Introduction
The Stroke Prevention by Aggressive Reduction in Cholesterol Levels (SPARCL) trial showed a significant risk reduction in cerebro-cardiovascular events with the use of intense-dose atorvastatin (80 mg/day) [1]. In particular, post-hoc analysis revealed that the relative risk reduction for those who achieved low-density lipoprotein cholesterol (LDL-C) levels below 70 mg/dL was 28%, which is greater than the overall result of 16%. Inspired by these results, the risk of cerebro-cardiovascular events between a low LDL-C target (<70 mg/dL) group (LT) and a high LDL-C target (100±10 mg/dL) group (HT) was compared in the Treat to Stroke Target (TST) trial [2]. A total of 2,860 patients from France and South Korea who had stroke and evidence of atherosclerosis were recruited. The results of the main TST trial showed a 22% risk reduction in the LT group compared to the HT group. A separate analysis of a French cohort showed similar results [3].
However, the optimal target of LDL-C in Asian patients has been well debated [4]. Because Asian stroke patients have distal cerebral artery pathology (e.g., intracranial atherosclerosis [ICAS] and small-vessel occlusion [SVO]) more often than their white counterparts [5], the results may differ from the French data. Indeed, Japan Statin Treatment Against Recurrent Stroke (J-STARS) data reported that pravastatin was effective in the secondary prevention of stroke in patients with atherosclerotic stroke but not those with lacunar infarction [4]. The present study aimed to determine whether the benefit of LT over HT is reproduced in the Korean stroke patients.
Methods
Trial design
Details of the TST have been described in a previous study [2]. The local institutional review board at each center approved the study protocol. Written informed consent was obtained from all the patients. This study was funded by Pfizer Inc. in 2020, and Pfizer Upjohn merged with Mylan to form Viatris. However, there was no industry involvement in the conduction, data gathering, or analysis of the trials.
Study population
All patients were >20 years old. For inclusion, the patient must have had an ischemic stroke within the previous 3 months and a modified Rankin Scale (mRS) score of 0 to 3, or a transient ischemic attack (TIA) within the previous 15 days that included limb weakness or speech disturbance lasting more than 10 minutes.
The patients were routinely evaluated using brain computed tomography (CT) or magnetic resonance (MR) imaging. The vascular status was assessed using brain CT, MR, or conventional angiographies. In addition, transesophageal echocardiography or CT angiography of the aorta was performed to detect aortic atheroma, when necessary. The imaging modality and diagnosis of atherosclerotic stenosis were confirmed by investigators at each center, who were experienced stroke neurologists practicing in tertiary stroke centers in Korea. To fulfill the enrollment criteria, atherosclerotic disease must have been present, including stenosis/occlusion of an extra- or intracranial cerebral artery, ipsilateral or contralateral to the side of brain ischemia, aortic arch atherosclerotic plaques ≥4 mm in thickness, or a previous history of coronary artery disease. If a patient had been taking an HMG-CoA reductase inhibitor (“statin”) before randomization, the LDL-C must have been at least 70 mg/dL (1.8 mmol/L), or at least 100 mg/dL (2.4 mmol/L) in the patient without prior statin use.
Randomization and follow-up
Eligible patients were randomly assigned in a 1:1 ratio to a target LDL-C of <70 mg/dL or a target of 100±10 mg/dL. Data on baseline demographics, body mass index, entry event, time since entry event, previous mRS score, vascular risk factors, and laboratory findings were collected. The statin type and dosage were not predefined. Investigators were asked to measure LDL-C levels 3 weeks after initiating treatment to adjust the statin dose and add other lipid-lowering agents, including ezetimibe, when necessary. Patients were followed up every 6 months to measure their LDL-C levels. In addition to face-to-face visits, phone calls were conducted with patients every 6 months to acquire LDL-C levels from the preceding visit and to evaluate potential trial endpoints using a structured questionnaire. If a trial outcome was discovered, the local investigator was contacted to confirm the event. The investigators were advised to maintain the blood pressure of the patient at ≤130/80 mm Hg in those with diabetes, and at ≤140/90 mm Hg in all others, to maintain a glycated hemoglobin level of less than 7% in those with diabetes, and to encourage the cessation of smoking.
Primary and secondary endpoints
The primary endpoints were: (1) a composite of nonfatal cerebral infarction or stroke of undetermined origin; (2) nonfatal acute coronary syndrome; unstable angina requiring urgent coronary artery revascularization; (3) neurological symptoms requiring urgent cerebral or carotid artery revascularization; and (4) cardiovascular death, including unexplained sudden death. The secondary endpoints were as follows: (1) cardiovascular events alone (nonfatal acute coronary syndrome, unstable angina requiring urgent coronary artery revascularization, and cardiovascular death that excluded unexplained sudden death); (2) ischemic stroke alone; (3) ischemic stroke including TIA; and (4) any stroke (ischemic stroke, TIA, or hemorrhagic stroke).
Ischemic strokes were further analyzed according to the Trial of Org 10172 in Acute Stroke Treatment (TOAST) classification and the location of atherosclerosis: extracranial atherosclerosis (ECAS) versus ICAS versus ECAS+ICAS [5]. In addition, among the patients with cerebral atherosclerosis, we categorized the patients into those with symptomatic ICAS (ischemic stroke on the side ipsilateral to the ICAS) and those with symptomatic ECAS (ischemic stroke on the side ipsilateral to the ECAS), and compared the secondary endpoints between these two groups. All clinical events were assessed by experienced neurologists blinded to the patient groups. Moreover, as adverse events, we investigated the occurrence of cerebral hemorrhage and newly diagnosed diabetes during the follow-up period.
Statistical analysis
Sample size estimation was previously performed, assuming that the enrollment of 3,786 patients would result in 385 primary endpoints, providing approximately 80% power to detect a 25% lower relative risk of major cerebro-cardiovascular events in the LT group than in the HT group. A total of 712 Korean patients were enrolled in this study. First, we compared baseline demographics, vascular risk factors, clinical variables, and laboratory findings between the LT and HT groups. The significance of differences was assessed using Student’s t-test, Mann-Whitney U test, and the chi-square test, as appropriate. The cumulative incidence of primary and secondary endpoints was evaluated using the Kaplan-Meier method and the log-rank test. Primary efficacy analysis was conducted using the Cox proportional hazards regression model. Age, sex, entry event (stroke or TIA), and time since entry were used as covariates.
In addition, we performed a subgroup analysis according to the TOAST classification (large-artery atherosclerosis [LAA] vs. SVO vs. other etiologies) and the location of atherosclerosis (ICAS vs. ECAS vs. ICAS+ECAS, and symptomatic ICAS vs. symptomatic ECAS). Moreover, we performed interaction analyses to evaluate the relationship between the location of symptomatic atherosclerosis (symptomatic ICAS vs. symptomatic ECAS) and LDL-C target (LT group vs. HT group) with respect to the secondary endpoints. Statistical significance was set at P<0.05. All data were analyzed using the R software (version 4.2.3; R Foundation for Statistical Computing, Vienna, Austria).
Results
Between May 2015 and February 2021, 712 patients from 14 centers in South Korea were recruited: 357 in the LT group and 355 in the HT group. Patients were followed up until January 31, 2023. The median follow-up period and interquartile range were 3.0 (0.6-5.8) years for the LT group and 2.8 (0.6-5.7) years for the HT group.
The baseline patient characteristics are shown in Table 1. Age, sex, and medical comorbidities did not differ between the two groups. The baseline mean LDL-C level was 128.7±39.6 mg/dL for the LT group and 130.1±37.8 mg/dL for the HT group. No significant differences were observed in the proportions of statin-naive patients, nor in terms of blood pressure or the laboratory findings.
Endpoints
At a median follow-up of 2.9 years, the mean LDL-C level was 71.0 mg/dL in the LT group and 86.1 mg/dL in the HT group (P<0.001) (Figure 1). Table 2 presents the primary and secondary endpoints of this study. The primary endpoint occurred in 24 (6.7%) patients in the LT group and 31 (8.7%) in the HT group (unadjusted hazard ratio [HR]=0.77, 95% confidence interval [CI]=0.45-1.32, P=0.333). After adjustment for covariates, there were no significant differences in the occurrence of primary endpoints between the two groups (adjusted HR 0.78, 95% CI 0.45-1.33, P=0.353) (Figure 2). Among the primary endpoints, the most frequent events were cerebral infarction or stroke of undetermined origin (18 [5.0%] vs. 15 [4.2%]), followed by urgent coronary revascularization (3 [0.8%] vs. 11 [3.1%]). In the secondary endpoint analysis, the risk of cardiovascular events alone was significantly higher in the HT group (HR 0.26, 95% CI 0.09-0.80, P=0.019) compared with the LT group. However, no significant differences were observed between the two groups in the occurrence of ischemic stroke alone (HR 1.26, 95% CI 0.64-2.49, P=0.503), ischemic stroke including TIA (HR 1.12, 95% CI 0.60-2.10, P=0.712), and any stroke (HR 1.35, 95% CI 0.75-2.42, P=0.321) (Figure 3).
Stroke recurrence according to the stroke mechanism and the location of atherosclerosis
The distribution of initial stroke mechanisms is shown in Supplementary Figure 1. Subgroup analyses based on the initial stroke mechanism (LAA, SVO, and other etiologies) revealed no significant differences between the LT and HT groups in the occurrence of ischemic stroke alone, ischemic stroke including TIA, or any stroke (Table 3). Similarly, no significant differences in the occurrence of secondary endpoints were found between the groups according to the location of atherosclerosis (ICAS vs. ECAS vs. ICAS+ ECAS) (Table 4) or the location of symptomatic cerebral atherosclerosis (symptomatic ICAS vs. symptomatic ECAS) (Table 5). However, there was a significant interaction between the location of symptomatic cerebral atherosclerosis and the LDL-C target concerning the secondary endpoints (P=0.0284 for interaction with ischemic stroke alone, P=0.0624 for interaction with ischemic stroke including TIA, and P=0.0324 for interaction with any stroke) (Figure 4).
Adverse events
Cerebral hemorrhages occurred in five patients (1.4%) in the LT group and one patient (0.3%) in the HT group (HR 5.00 [95% CI 0.58-50.00], P=0.142). Newly diagnosed diabetes occurred in nine patients (2.5% vs. 2.5%) in each group, showing no significant differences (HR 1.01 [95% CI 0.40-2.56], P=0.979).
Discussion
The main TST trial [2] and subsequent analysis of French data [3] showed a significant risk reduction for major cerebro-cardiovascular events in patients with stroke and TIA in the LT group compared with the HT group. However, in Korean patients, the incidence of primary outcomes was similar between groups. The neutral result may in part be attributed to the relatively narrow gap in the mean LDL-C level between the LT and HT groups: 71.0 mg/dL versus 86.1 mg/dL, respectively, as compared to 66 mg/dL versus 96 mg/dL, respectively, in French patients [2]. In addition, the follow-up period for Korean patients (median, 2.9 years) was shorter than that of French patients (median, 5.3 years), which was attributed to the later involvement of the TST trial in Korean versus French centers.2 Nevertheless, a detailed analysis of our data suggests that this difference may be real.
In the present Korean data, cardiovascular events alone (coronary artery disease and urgent coronary artery revascularization) were significantly lower in the LT group compared to the HT group (0.26 [0.09-0.80], P=0.019). Similarly, in the French data, a strong trend was observed for the cardiovascular events to occur less frequently in the LT group than in the HT group, although the difference was not statistically significant (0.66 [0.37-1.20], P=0.180). These results suggest that a lower LDL-C target level is better than a higher target level, regardless of ethnicity, at least for the prevention of coronary artery disease.
In contrast, there was no significant benefit in the prevention of cerebrovascular events (cerebral infarction and cerebral artery revascularization) in the LT group (Table 2). This was in contrast to the results of the French study, in which a significantly lower incidence of new cerebral infarction was observed in the LT group (6.7%) than in the HT group (9.1%) [3].
We suspect that these differences may be attributed to the differences in stroke subtypes [6]. In the Asian population, ICAS and SVO are more prevalent than in Western counterparts [5]. Although we enrolled stroke patients with evidence of atherosclerosis, this does not necessarily mean “stroke caused by atherosclerosis”; patients with SVO were enrolled if there was evidence of atherosclerosis in other organs (e.g., coronary disease). Thus, we categorized the enrolled patients as having LAA (ICAS, ECAS, or both) or SVO.
Comparison of the LT and HT groups demonstrated no significant difference in the incidence of recurrent stroke in patients with either LAA or SVO. However, the HR was higher in patients with SVO (approximately 3.0) compared of those with LAA (approximately 0.9). Although the data did not reach statistical significance, most likely due to the small number of endpoints, our results suggest that the efficacy of statins may differ according to stroke subtype.
In both patients with ICAS and ECAS, no significant differences were observed between the LT and HT groups in the prevention of recurrent stroke regarding the location of atherosclerosis. However, the HR appeared to be smaller in the ECAS group (approximately 0.5) than in the ICAS group (approximately 1.2), suggesting that LT may be more effective than HT in preventing stroke in patients with ECAS than in those with ICAS. Further analysis showed a significant interaction in terms of stroke recurrence between the location of symptomatic cerebral atherosclerosis (ICAS vs. ECAS) and the LDL-C target (Table 5). Although the numbers of patients and outcome events were too small to make a reliable conclusion, our data seem to be consistent with the SPARCL data [7]. Although the benefit of preventing the primary outcome (overall odds ratio) was 16% in this study, there was a greater (33%) reduction in the primary outcome in patients who had carotid artery disease [7], suggesting that the benefit of high-dose statins may be much less in stroke patients without carotid stenosis [7].
Thus, the results indicate that the principle “the lower the LDL-C level, the better the clinical outcome” is likely to be true for patients with ECAS but not for those with more distal cerebral artery pathologies, such as SVO or ICAS [6]. It has been shown that statin therapy is associated with increased occurrence of intracerebral hemorrhage [8], which is also a manifestation of SVO. Studies have shown that LDL-C level is less closely associated with reducing recurrent SVO than atherothrombotic infarction [9,10]. A prospective, long-term study from Japan found a significant association between LDL-C level and the risk for atherothrombotic infarction, whereas such a relationship was not evident for SVO [11]. Regarding ICAS, although a study showed that high-dose atorvastatin improved high-resolution MR findings of intracranial pathologies in patients with ICAS, this was a non-controlled study, and the optimal target for LDL-C level in these patients remains unknown [12]. In addition, previous studies showed that hyperlipidemia [13] or increased LDL-C level [14] was more closely associated with ECAS than with ICAS. These pieces of evidence may explain why the stroke recurrence rate in the LT group was not lower than in the HT group in the population who frequently have ICAS and SVO than ECAS.
These results should be interpreted with caution since patients with SVO have atherosclerosis in other arteries. Therefore, the actual benefit of LT may have been even lower if we had enrolled patients with SVO without any evidence of atherosclerosis. Moreover, because no control subjects were included in the TST trial, our findings cannot guide the use of statins or determine the optimal target of LDL-C in patients with ICAS or SVO.
Conclusions
In summary, the results presented in this study are consistent with those in the literature [6] demonstrating that LT is better than HT for the prevention of coronary diseases in patients with ischemic stroke. However, unlike the French TST results, the Korean TST data failed to show a benefit of LT over HT in the prevention of cerebrovascular events. This is most likely related to the different stroke subtypes in different ethnicities. Given the smaller number of Korean patients compared to French patients, studies with a larger number of patients are warranted. Further research is needed to determine whether statins are effective and, if so, what would be an appropriate LDL-C target in Asian patients who more often have ICAS and SVO compared to Caucasians.
PDF Links
PubReader
ePub Link
XML Download
Full text via DOI
Download Citation
Supplement
Print
