapeutic effects within a 6 h time frame [17].

The PROACT study demonstrated a partial or complete recanalization rate of initially occluded cerebral vessels of 57.7% with Pro-UK versus 15.4% in patients receiving placebo, substantially higher than the 34.4% observed with double-chain rt-PA given intravenously [5, 18]. In all, 40% of Pro-UK patients and 25% of control patients had a mRS of 2 or less at 90 days, despite sICH within 24 h occurring in 10% of Pro-UK patients and 2% of control patients [19]. Thus, IA local delivery Pro-UK within 6 h of the onset of AIS significantly improved recanalization efficacy and clinical outcome at 90 days, despite an increased risk of early sICH. Takano et al. [20] showed that both IA and IV Pro-UK significantly promoted reperfusion and decreased lesion in an embolic stroke model, indicating that IA and IV therapy with Pro-UK are equally effective in promoting reperfusion and inhibiting the development of focal ischemic injury. Takano et al. mentioned that, if the current IA efficacy trial of Pro-UK provides evidence that this novel thrombolytic agent improves clinical outcome when given to AIS patients within the first 6 h after onset, then it is likely that future assessment of Pro-UK treatment will occur.

Our previous study suggested that rhPro-UK intravenous thrombolysis therapy is effective for patients within 4.5 h after stroke onset [6]. In the present study, we compared the efficacy and safety between different doses of IV rhPro-UK for AIS in the 4.5–6 h time window. Without increased mortality and ICH risk, more than 60% of patients at either dose achieved functional independence at 90 days. The proportion of patients with favorable outcomes was similar to previous studies. In a meta-analysis of five randomized trials [21], the crude cumulative rates of disability-free (mRS 0–1) outcome at 3 months were tenecteplase 57.9% versus alteplase 55.4%, the crude summary sICH rates were tenecteplase 3% versus alteplase 3%, and the crude mortality rates at 3 months were tenecteplase 7.6% versus alteplase 8.1%.

There was a trend with rhPro-UK 35 mg toward an increase in the rate of patients with favorable outcome and a reduction in the mean NIHSS score compared with those in the rhPro-UK 50 mg group. There was a dose-response relationship for Pro-UK [22]. Efficacy increased from the lowest to the highest dose and increased as the time window shortened [17]. Meanwhile, some researchers found that conversion rate from single-chain Pro-UK to double-chain UK was not only related to the dosage, but also associated with the administration regimen [4]. When a 25% bolus was followed by a 60 min infusion, the conversion rates of 35 mg dosage and 50 mg dosage were 15.4% and 27.9%, respectively. When the administration regimen was changed to a 20 mg bolus first followed by a 30 min infusion, the conversion rate was significantly decreased. The conversion rate was only 14% for the 65 mg dosage, and 18.9% for the 85 mg dosage. The better trend in rhPro-UK 35 mg versus 50 mg groups might be related to the different conversion rates due to the administration regimen. This finding needs to be confirmed in future studies.

Our randomized study is limited by its small sample size. The potential benefit could be more obvious in a larger randomized cohort study. The treatment allocation was not blinded, and there was no control group. Despite taking several measures, such as selection of stroke centers with Good Clinical Practice (GCP) certification, conduction of uniform training for researchers, and regular supervision and inspection of centers, certain measure biases with the open design may not have been completely ruled out.

At the time when the trial was designed, there were no recognized safe and effective thrombolytic drugs approved for use within 4.5–6 h of onset in patients with ischemic stroke. As a phase II clinical trial for a new drug application with the primary objective of exploring the appropriate dose, only the safety and efficacy of intravenous rhPro-UK 35 mg and 50 mg were compared. In a subsequent phase III clinical trial, a randomized controlled design was used to further compare the rhPro-UK 35 mg dose group with the standard treatment. Furthermore, the optimal strategy and the influencing factors of curative effect need to be confirmed in further studies. Whether this strategy could be beneficial in improving long-term clinical outcomes needs to be clarified in larger studies. Finally, advanced imaging about perfusion lesions or vessel occlusion was not mandated. Information of reperfusion and recanalization on imaging was absent. However, the use of CT scans in the present study gives a wider applicability of this drug, as CT scans are widely available, even in smaller cities.