Systematic Review - Combining Neuroprotection With Reperfusion in Acute Ischemic Stroke

Background Clinical trials of neuroprotection in acute ischemic stroke (AIS) have provided disappointing results. Reperfusion may be a necessary condition for positive effects of neuroprotective treatments. This systematic review provides an overview of efficacy of neuroprotective agents in combination with reperfusion therapy in AIS. Methods A literature search was performed on the following databases, namely PubMed, Embase, Web of Science, Cochrane Library, Emcare. All databases were searched up to September 23rd 2021. All randomized controlled trials in which patients were treated with neuroprotective strategies within 12 h of stroke onset in combination with intravenous thrombolysis (IVT), endovascular therapy (EVT), or both were included. Results We screened 1,764 titles/abstracts and included 30 full reports of unique studies with a total of 16,160 patients. In 15 studies neuroprotectants were tested for clinical efficacy, where all patients had to receive reperfusion therapies, either IVT and/or EVT. Heterogeneity in reported outcome measures was observed. Treatment was associated with improved clinical outcome for: 1) uric acid in patients treated with EVT and IVT, 2) nerinetide in patients who underwent EVT without IVT, 3) imatinib in stroke patients treated with IVT with or without EVT, 4) remote ischemic perconditioning and IVT, and 5) high-flow normobaric oxygen treatment after EVT, with or without IVT. Conclusion Studies specifically testing effects of neuroprotective agents in addition to IVT and/or EVT are scarce. Future neuroprotection studies should report standardized functional outcome measures and combine neuroprotective agents with reperfusion therapies in AIS or aim to include prespecified subgroup analyses for treatment with IVT and/or EVT.


INTRODUCTION
Intravenous thrombolytic therapy (IVT) has become standard care for acute ischemic stroke (AIS), but only a small minority (12%) of patients is eligible for IVT because of the limited time window and contra-indications (1). The absolute benefit of treatment with IVT is limited and is estimated to be 4-10% (2). In the last decade, endovascular therapy (EVT) to mechanically reopen the occluded cerebral artery has led to an improvement of functional outcome in patients with AIS caused by large vessel occlusion (LVO) (3). However, despite high recanalization rates (70-90%) chances of good functional outcome after EVT remain relatively low (30-60%) (3,4). Currently only 10% of patients after EVT are without stroke symptoms at 3 months follow-up with a modified Rankin Scale (mRS) score of 0 (3,5). This implies the need for additional treatment and systems-based interventions to further improve recovery of patients with AIS. A wide range of neuroprotective agents has been investigated in the past to reduce brain injury and thereby improve patient recovery. Despite promising results from animal studies, none of the tested neuroprotective strategies appeared effective in clinical trials (6). Earlier trials may have failed due to a lack of recanalization in treating patients with AIS. As ischemic tissue will eventually become infarcted if blood flow is not restored, adequate reperfusion is probably a necessary condition for recovery with or without additional neuroprotective treatments (4,(7)(8)(9). The four primary treatment targets are reduction of excitotoxicity, oxidative stress, inflammation, and cellular apoptosis (10). In patients with adequate recanalization, another targeted mechanism is reducing reperfusion injury (7,11). With the introduction of IVT and EVT, drugs with neuroprotective properties can now be investigated in combination with reperfusion therapy. This systematic review provides an overview of randomized controlled trials (RCTs) of neuroprotective agents in AIS as an adjunct to IVT and/or EVT.

Search Strategy
This systematic review adhered to The Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) 2020 statement and focused on RCTs in patients with AIS undergoing reperfusion therapy and neuroprotective therapy within 12 h of stroke onset. We selected studies by searching electronic databases (PubMed, Embase, Web of Science, Cochrane Library, Emcare) up to and including September 23rd 2021 using the search query specified in Box 1 (Supplementary Material). Two authors (EMV and VJG) independently screened the search results by reading titles and abstracts. All disagreements between raters were included for full text screening. Full text screening was done by two authors authors (EMV and IRW), all disagreements were re-evaluated until consensus was achieved.

Selection Criteria
RCTs studying neuroprotective agents in patients with AIS in addition to any form of reperfusion therapy within 12 h of stroke onset were selected. Exclusion criteria included non-English articles, an unclear inclusion interval or unclear timing of treatment with the neuroprotective agent, and articles not reporting exact numbers of patients treated with reperfusion treatments (Figure 1). The reference lists of all relevant articles were screened for additional studies.

Data Extraction
Analyzed primary outcomes were mRS, National Institutes of Health Stroke Scale (NIHSS), Barthel Index (BI), mortality, and risk of hemorrhage. Analyzed secondary outcomes were radiological markers including rate of recanalization measured with the thrombolysis in cerebral infarction (TICI) grading scale, and infarct volume measured by CT or MRI if available. Odds ratios (OR) for good functional outcome (mRS 0-2 vs. [3][4][5][6] and mortality (mRS 6) were calculated with reported mRS score distributions if available. All populations were assessed for type of reperfusion therapy (IVT, EVT, or both) and percentage of LVO. When no data were available on percentage of LVO in the cohort, associated radiological information was documented. Time to reperfusion therapy was noted. Neuroprotective treatment regimens were extracted including time between start of reperfusion therapy and neuroprotective treatment. Whether neuroprotective treatment was started before or after reperfusion therapy was noted. Results were grouped according to the main mechanism of action.

Availability
The extraction form is available in the Supplementary Material. All data may be shared upon request.

Assessment of Bias
All included articles were evaluated with the modified Cochrane collaboration tool to assess risk of bias for RCTs.

Role of the Funding Source
The authors received no funding for this study. Funders of the trials included in this review did not have any influence on the design, interpretation, and reporting of this review.

RESULTS
The search performed on September 23rd 2021 (initial search date: November 13th 2020) yielded 1,764 articles, of which 117 were selected for full text screening. Ultimately, 30 studies were included in this systematic review, encompassing a total of 16160 patients. Results are presented in Tables 1, 2, subdivided into studies combining neuroprotection with either IVT or EVT, and further categorized by presumed mechanism of action: reduction of excitotoxicity, oxidative stress, inflammation, cellular apoptosis, and other/non-pharmacological. Most neuroprotective agents have an effect on multiple mechanisms. Figure 2 shows an overview of neuroprotective agents and their assumed mechanisms of action. In the majority of studies (26/30), patients were treated within minutes to several hours after IVT or before completion of EVT. Three studies reported on pre-hospital administration of neuroprotective treatment; only one specifically administered neuroprotective treatment after recanalization with EVT. Fifteen studies were specifically designed to investigate effects of neuroprotective treatment in addition to reperfusion therapy, i.e., all patients had to receive reperfusion therapy, either IVT and/or EVT. Of these 15 studies with 100% reperfusion therapy, in only 3 studies all patients were treated with EVT. Five of these 15 studies showed a positive result of neuroprotective treatment on functional outcome. In the other 15 studies not all patients were treated with reperfusion therapy. The number of patients treated with IVT or EVT was limited or predefined subgroup analyses on patients receiving IVT or EVT were not reported. Our calculated ORs for good clinical outcome (mRS 0-2) did not reach statistical significance in any of the reported trials. The OR for mortality (mRS 6) was only significantly lower in the study reporting on normobaric oxygen.
An overview of unadjusted ORs for good functional outcome and mortality is presented in the Supplementary Material. Risk of bias assessment is presented in Table 3. Studies in which <25% of patients was treated with IVT are summarized in Tables 1, 2, but not separately mentioned in the results below.

Clomethiazole
Clomethiazole is a positive allosteric modulator of GABA (42). The CLASS-T trial (15) tested the effect of clomethiazole in patients concomitantly treated with IVT. All patients (n = 190)

Magnesium
The use of pre-hospital intravenous magnesium was investigated in 2015 (26). In total 1700 patients were treated, of whom 73% (n = 1,246) had ischemic stroke. Of these patients, 36% received IVT. Main outcome was the mRS at 90 days. Mean mRS scores did not differ between the magnesium group and controls (2.7 in each group). Good functional outcome was observed in 52.4% of the magnesium treated group and 52.8% in controls (OR 1.0; 95%CI 0.8-1.2).

Transdermal Glyceryl Trinitrate
Transdermal glyceryl trinitrate (GTN), a nitric oxide donor, was tested in patients presenting with symptoms of stroke in 2015 (43). Additional analyses were published on a predefined subgroup of early treated patients (<6 h) in this trial (27). All types of stroke were included, of which 76% (n = 208) was ischemic stroke. One third of AIS patients was treated with IVT. Patients that randomized early to GTN had a significant shift toward a lower mRS (adjusted common OR 0.5; 95%CI 0.3-0.8). Good functional outcome was 48,6% in the GTN group and 39,5% in controls (OR 1.4; 95%CI 0.9-2.3). No subgroup analysis for patients treated with IVT was performed.

NXY-059
NXY-059 is a free radical trapping agent. Two large, RCTs (18) included a total of 4946 patients who were treated with intravenous NXY-059 within 6 hof stroke onset; over one third received IVT as well. Primary outcome was the mRS at 3 months. Among patients treated with IVT, there was no association between NXY-059 treatment and mRS score (mRS 0-2 43.0 vs. 42.4%, OR 1.1; 95%CI 0.9-1.1).

Cerebrolysin
Cerebrolysin is a porcine brain-derived preparation of lowmolecular-weight neuropeptides with free amino acids that show pharmacodynamic properties similar to those of naturally occurring neurotrophic factors. In 2013, cerebrolysin was studied

Epigallocatechin Gallate
Epigallocatechin gallate (EGCG) is a naturally occurring polyphenol from green tea extracts. It was tested in 371 AIS patients all treated with IVT (30). The trial reported positive results with lower NIHSS scores in the EGCG treated group in patients treated 3-5 h compared to placebo, though it is unclear whether these differences were significant since no ratios or confidence intervals were reported. MRS scores were not reported.

Ebselen
Ebselen is a lipid-soluble seleno-organic compound that was tested in 99 stroke patients for its possible neuroprotective properties reducing oxidative stress (12). It was administered orally within 12 h of stroke onset. More than two thirds of patients were treated with IVT. Primary outcome was CT infarct volume and the Glasgow Outcome Scale. Mean CT infarct volume was 11.4 ml in Ebselen treated patients vs. 13

Cellular Apoptosis
No studies fulfilled the inclusion criteria.

Other/Non-pharmacological Mechanisms
Hypothermia ICTuS-L tested induced hypothermia at 33 • C in AIS patients who also received IVT (21). In total 58 patients were enrolled and 48 (79%) were treated with IVT. Though the primary objective of the study was to investigate safety, assessed efficacy outcomes were NIHSS and mRS. There were no differences in outcome, neither in the overall population, nor in subgroups according to IVT. At 3 months, 18% of patients in the hypothermia group had a mRS score of 0-1 vs. 24% in controls (P < 0.77). No OR for good clinical outcome was calculated as exact data on the distribution of mRS scores was not reported.

Remote Ischemic Perconditioning
Remote ischemic perconditioning (rPerc) is inducing ischemia in one organ to increase tolerability in another. Four trials investigated the use of rPerc in stroke patients receiving reperfusion therapy: two combining rPerc with IVT, and two trials also treating patients with EVT (see paragraph on neuroprotection and EVT). In 2020 a single center trial (n = 68) reported significant results in a trial of rPerc in which patients with AIS were treated with IVT and subsequent ischemic perconditioning (37). Excellent recovery (mRS 0-1) at 3 months was obtained in 72% of patients in the rPerc group vs. 50% in controls (adjusted risk ratio, 9.85; 95%CI, 1.5 to 63.2), though patients with severe stroke (NIHSS >24) were excluded from this trial. A second trial investigated rPerc in 49 patients all treated with IVT (39). Primary outcome was the occurrence of hemorrhagic transformation or other adverse events. Secondary outcome measures included the ratio of mRS 0-1 at 90 days, distribution of mRS, and NIHSS scores. There were no differences in recovery between groups treated with and without rPerc (mRS 0-1 62.5 vs. 68.0% in controls, p = 0.69).

Nerinetide
The eicosapeptide nerinetide was investigated in the NA-1 trial, comparing nerinetide treatment with placebo in 1105 patients with LVO eligible for EVT (36). The primary outcome was favorable outcome defined as an mRS 0-2 at 90 days. Concomitant alteplase was administered in 60% of patients. The  The authors hypothesize that an interaction between nerinetide and the use of IVT might have nullified treatment effect with nerinetide in the IVT subgroup.

Oxidative Stress
Albumin Intravenous administration of albumin in 422 AIS patients was reported in 2013 (23). The majority of patients received IVT (68%), with a small subgroup receiving additional EVT (16%) or EVT alone (4%). Primary endpoint was favorable outcome (mRS 0-1, NIHSS 0-1, or both). The trial was halted prematurely because of futility after an unscheduled third interim analysis. The primary outcome was adjusted for the use of IVT and baseline NIHSS and did not differ between patients in the albumin group and controls (44.0 vs. 44.0%; risk ratio 1.0; 95%CI 0.8-1.1). The proportion of patients with good functional outcome did not differ between treatment groups (56.7 vs. 57.0%, OR 1.0; 95%CI 0.7-1.3). No subgroup analysis was done for the group of patients treated with EVT, and rates of successful recanalization were not reported.

Uric Acid
Uric acid is a naturally occurring antioxidant. A pilot study was conducted in 2007 treating 16 patients with uric acid after IVT proving its safety (17). The URICO-ICTUS trial consequently studied a population of 411 stroke patients (44). In 2017 results from a prespecified subgroup analysis were reported with 45 patients who received IVT and EVT and were all treated with intravenous uric acid or placebo within 4.5 h of stroke onset (31). Recanalization was achieved in >80% of patients. Primary outcome was good functional outcome (mRS 0-2) and was observed in 16/24 (67%) patients treated with uric acid and 10/21 (48%) in controls (adjusted OR, 6.1, 95%CI 1.1-34.6). The unadjusted OR analysis did not show a significant effect (OR 2.0 95%CI 0.7-7.3) of uric acid on good functional outcome.

Dodecafluoropentane
Dodecafluoropentane (DDFPe) is an oxygen transporting nanodroplet with neuroprotective properties (33). In 2019 administration in 24 patients with AIS was investigated, starting infusion within 12 h of stroke onset (44). Three different dose tiers were used. In total of 16/18 DDFPe treated patients received reperfusion therapy (IVT = 61%, EVT 6%, IVT + EVT 22%). Outcomes were (serious) adverse events, NIHSS and mRS scores. When only the high-dose group of DDFPe was compared to placebo, there was an improvement in both 30-and 90-day mRS (P = 0.01 and P = 0.03, respectively). No OR for good clinical outcome was calculated as exact data on the distribution of mRS scores were not reported. Sample sizes were too small for subgroup analyses in patients with LVO.

Theophylline
Theophylline is speculated to be neuroprotective due to its cerebral vasoactive properties (31). It was investigated as addon to IVT in 64 patients with AIS (38). All patients received theophylline intravenously within 30 min of IVT and 14% of patients (n = 9) was also treated with EVT. A limited number of patients with LVO was included as the majority of patients with severe stroke was unable to provide informed consent. Outcomes were NIHSS improvement and infarct volume growth at 24 h. After 3.5 years the steering committee decided to stop the trial due to slow recruitment. Patients treated with theophylline improved on the NIHSS by 4.7 points (SD, 5.6) compared with an improvement of 1.3 points (SD, 7.5) after IVT alone (P = 0.044), though results did not reach significance after correction for multiplicity [adjusted estimated improvement 3.6 (95%CI, 0.

Inflammation
Otaplimastat Otaplimastat is a cerebroprotectant that inhibits the matrix metalloprotease pathway (34). It was tested in AIS patients receiving IVT in 2019 (34). The trial was divided in a safety trial (n = 11) and randomized controlled efficacy trial (n = 69). The latter part compared two dose tiers (40/80 mg) vs. placebo. In total 29% of patients underwent EVT in addition to IVT. Primary endpoints were intracranial hemorrhage, systemic bleeding and the mRS. No significant differences in the occurrence of hemorrhagic transformation were found between groups. A nonsignificant shift in distribution of mRS was reported in the otaplimastat 40 mg group vs. placebo with an adjusted OR of 3.2 (95%CI 0.9-10.9). Good functional outcome did not differ between groups (75.0 vs. 76.2%, OR 0.9; 95%CI 0.3-3.2). Of patients treated with EVT, 77% achieved good recanalization (mTICI score ≥2b).

Cellular Apoptosis
Imatinib Imatinib is a tyrosine kinase inhibitor which reduces cellular apoptosis by blocking the signaling of platelet-derived growth factor alpha. It was investigated in 60 stroke patients in three different dose levels (400, 600, and 800 mg, orally) (29). All patients received IVT and 42% of patients underwent EVT. Primary outcome measure was the occurrence of adverse events. Secondary outcomes included infarct volume, NIHSS improvement and good functional outcome (mRS 0-2). Treating AIS patients with imatinib was safe. Patients treated with imatinib had more NIHSS improvement than controls with a significant difference measured in the high dose group, adjusted for treatment with EVT [low, medium, and high dose: 2 points (p = 0.259), 3 points (p = 0.106), and 5 points (p =

Remote Ischemic Perconditioning
Two trials investigated the use of rPerc in stroke patients treated with EVT. The first trial investigated the pre-hospital use of rPerc in 443 stroke patients (25). All MRI proven stroke patients who underwent IVT were analyzed. Primary end point was penumbral salvage (volume of perfusion-diffusion mismatch not progressing to infarction after 1 month These results indicate a high potential for some neuroprotective treatments in addition to reperfusion. Studies investigating neuroprotective treatment in addition to IVT that report a positive effect of neuroprotection on clinical outcome are the trials investigating GTN, imatinib and remote ischemic perconditioning (rPerc). Unfortunately, two thirds of patients did not receive reperfusion therapy in the GTN trial. Early treatment of stroke patients with GTN was recently investigated in the prehospital transdermal glyceryl trinitrate in patients with ultraacute presumed stroke (RIGHT-2) trial (45). The trial did not show a beneficial effect of GTN in pre-hospital treatment of stroke. Since no subgroup analyses on patients treated with IVT were reported this study fell outside the scope of our current search. The trial investigating imatinib was primarily a safety trial with a limited number of patients. A follow up trial is currently ongoing and estimated to be completed in 2023. One of the four trials on rPerc found a larger probability of excellent outcome (mRS 0-1) with rPerc than without rPerc. These results should be interpreted with caution since outcome assessment in this trial was not blinded and the study was susceptible to allocation concealment bias. Studies investigating neuroprotective treatment in addition to EVT that report a significant effect of neuroprotection on clinical outcome are the trials in which patients were treated with 1) the combined treatment of IVT and EVT with uric acid, 2) the use of nerinetide in stroke patients treated with EVT, and 3) the use of high-flow NBO in stroke patients successfully treated with EVT. The article on uric acid reported results from a predefined subgroup analysis with patients treated with IVT and EVT. All three studies were methodologically sound with a low risk of bias. A new trial investigating nerinetide (ESCPAPE-NEXT) is already recruiting patients to confirm these results (NCT04462536). A follow-up trial for uric acid has been planned as well (11).
By limiting our search to RCTs with the terms neuroprotection (and variants) and stroke, it is possible that some trials on neuroprotective agents and reperfusion therapy in AIS were not found. Indeed, by cross-checking references, two more articles were added to our results. A large number of neuroprotective agents has been investigated in AIS in the past. It is possible that a number of articles fell outside our search, despite the use of reperfusion therapy, simply because no data on the use of reperfusion therapy were reported. As we limited our search strategy to neuroprotection trials using efficacy as outcome, safety trials may have fallen outside the scope of our search. Different studies reported various outcome measures. Our calculated ORs for good clinical outcome (mRS 0-2) did not reach statistical significance in any of the reported trials. The heterogeneity in outcome, as well as variations in treatment administration (e.g. timing, dosage), and the lack of replication-studies impaired the development of a quantitative synthesis. Future studies should aim to report standardized functional outcome measures, most notably the mRS, as was proposed by the Stroke Therapy Academic Industry Roundtable (STAIR) (46), as well as strive to replicate and validate promising findings.
Recanalization rates in stroke patients treated with EVT are higher (60-90%) (4) than in patients treated with IVT (10-30%) (47,48). Neuroprotective treatment may be more effective in these patients. As EVT is a relatively new treatment, a very limited number of neuroprotective agents has been tested in this setting. Future research should seek to combine EVT with neuroprotective strategies to test the hypothesis that neuroprotective treatment in combination with successful reperfusion leads to better outcomes in patients with AIS. Rapid reperfusion can, however, also lead to reperfusion injury, induced by high levels of free oxygen radicals, when oxygen becomes again available in previous hypoxic-ischemic tissue. For that reason treatments, which target reduction in oxidative stress might be beneficial in the prevention of this reperfusion injury.
An overview of ongoing studies investigating neuroprotective agents in addition to EVT using efficacy outcome measures is shown in Table 4. Future pre-clinical research is needed to improve the targeting of neuroprotective agents as well as their action mechanism. STAIR X proposed the term neurovascular unit (NVU) to better recognize different parts of the brain susceptible to ischemia. The NVU consists of various cell types including astrocytes, neurons and endothelial cells with a differential vulnerability to ischemic injury. New research strategies may focus on neuroprotective strategies targeting one or multiple parts of the NVU (13). Indeed, the importance of solid pre-clinical research in neuroprotective strategies has already been recognized by the initiation of the Stroke Preclinical Assessment Network (SPAN), a project started by the National Institutes of Health (NIH) in 2018 in which 7 different laboratories seek to employ rigorous, clinical research practices while assessing the effectiveness of potential neuroprotective therapies in reducing ischemic brain injury. New pre-clinical research strategies may focus on neuroprotective strategies targeting one or multiple parts of the NVU.

CONCLUSION
Studies specifically investigating neuroprotection in addition to reperfusion treatment are scarce. Five of 15 studies report a positive treatment effect of neuroprotection in addition to reperfusion therapy. Positive results are reported for combining EVT and IVT with uric acid, nerinetide in addition to EVT without IVT, rPerc with IVT, and the use of imatinib and normobaric oxygen in patients treated with EVT and/or IVT, although various clinical outcome measures were reported and validation of these results needs to be established. Future neuroprotection studies should combine neuroprotective agents with reperfusion therapies in AIS or include prespecified subgroup analyses for treatment with IVT and/or EVT.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding author.