Platelet‐activating factor antagonist‐based intensive antiplatelet strategy in acute ischemic stroke: A propensity score matched with network pharmacology analysis

Abstract Background Diterpene ginkgolides meglumine injection (DGMI) is a platelet‐activating factor receptor (PAFR) antagonist that can be used to treat acute ischemic stroke (AIS). This study evaluated the efficacy and safety of an intensive antiplatelet strategy based on PAFR antagonists and explored the underlying mechanisms of PAFR antagonists in AIS treatment. Methods This is a retrospective study applying propensity score methods to match AIS patients treated with DGMI to nontreated patients. The primary outcome was functional independence (modified Rankin Scale [mRS] 0–2) at 90 days. The safety outcome was bleeding risk. We used McNemar test to compare the efficacy outcome. Subsequently, the network pharmacology analysis was performed. Results 161 AIS patients treated with DGMI in the study were matched with 161 untreated patients. Compared with untreated patients, DGMI‐treated patients had a significantly higher rate of mRS ranking 0–2 at 90 days (82.0% vs. 75.8%, p < 0.001), without increased risk of bleeding. The gene enrichment analysis showed that the overlap genes of DGMI targeted and AIS‐related enriched in thrombosis and inflammatory‐related signaling pathways. Conclusions An intensive antiplatelet strategy of DGMI plus traditional antiplatelet agents is effective in treating AIS and may work by mediating post‐stroke inflammation and thrombosis.


| INTRODUC TI ON
Acute ischemic stroke (AIS) sufferers are at high risk of subsequent disability and mortality, particularly when it is caused by arterial occlusion. 1Intravenous thrombolysis and endovascular thrombectomy are the most recommended management for the acute phase of AIS but are time-critical. 2However, a proportion of AIS patients still did not achieve complete recanalization, 3 and approximately 40%-50% of successful reperfusion patients retained unfavorable outcomes. 4,5The differences in treatment outcomes in patients with AIS may be attributed in part to the ongoing thromboembolic formation and early arterial re-occlusion by platelet activation and subsequent platelet aggregation, irrespective of the patient receiving revascularization therapy. 6,7Therefore, antiplatelet agents remain the cornerstone among the early treatments of AIS.][10][11][12] Given the balance between the benefits of intensive antiplatelet therapy and an increased risk of bleeding, the previous guidelines only recommend the short-term use of combined antiplatelet therapy with aspirin and clopidogrel following a minor ischemic stroke or a high-risk transient ischemic attack (TIA). 13Therefore, exploring safe and effective antithrombotic therapies against early platelet aggregation in AIS has received great concern.
5][16] Diterpene ginkgolides meglumine injection (DGMI), a novel PAF antagonist, is an extract from Ginkgo leaves, comprising mainly of ginkgolide A, B, and K (≥98%).It has been previously reported to be the highly selective antagonists of the PAF receptor (PAFR). 17In terms of the mechanism of drug-targeted action, DGMI may be a promising antiplatelet therapy for AIS independent of Aspirin and P2Y12 receptor inhibitors.
A multicenter, randomized controlled clinical trial (RCT) RCT has shown ginkgolide plus aspirin to be more effective than aspirin monotherapy for AIS patients with intracranial artery stenosis (ICAS). 18However, an intensive antiplatelet strategy combining PAF antagonists with guideline-recommended antiplatelet regimens has not been validated in real-world AIS study.Therefore, to investigate the efficacy and safety of an intensive antiplatelet regimen based on PAFR antagonists in AIS, we compared the different antiplatelet routing: DGMI plus conventional antiplatelet strategy versus conventional antiplatelet strategy for AIS patients without vessel reperfusion treatment.Furthermore, the underlying antithrombosis and anti-inflammatory mechanisms of DGMI as a PAFR antagonist in AIS treatment were also explored.

| Study populations selection
This was a retrospective study in compliance with STROBE criteria, 19 which involved consecutive patients with AIS hospitalized in the The patient's written informed consent was waived because all data analyzed in this study were anonymized.

| Clinical data
Baseline demographics, medical history (diabetes mellitus, hypertension, hyperlipidemia, ischemic stroke, and hyperlipidemia), smoking status, National Institute of Health stroke scale (NIHSS) score at admission, modified Rankin Scale (mRS) prior to onset, and stroke subtypes were included.The routine blood examinations were conducted on admission within the first 24 h.Stroke etiology was categorized following TOAST classification. 20 employed to assess the severity of stroke. 22Patient comorbidities, stroke severity, treatment information, safety, and efficacy outcomes were recorded as well.

| Antiplatelet intervention
The administration of DGMI was at the discretion of the attending neurologist.According to the antiplatelet regimen administered in the 7 days after admission, the participants were assigned to two groups: DGMI plus conventional antiplatelet-treated group and conventional antiplatelet-treated group.In general, the DGMI treatment was treated immediately after admission as follows: 25 mg of DGMI (Jiangsu Kanion Pharmaceutical Co., Ltd.) diluted in 250 mL of 0.9% NaCl injection was intravenously injected daily for 7 days.The conventional antiplatelet regiment was defined as the aspirin (100 mg oral daily)/clopidogrel (75 mg oral daily) monotherapy or dual antiplatelet therapy (DAPT) recommended by current guidelines on an individual basis.In addition, all patients enrolled were treated individually according to the guidelines, 21 such as antihypertension, lipid and glucose-lowering, and neuroprotection for AIS.

| Efficacy and safety outcomes
The primary outcome was functional independence (90 days mRS 0-2).The secondary outcomes included freedom from disability (90-days mRS 0-1), and the incidence of early neurological deterioration (END).END was measured as an increase of 4 or more points in NIHSS score within 7 days after admission compared with baseline. 23The primary safety outcome was symptomatic intracranial hemorrhage (sICH), defined as computed tomography (CT) evidence of intracranial hemorrhage associated with a 4point increase in NIHSS score. 24Other safety outcomes included the presence of asymptomatic intracranial hemorrhage (aICH) without clinical worsening despite evidence of hemorrhagic transformation on the brain CT, 25 and the presence of any bleeding events.CT imaging obtained after treatment was in response to adverse patient clinical changes per the discretion of the treating team or patients who had an NIHSS ≥4-point worsening of NIHSS at 24 h.Clinical assessment evaluation at 90 days follow-up was performed by a neurologist or a trained specialist nurse who is certified in mRS assessment from the senior stroke center through face-to-face or telephone interviews.

| Assessment of platelet function
This study included platelet function testing results to assess the PAFR antagonist synergistic antiplatelet effect in AIS patients.(3) angle (α, degree), signifying the rate of clot formation; (4) maximum amplitude (MA), signifying the maximum strength of the clot; (5) MAADP, signifying the strength of the clot induced by adenosine diphosphate (ADP); (6) arachidonic acid (AA) inhibition rate (AA%), signifying the reaction to aspirin to aspirin; and (7) ADP inhibition rate (ADP%), signifying the reaction to clopidogrel.

| Identification of the potential mechanism of DGMI in AIS through bioinformatics and network pharmacology
To find the differentially expressed genes (DEGs) of AIS, we downloaded the gene expression profile from the Gene Expression Omnibus (GEO) (https://www.ncbi.nlm.nih.gov/geo/):63 blood samples (39 samples with AIS and 24 health controls) in GSE16561. 26 used the "limma" R package to normalize the expression data (Figure S1) and to identify the DEGs between AIS and the control blood samples with a threshold of |logFC| >0.5 and an adjusted p value <0.05 We input DGMI active compounds (Ginkgolides A, B, and K) into the PubChem Database (https://pubch em.ncbi.nlm.nih.gov) to achieve their smile numbers. 27The smiles numbers were imported into SwissTargetPrediction (http://www.swisstarge tpred iction.ch) to predict the drug-target, 28 and then they were imported into UniProt (http://www.uniprot.org/) to obtain the official gene symbol name, setting the organism "Homo sapiens". 29Using these web tools, we acquired predicted drug-target genes of each compound.
After merging the duplicate data, we chose the target genes as potential targets of DGMI.Additionally, we calculated and made Venn diagrams for overlap genes of AIS-DEGs and DGMI-target genes using the Sangerbox 3.0 tool (http://www.http://sange rbox.com/home.html). 30 further indicate the biological processes (BPs) and signaling pathways that involve DGMI in the treatment of AIS, the gene ontology (GO) and Reactome pathways enrichment analysis were performed by submitting overlapping genes of DGMI and AIS-DEGs, based on the R "clusterProfiler" package.The cut-off criteria for enrichment analysis were p < 0.05 and false discovery rate (FDR) < 0.05.

| Statistical analysis
Results are presented as mean ± standard deviation (SD) for normally distributed continuous variables, as median (interquartile range [IQR]) for non-normally distributed data, and as number (%) for categorical variables.The Shapiro-Wilk normality test was performed on all continuous variables before analysis.The missing data were imputed using the k-Nearest Neighbor (kNN) algorithm.The data used and analyzed during the current study are available from the corresponding author on reasonable request.

| RE SULTS
In total, 2586 AIS hospitalization records were screened for eligibility, and 482 admissions met the selection criteria and were eligible for the analysis (Figure 1).Patients prescribed DGMI (n = 280) were slightly younger, had a higher prevalence of smoking, stroke etiology of large-artery atherosclerosis (LAA), had a higher PT level and a lower Scr level at baseline, and were more likely to receive DAPT during hospitalization before PSM (Table 1).difference <0.1 (Table 1).The main cause of END in our study was stroke progression, symptom fluctuations, and brain edema.

| Clinical outcomes
When compared with the conventional antiplatelet group, the DGMI plus conventional antiplatelet group had a significantly higher proportion of mRS ranking 0-2 at 90 days (82.0% vs. 75.8%,p < 0.001), a significantly higher proportion of mRS ranking 0-1 at 90 days (69.6% vs. 62.1%, p < 0.001), and a lower incidence of END (10.6% vs. 12.4%, p < 0.001; Table 2, Figure 2).There was TA B L E 1 Baseline patient characteristics in patients treated with and without DGMI before and after Propensity Score Matching.No significant difference was observed between the two groups in levels of all TEG variables reflecting coagulation and platelet function after antiplatelet therapy (all p > 0.05, Table S1).

| Bioinformatics and network pharmacology analysis
As a result, a total of 669 AIS-related DEGs were screened out from  S2.
Reactome pathways enrichment analysis showed that the overlapping genes were primarily enriched in interleukin-4 and interleukin-13 signaling, neutrophil degranulation, biosynthesis of DHA-derived SPMs, signaling by interleukins (Figure 4, Table S3).

| DISCUSS ION
Our study has demonstrated the effectiveness of DGMI plus con- intracranial artery stenosis patients. 18The post hoc analyses of GISAA also confirmed the safety and efficacy of the Ginkgolidebased DAPT strategy in elderly patients. 16The results of the POINT and CHANCE trials have demonstrated the superiority of short-term DAPT to aspirin alone in minor ischemic stroke and TIA, Although previous studies suggested that PAFR antagonists could inhibit platelet aggregation in vivo and in vitro, [32][33][34] data available investigating the effects of PAFR antagonists on coagulation and platelet function in patients with AIS are scarce.The GISAA study found that the effect of Ginkgolide in AIS was associated with the PAF pathway but not with the ADP or thromboxane A2 (TxA2) pathways. 18Similarly, the current study found that neither AA- Experimental evidence suggested that PAFR antagonists may exert neuroprotective effects in ischemic stroke mainly through anti-apoptosis, 35 anti-inflammation, 36 turnover of harmful autophagy, 37 pro-angiogenic activity, 38 and attenuating neuronal pyroptosis. 39Our work integrating bioinformatics combined with network pharmacology approaches further offers new mechanistic insight into our observations.According to the functional enrichment results, DGMI's therapeutic effect was directed through mediating inflammatory-related signaling pathways, such as IL-4 and IL-13 signaling.Microglia/macrophages are highly plastic immune cells with different polarizations and can be either pro-inflammatory (M1 polarization) or anti-inflammatory (M2 polarization) when responding to specific microenvironmental signals after cerebral ischemia. 40IL-4 production in the central nervous system is essential for the induction of M2 polarization, with the subsequent expression of different molecules related to scavenger receptors and pro-angiogenic factors, such as mannose receptors, ectin-1, and arginase, which perform neuroprotective effect. 41,42IL-13 acts as an anti-inflammatory cytokine and plays a role in regulating the polarization of microglia/ macrophages. 43At early cerebral ischemia, microglia, and macrophages can activate T cells and produce IL-17. 44Previous studies have found that IL-17 exerts a pro-inflammatory cytokine effect by activating NF-κB/MAPK pathway signaling and a decrease in serum IL-17A in patients with good clinical outcomes after DGMI treatment. 32 addition, XQ-1H, a new derivative of ginkgolide B, was found to protect mice from ischemic injury by regulating microglia polarization via the PPARγ pathway. 45These findings suggested that DGMI reduced inflammation in AIS patients by modulating the balance of pro-and anti-inflammatory microglia polarization.
Growing evidence suggests that inflammation is strongly associated with thrombosis through VWF-mediated reactivity, which interacts with tissue factor-initiated fibrin formation and may lead to extensive thrombotic disease. 46 Zhaoqing First People's Hospital from June 2017 to May 2022.The inclusion criteria were as follows: (a) over the age of 18; (b) admitted to hospital within 48 h of onset of symptoms; (c) diagnosed with AIS based on clinical symptoms, brain magnetic resonance imaging (MRI) manifestations, as well as the criteria of the Trial of Org 10172 in Acute Stroke Treatment (TOAST) 20 ; (d) treated with the standard antiplatelet protocol (Aspirin/Clopidogrel monotherapy, or Aspirin plus Clopidogrel) recommended by the Chinese guideline 21 ; (e) having received platelet function testing after 7 days of antiplatelet treatment with a documented 3-month follow-up of onset.Patients were excluded in the presence of any of the following conditions: (a) had incomplete medical records affecting subsequent analysis; (b) had other medications within hospitalization affecting hematological coagulation, such as argatroban, tirofiban, cilostazol, heparin, warfarin, dabigatran, and factor Xa inhibitors; (c) had a history of malignancies, systemic inflammatory disorders, hematological disorders, severe liver, kidney, or heart dysfunction; (d) had been undergoing intravenous thrombolysis or endovascular thrombectomy after the admission.This study was approved by the Zhaoqing First People's Hospital Ethics Committee (No. of approval: B2021-11-02).
Platelet reactivity was assessed by the thromboelastography (TEG) platelet mapping (BVCA-VIII, Bring Biology) at 7 ± 2 days after the initiation of antiplatelet therapy.Seven TEG parameters were examined: (1) reaction time (R, min), signifying the time from clotting factor activation to initial clot formation; (2) coagulation time (K, min), signifying the time from clot formation to reach a 20 mm amplitude;

| 4085 HAN
et al.Patients were stratified into two groups depending on treatment with DGMI or not.Given the selection bias inherent to retrospective studies and the confusion bias, we used propensity score matching (PSM) applying a logistic regression model for balancing the distribution of the baseline characteristics and potential confounders between groups, including demographic data (age, sex), previous history (history of diabetes mellitus, hypertension, hyperlipidemia, ischemic stroke, and smoking status), baseline clinical characteristic (TOAST subtypes, carotid atherosclerosis, and blood white blood cell [WBC], serum creatinine [Scr], blood cholesterol [CHOL], glycated hemoglobin [HbA1c], prothrombin time [PT], fibrinogen [Fib], NIHSS at admission, mRS before admission, and antiplatelet treatment strategy).Using the R "MatchIt" package, patients were matched 1:1 using the nearest neighbor matching with a caliper of 0.05 on a logit scale.Figure S2 displays the distribution of the propensity score before and after matching in DGMI-treated group and un-treated group.There is adequate overlap between the two treatment groups.Characteristics of patients included in generating the propensity scores were compared before and after matching using a standardized differences. 31After matching, binary outcomes were compared using McNemar test, and TEG parameters were compared using paired t tests.Analyses were performed by R software (version 4.1.5,R Foundation), and a p < 0.05 was considered as statistical significance.

After 1 :
1 propensity score matching, a total 161 of patients were allocated in each group and the majority of baseline characteristics of confounders were well balanced with a standardized F I G U R E 1 Flow chart of admission record selection.AIS, acute ischemic stroke; DGMI, diterpene ginkgolides meglumine injection; PSM, propensity score matching; TEG, thromboelastography.
GSE16561 dataset.The result included 504 down-regulated and 165 upregulated genes and is shown in FigureS3A.A total of 152 target genes of each candidate DGMI compound were obtained and we defined 14 overlapping genes between the targets of DGMI and AIS-related DEGs (FigureS3B).To indicate the specific circumstances of the association of each active constituent of DGMI with the potential targets, a "Drug-Active Constituent-Potential Target" interactive network is displayed in FigureS3C.The overlapping genes represent targets for the combined application of DGMI for the treatment of AIS and were analyzed to obtain the gene functional enrichment analysis.The top 10 terms for biological process BP, cell composition (CC), molecular function (MF) enrichment of GO were listed in Figure 3.The overlapping genes were most significantly enriched in complement receptor-mediated signaling pathway, regulation of inflammatory response for BP, ficolin-1-rich granule for CC, complement receptor activity, oxidoreductase activity for MF.The top 20 terms enriched in GO enrichment analysis are listed in Table

F I G U R E 2 F I G U R E 3
Abbreviations: DGMI, diterpene ginkgolides meglumine injection; mRS, modified Rankin score.
nor ADP-mediated platelet reactivity tested by TEG differed significantly between the group treated with DGMI and that with not.It suggested that the therapeutic benefit of DGMI might not come from improving antiplatelet drug resistance to aspirin or clopidogrel.The underlying synergistic antithrombotic mechanisms of DGMI and conventional antiplatelet regimens remain unknown.The Reactome pathways enrichment analysis of AIS-DEGs and DGMI overlaps genes significantly enriched in the formation of fibrin clots and arachidonic acid metabolism signaling pathways, implying that these thrombosis-related signaling pathways could be the mechanisms of the DGMI-mediated AIS antithrombotic process.Based on the results of the current study, we speculated that DGMIbased intensive antiplatelet regimens do not increase bleeding risks since DGMI selectively impedes thrombus formation in the brain in AIS without affecting systematic coagulation or platelet function.

5 |
Functional and genetic evidence of DGMI against AIS indicated enrichment in multiple pathways mainly implicated in thrombosis and inflammation.Similar to VWF, PAF might be the link between inflammation and thrombosis, which makes it an ideal target for AIS therapy.The strengths of the study include the observational study design using the PSM method and in-depth mechanism of investigation that utilizes transcriptomics with network pharmacology.However, several limitations are noteworthy.First of all, this study is a retrospective observational study that enrolled participants from a Chinese monocenter, which could result in selection bias.Second, because the antiplatelet regimen administration route was based on physicians' decision, the potential for confounding associated with clinical outcomes might not be balanced.However, the inclusion of variables such as NIHSS at admission and stroke etiology could improve this limitation, and statistical models can take the effect of latent variables into account in statistical analyses.Furthermore, although we evaluated platelet function after DGMI treatment, we did not have the dynamic measurement of PAF, the influence of DGMI plus conventional antiplatelet agents on PAF-dependent platelet aggregation after AIS was not available in the present study.In addition, Figure2showed that when compared with the conventional antiplatelet group, the DGMI plus conventional antiplatelet group had a higher proportion of mRS ranking 5-6 at 90 days (6 [3.7%] vs. 2 [1.2%]).Even though hemorrhagic transformation was not the reason for a higher proportion of mRS 5-6 in the DGMI group, the mechanism of DGMI-based antiplatelet therapy could bring significant beneficial effects (improving functional outcomes at 3 months) to AIS patients, but without reducing the rate of mRS 5-6 at 3 months when compared with non-DGMI group remains unknown.Further studies focus on improving the above deficiencies and the interplay mechanism of PAF-mediated inflammation and thrombosis in AIS pathogenesis are warranted.CON CLUS IONWe found that DGMI plus conventional antiplatelet agents were a promising intensive antiplatelet strategy in AIS without any increase in bleeding events compared with patients who received conventional antiplatelet therapy.The analysis integrating transcriptomic and pharmacology identified that DGMI exerts a therapeutic effect in patients with AIS via inflammatory and thrombosis-related pathways.Since ischemic stroke has multiple and different underlying etiologies, PAFR-antagonist-based intensive antiplatelet strategy might shed a new light on the prevention and treatment of AIS.
NIHSS score was One of the major limitations of intensive therapy in AIS is that the increased risk of bleeding events might offset the efficacy of the antithrombotic treatments.Similar to previous studies, our study has established the safety of PAFR antagonists combined with other antiplatelet drugs in AIS, even in patients with more severe stroke or with DAPT.
8,10but it is unclear whether an intensive antiplatelet strategy (DAPT or triple