Research JAMA Neurology | Original Investigation Efficacy and Safety of Ticagrelor and Aspirin in Patients With Moderate Ischemic Stroke An Exploratory Analysis of the THALES Randomized Clinical Trial Yongjun Wang, MD; Yuesong Pan, PhD; Hao Li, PhD; Pierre Amarenco, MD; Hans Denison, MD, PhD; Scott R. Evans, PhD; Anders Himmelmann, MD, PhD; Stefan James, MD, PhD; Filip Birve, BSc; Per Ladenvall, MD, PhD; Carlos A. Molina, MD, PhD; S. Claiborne Johnston, MD, PhD; for the THALES Steering Committee and Investigators Supplemental content IMPORTANCE Prior trials of dual antiplatelet therapy excluded patients with moderate ischemic stroke. These patients were included in the Acute Stroke or Transient Ischaemic Attack Treated With Ticagrelor and ASA for Prevention of Stroke and Death (THALES) trial, but results have not been reported separately, raising concerns about safety and efficacy in this subgroup. OBJECTIVE To evaluate the efficacy and safety of ticagrelor plus aspirin in patients with moderate ischemic stroke (National Institutes of Health Stroke Scale [NIHSS] score of 4 to 5). DESIGN, SETTING, AND PARTICIPANTS The THALES trial was a randomized trial conducted at 414 hospitals in 28 countries in January 2018 and December 2019. This exploratory analysis compared patients with moderate stroke (baseline NIHSS score of 4 to 5) with patients with less severe stroke (NIHSS score of 0 to 3). A total of 9983 patients with stroke were included in the present analysis, after excluding 2 patients with NIHSS scores greater than 5 and 1031 patients with transient ischemic attack. Data were analyzed from March to April 2021. INTERVENTIONS Ticagrelor (180-mg loading dose on day 1 followed by 90 mg twice daily on days 2 to 30) or placebo within 24 hours after symptom onset. All patients received aspirin, 300 to 325 mg, on day 1 followed by aspirin, 75 to 100 mg, daily on days 2 to 30. Patients were observed for 30 additional days. MAIN OUTCOMES AND MEASURES The primary outcome was time to stroke or death within 30 days. The primary safety outcome was time to severe bleeding. RESULTS In total, 3312 patients presented with moderate stroke and 6671 presented with less severe stroke. Of those in the moderate stroke group, 1293 (39.0%) were female, and the mean (SD) age was 64.5 (10.8) years; of those in the less severe stroke group, 2518 (37.7%) were female, and the mean (SD) age was 64.8 (11.2) years. The observed primary outcome event rate in patients with moderate stroke was 7.6% (129 of 1671) for those in the ticagrelor group and 9.1% (150 of 1641) for those in the placebo group (hazard ratio, 0.84; 95% CI, 0.66-1.06); the primary outcome event rate in patients with less severe stroke was 4.7% (158 of 3359) for those in the ticagrelor group and 5.7% (190 of 3312) for those in the placebo group (hazard ratio, 0.82; 95% CI, 0.66-1.01) (P for interaction = .88). Severe bleeding occurred in 8 patients (0.5%) in the ticagrelor group and in 4 patients (0.2%) in the placebo Author Affiliations: Author group in those with moderate stroke compared with 16 patients (0.5%) and 3 patients (0.1%), affiliations are listed at the end of this respectively, with less severe stroke (P for interaction = .26). article. Group Information: The THALES CONCLUSIONS AND RELEVANCE In this study, patients with a moderate ischemic stroke had Steering Committee and consistent benefit from ticagrelor plus aspirin vs aspirin alone compared with patients with Investigators appear at the end of the less severe ischemic stroke, with no further increase in the risk of intracranial bleeding or article. other severe bleeding events. Corresponding Authors: S. Claiborne Johnston, MD, PhD, Dean’s TRIAL REGISTRATION ClinicalTrials.gov Identifier: NCT03354429 Office, Dell Medical School, University of Texas at Austin, 1912 Speedway, Ste 564, Austin, TX 78712 (clay.johnston@utexas.edu); Yongjun Wang, MD, Department of Neurology, Beijing Tiantan Hospital, Capital Medical University, No. 119, South 4th JAMA Neurol. 2021;78(9):1091-1098. doi:10.1001/jamaneurol.2021.2440 Ring West Rd, Beijing, China, 100070 Published online July 9, 2021. (yongjunwang@ncrcnd.org.cn). (Reprinted) 1091 Downloaded From: https://jamanetwork.com/ on 03/22/2023 Research Original Investigation Efficacy and Safety of Ticagrelor and Aspirin in Patients With Moderate Ischemic Stroke D ual antiplatelet therapy (DAPT) is a pivotal strategy for early management of minor noncardioembolic ische- Key Points mic stroke and transient ischemic attack (TIA).1-4 The Question Is dual antiplatelet therapy with ticagrelor plus aspirin efficacy and safety of DAPT with clopidogrel and aspirin have of benefit for patients with moderate acute ischemic stroke been evaluated in patient populations with TIA or acute is- (National Institutes of Health Stroke Scale [NIHSS] score of 4 to 5)? chemic stroke with baseline National Institutes of Health Stroke Findings In this exploratory post-hoc analysis of the THALES trial Scale (NIHSS) scores of 3 or less (range, 0-42; higher scores including 9983 patients with moderate or less severe stroke indicate more severe stroke) in the Clopidogrel in High-Risk (NIHSS score of 0 to 3), treatment with ticagrelor plus aspirin Patients With Acute Non-Disabling Cerebrovascular Events showed similar efficacy and safety vs aspirin alone among patients (CHANCE) trial5 and Platelet-Oriented Inhibition in New TIA presenting with moderate acute ischemic stroke and those and Minor Ischemic Stroke (POINT) trial.6 Based on these stud- presenting with less severe ischemic cerebrovascular events. ies, the current guidelines recommend early, short (21 days) Meaning Patients with moderate ischemic stroke may benefit combination treatment with clopidogrel and aspirin for pa- from dual antiplatelet therapy with ticagrelor and aspirin. tients with high-risk TIA and minor noncardioembolic ische- mic stroke with an NIHSS score of 3 or less.1,2,4 In the Acute Stroke or Transient Ischaemic Attack Treated Written informed consent was obtained from all the partici- With Ticagrelor and ASA for Prevention of Stroke and Death pants or their representatives before enrollment. The trial pro- (THALES) trial,7 DAPT with ticagrelor and aspirin was evalu- tocol can be found in Supplement 1. ated in patients with TIA or acute ischemic stroke with NIHSS Patients eligible for inclusion in the THALES trial were 40 scores of 5 or less, within which approximately 30% of the pa- years or older, had an acute noncardioembolic ischemic stroke tient population had NIHSS scores of 4 or 5 at randomization. with an NIHSS score of 5 or less or a high-risk TIA (ABCD2 stroke The THALES trial demonstrated that compared with aspirin risk score [scores assessing the risk of stroke on the basis of alone, DAPT with ticagrelor and aspirin reduced the risk of age, blood pressure, clinical features, duration of TIA, and pres- stroke or death within 30 days. Few severe bleeding events ence or absence of diabetes; range, 0 to 7] of 6 or greater or were observed; however, these were more common in the ti- symptomatic intracranial or extracranial arterial stenosis [50% cagrelor plus aspirin group.7 Patients with stroke with base- or greater narrowing in the diameter of the lumen of an artery line NIHSS scores of 4 to 5 could potentially respond differ- that could account for the TIA]), and could undergo random- ently than those with an NIHSS score of 0 to 3 to DAPT, owing ization within 24 hours after symptom onset. Participants’ race to more severe neurological impairment and larger areas of is- was classified by investigators, and options for race were pre- chemic injury, eg, they may have a greater risk of brain hem- specified. Participant race was included as part of an initia- orrhage. To our knowledge, no randomized trials have previ- tive to ensure that a variety of nationalities, ethnicities and ously evaluated the efficacy and safety of DAPT in patients with races were represented, that the results of the study would be moderate stroke (NIHSS score of 4 to 5), and whether the ben- widely applicable, and to allow for subgroup analysis of pa- efit of DAPT can extend to patients with ischemic stroke with tients by race in a Cox proportional hazards model. Detailed a severity of NIHSS score 4 to 5 has rendered interest from information on inclusion and exclusion criteria is provided in stroke neurologists. In this exploratory subgroup analysis of the study protocol.7,8 the THALES trial, we aimed to characterize the efficacy and An exploratory analysis of data from patients with stroke safety of DAPT with ticagrelor plus aspirin in the first 30 days with a baseline NIHSS score of 4 to 5 who participated in the in patients with an acute ischemic stroke with a baseline NIHSS THALES trial was performed and reported in this article. This score of 4 or 5. analysis was performed to compare patients with moderate stroke (baseline NIHSS score of 4 to 5) with those with less se- vere stroke (baseline NIHSS score of 0 to 3) among patients with ischemic stroke as a qualifying event as the main analysis and Methods among the entire population regardless of qualifying event Overview of the THALES Trial (ischemic stroke or TIA) as a sensitivity analysis. Details on the rationale, design, and main results of the THALES trial have been published previously.7,8 The THALES trial was Outcomes a multicenter, double-blind, placebo-controlled, parallel- The primary efficacy outcome was time from randomization group randomized clinical trial, which was sponsored by to the first occurrence of new stroke (ischemic or hemor- AstraZeneca and conducted at 414 sites in 28 countries be- rhagic) or death within 30 days. Stroke events included both tween January 22, 2018, and December 13, 2019. The objec- progression of index stroke (defined by rapid worsening of an tive of the THALES trial was to compare the efficacy and safety existing focal neurological deficit attributable to a new infarc- of ticagrelor (180-mg loading dose on day 1 followed by 90 mg tion or extension of a previous infarction in the same vascu- twice daily on days 2 to 30) with placebo in patients with acute lar bed) or new stroke events (including ischemic, hemor- minor noncardioembolic ischemic stroke or TIA. All patients rhagic, and undetermined strokes). Death included all causes took aspirin (300 to 325 mg on day 1 followed by 75 to 100 mg of death. The secondary outcomes included time to first sub- daily on days 2 to 30) in the first 30 days. The trial was ap- sequent ischemic stroke within 30 days and overall disability proved by the ethics committee at each participating site. measured as a modified Rankin Scale (mRS) score greater than 1092 JAMA Neurology September 2021 Volume 78, Number 9 (Reprinted) jamaneurology.com Downloaded From: https://jamanetwork.com/ on 03/22/2023 Efficacy and Safety of Ticagrelor and Aspirin in Patients With Moderate Ischemic Stroke Original Investigation Research Figure 1. Flow Diagram 11 073 Patients enrolled 57 Excluded 52 Not eligible 5 Patient decision 11 016 Randomized 5523 Randomized to ticagrelor, 90 mg, twice daily 5493 Randomized to placebo 5506 Received ≥1 dose 5470 Received ≥1 dose 17 Never received a dose 23 Never received a dose 5032 Had ischemic stroke as qualifying event 4953 Had ischemic stroke as qualifying event 2 With NIHSS score >5 1641 With NIHSS score of 4-5 1671 With NIHSS score of 4-5 3312 With NIHSS score of 0-3 3359 With NIHSS score of 0-3 540 Had TIA as qualifying event 491 Had TIA as qualifying event 5 With NIHSS score of 4-5 10 With NIHSS score of 4-5 487 With NIHSS score of 0-3 442 With NIHSS score of 0-3 48 With NIHSS score missing NIHSS indicates National Institutes of 39 With NIHSS score missing Health Stroke Scale; TIA, transient ischemic attack. 1 at the end-of-treatment visit 30 to 34 days after randomiza- of event). Differences in time to the first events between study tion. Disabling stroke (subsequent stroke with an mRS score groups within the NIHSS subgroups were evaluated using Cox greater than 1 at day 30) was an exploratory end point. The proportional hazards regression models, and hazard ratios safety outcomes included time to first severe bleeding event (HRs) were reported with 95% CIs. For the outcome of overall according to the Global Utilization of Streptokinase and Tis- disability and disabling stroke, differences between study sue Plasminogen Activator for Occluded Coronary Arteries groups within the baseline NIHSS groups were evaluated using (GUSTO) definition,7,9 a composite of the first intracranial hem- logistic regression models, and odds ratios (ORs) with their 95% orrhage or fatal bleeding event, the first moderate or severe CIs are reported. bleeding event according to the GUSTO definition, and pre- With 279 primary events in the moderate stroke group mature permanent discontinuation of study drugs because of (baseline NIHSS score of 4 to 5), the power was 31% to detect bleeding. All efficacy and safety analyses were based on an effect with an HR of 0.84. No adjustment for multiple com- investigator-assessed events.7 The definitions of stroke, death, parisons was made, and all P values were nominal, since all disability, and bleeding outcomes in the THALES trial have been analyses presented were exploratory. All analyses were per- reported previously.7 formed with SAS software version 9.4 (SAS Institute). Statistical Analysis This post hoc analysis was exploratory and hypothesis gener- ating across prespecified subgroups. Baseline characteristics Results were presented by antiplatelet treatment groups (ticagrelor or Study Participants placebo) and baseline NIHSS groups (NIHSS score of 0 to 3 vs For the present subgroup analysis, 9983 patients with stroke 4 to 5). Categorical variables were presented as percentages and were included of the 11 016 randomized patients in the THALES continuous variables as medians with interquartile ranges or trial, including 3312 patients (30.1%) with moderate stroke and means with standard deviations, as appropriate. 6671 patients (60.6%) with less severe stroke, excluding 2 pa- All efficacy and safety analyses were based on the inten- tients with stroke with an NIHSS score greater than 5 and 1031 tion-to-treat principle and included all randomized patients. patients with TIA (Figure 1). Of those in the moderate stroke Interaction between treatment assignment on all outcomes and group, 1293 (39.0%) were female, and the mean (SD) age was baseline NIHSS groups were evaluated by including terms for 64.5 (10.8) years; of those in the less severe stroke group, 2518 treatment assignment (ticagrelor or placebo), baseline NIHSS (37.7%) were female, and the mean (SD) age was 64.8 (11.2) group (NIHSS score of 0 to 3 or 4 to 5), and treatment × NIHSS years. Among the 3312 patients with moderate stroke, 1671 pa- group interaction as covariates in Cox or logistic regression tients were assigned to the ticagrelor group and 1641 to the pla- models. Interaction terms with a 2-tailed P value less than .05 cebo group. Baseline characteristics of patients with baseline were considered statistically significant. For outcomes of stroke NIHSS scores of 0 to 3 and 4 to 5 among those with ischemic or death, ischemic stroke, and bleeding events, we estimated stroke as a qualifying event and those in the entire popula- event rates by Kaplan-Meier method and presented the time tion are presented in Table 1 and eTable 1 in Supplement 2, re- to the first event using Kaplan-Meier curves (1 − proportion free spectively. Compared with those with baseline NIHSS scores jamaneurology.com (Reprinted) JAMA Neurology September 2021 Volume 78, Number 9 1093 Downloaded From: https://jamanetwork.com/ on 03/22/2023 Research Original Investigation Efficacy and Safety of Ticagrelor and Aspirin in Patients With Moderate Ischemic Stroke Table 1. Baseline Characteristics of Included Patients With Stroke by Baseline National Institutes of Health Stroke Scale (NIHSS) Group No. (%) NIHSS score of 0-3 NIHSS score of 4-5 Ticagrelor Placebo Total Ticagrelor Placebo Total Characteristic (n = 3359) (n = 3312) (n = 6671) (n = 1671) (n = 1641) (n = 3312) Age, mean (SD), y 65.0 (11.1) 64.6 (11.3) 64.8 (11.2) 64.4 (10.8) 64.6 (10.8) 64.5 (10.8) Female 1237 (36.8) 1281 (38.7) 2518 (37.7) 650 (38.9) 643 (39.2) 1293 (39.0) Race White 1765 (52.5) 1711 (51.7) 3476 (52.1) 866 (51.8) 873 (53.2) 1739 (52.5) Black 15 (0.4) 19 (0.6) 34 (0.5) 4 (0.2) 9 (0.5) 13 (0.4) Asian 1529 (45.5) 1526 (46.1) 3055 (45.8) 703 (42.1) 668 (40.7) 1371 (41.4) Other 50 (1.5) 56 (1.7) 106 (1.6) 98 (5.9) 91 (5.5) 189 (5.7) Region Asia and Australia 1546 (46.0) 1536 (46.4) 3082 (46.2) 704 (42.1) 665 (40.5) 1369 (41.3) Europe 1652 (49.2) 1624 (49.0) 3276 (49.1) 844 (50.5) 846 (51.6) 1690 (51.0) North America 7 (0.2) 6 (0.2) 13 (0.2) 1 (0.1) 2 (0.1) 3 (0.1) Central and South America 154 (4.6) 146 (4.4) 300 (4.5) 122 (7.3) 128 (7.8) 250 (7.5) Blood pressure, median (IQR), mm Hg Systolic 148.0 147.0 148.0 150.0 150.0 150.0 (132.0-161.0) (132.0-160.0) (132.0-160.0) (138.5-166.0) (135.0-167.0) (137.0-166.0) Diastolic 82.0 83.0 83.0 85.0 85.0 85.0 (78.0-91.0) (77.0-91.0) (78.0-91.0) (80.0-92.0) (80.0-91.0) (80.0-92.0) a Body mass index, median (IQR) 25.7 25.6 25.7 26.0 25.8 25.9 (23.2-28.9) (23.1-28.7) (23.1-28.7) (23.2-29.1) (23.2-29.3) (23.2-29.1) Medical history Current smoker 937 (27.9) 860 (26.0) 1797 (26.9) 475 (28.4) 466 (28.4) 941 (28.4) Hypertension 2524 (75.1) 2416 (72.9) 4940 (74.1) 1370 (82.0) 1343 (81.8) 2713 (81.9) Dyslipidemia 1278 (38.0) 1182 (35.7) 2460 (36.9) 627 (37.5) 614 (37.4) 1241 (37.5) Diabetes (type 1 and 2) 872 (26.0) 846 (25.5) 1718 (25.8) 483 (28.9) 455 (27.7) 938 (28.3) Prior ischemic heart disease 309 (9.2) 301 (9.1) 610 (9.1) 169 (10.1) 169 (10.3) 338 (10.2) Congestive heart failure 104 (3.1) 103 (3.1) 207 (3.1) 85 (5.1) 86 (5.2) 171 (5.2) Previous ischemic stroke 524 (15.6) 547 (16.5) 1071 (16.1) 308 (18.4) 272 (16.6) 580 (17.5) Previous TIA 128 (3.8) 107 (3.2) 235 (3.5) 57 (3.4) 49 (3.0) 106 (3.2) Taking aspirin prior to index event 476 (14.2) 401 (12.1) 877 (13.1) 184 (11.0) 159 (9.7) 343 (10.4) Taking clopidogrel 49 (1.5) 53 (1.6) 102 (1.5) 15 (0.9) 10 (0.6) 25 (0.8) prior to index event Time from symptom onset to randomization, h <12 1145 (34.1) 1109 (33.5) 2254 (33.8) 460 (27.5) 448 (27.3) 908 (27.4) ≥12 2214 (65.9) 2203 (66.5) 4417 (66.2) 1211 (72.5) 1193 (72.7) 2404 (72.6) Qualifying event Ischemic stroke 3359 (100) 3312 (100) 6671 (100) 1671 (100) 1641 (100) 3312 (100) TIA 0 0 0 0 0 0 Persistent signs or symptoms at the 3233 (96.2) 3157 (95.3) 6390 (95.8) 1670 (99.9) 1641 (100) 3311 (100) time of randomization Acute ischemic brain lesion at the 1633 (48.6) 1655 (50.0) 3288 (49.3) 875 (52.4) 835 (50.9) 1710 (51.6) time of randomization Abbreviations: IQR, interquartile range; TIA, transient ischemic attack. a Calculated as weight in kilograms divided by height in meters squared. of 0 to 3, patients with NIHSS scores of 4 to 5 were less likely Efficacy Outcomes to be Asian, to take aspirin and clopidogrel prior to the index Overall, the rate of the primary efficacy end point was greater event, and to be randomized within 12 hours from symptom in patients with moderate stroke compared with those with less onset but more likely to have hypertension, diabetes, and con- severe stroke (7.6% [129 of 1671] vs 4.7% [158 of 3359] in the gestive heart failure. There were no major imbalances be- ticagrelor group; 9.1% [150 of 1641] vs 5.7% [190 of 3312] in the tween the 2 treatment groups within the moderate and less placebo group) (Table 2). No significant interaction between severe stroke groups. treatment assignment and NIHSS group for the primary effi- 1094 JAMA Neurology September 2021 Volume 78, Number 9 (Reprinted) jamaneurology.com Downloaded From: https://jamanetwork.com/ on 03/22/2023 Efficacy and Safety of Ticagrelor and Aspirin in Patients With Moderate Ischemic Stroke Original Investigation Research Table 2. Outcomes of Patients Receiving Ticagrelor or Placebo by Baseline National Institutes of Health Stroke Scale (NIHSS) Group Ticagrelor (n = 5030) Placebo (n = 4953) Event rate Event rate Total (KM Total (KM patients, Events, estimate), patients, Events, estimate), P P value for Outcome No. No. (%) % No. No. (%) % Hazard ratio (95% CI) value interaction Primary efficacy end point Stroke or death NIHSS score of 0-3 3359 158 (4.7) 4.7 3312 190 (5.7) 5.7 0.82 (0.66-1.01) .06 .88 NIHSS score of 4-5 1671 129 (7.7) 7.6 1641 150 (9.1) 9.1 0.84 (0.66-1.06) .14 Secondary efficacy end point Ischemic stroke NIHSS score of 0-3 3359 143 (4.3) 4.2 3312 182 (5.5) 5.4 0.77 (0.62-0.96) .02 .64 NIHSS score of 4-5 1671 121 (7.2) 7.2 1641 141 (8.6) 8.6 0.83 (0.65-1.06) .14 Safety end points GUSTO severe bleeding NIHSS score of 0-3 3359 16 (0.5) 0.5 3312 3 (0.1) 0.1 5.28 (1.54-18.10) .008 .26 NIHSS score of 4-5 1671 8 (0.5) 0.5 1641 4 (0.2) 0.2 1.97 (0.59-6.53) .27 Intracranial hemorrhage or fatal bleeding NIHSS score of 0-3 3359 13 (0.4) 0.4 3312 3 (0.1) 0.1 4.29 (1.22-15.04) .02 .41 NIHSS score of 4-5 1671 6 (0.4) 0.4 1641 3 (0.2) 0.2 1.97 (0.49-7.86) .34 Fatal bleeding NIHSS score of 0-3 3359 3 (0.1) 0.1 3312 1 (0.0) 0.0 2.96 (0.31-28.47) .35 .75 NIHSS score of 4-5 1671 5 (0.3) 0.3 1641 1 (0.1) 0.1 4.91 (0.57-42.05) .15 Intracranial hemorrhage NIHSS score of 0-3 3359 13 (0.4) 0.4 3312 3 (0.1) 0.1 4.29 (1.22-15.04) .02 .32 NIHSS score of 4-5 1671 5 (0.3) 0.3 1641 3 (0.2) 0.2 1.64 (0.39-6.86) .50 Hemorrhagic stroke NIHSS score of 0-3 3359 5 (0.1) 0.1 3312 2 (0.1) 0.1 2.47 (0.48-12.73) .28 .99 NIHSS score of 4-5 1671 3 (0.2) 0.2 1641 0 NA NA NA GUSTO moderate or severe bleeding NIHSS score of 0-3 3359 20 (0.6) 0.6 3312 5 (0.2) 0.2 3.95 (1.48-10.54) .006 .33 NIHSS score of 4-5 1671 12 (0.7) 0.7 1641 6 (0.4) 0.3 1.98 (0.74-5.28) .17 Premature permanent discontinuation of study drugs due to bleeding NIHSS score of 0-3 3359 90 (2.7) 2.9 3312 19 (0.6) 0.6 4.75 (2.89-7.79) <.001 .79 NIHSS score of 4-5 1671 47 (2.8) 3.0 1641 11 (0.7) 0.7 4.24 (2.20-8.17) <.001 Abbreviations: GUSTO, Global Utilization of Streptokinase and Tissue Plasminogen Activator for Occluded Coronary Arteries; KM, Kaplan-Meier; NA, not applicable. cacy end point of stroke or death was identified (Table 2), in- among those with moderate stroke (HR, 0.83; 95% CI, 0.65 to dicating consistent efficacy of ticagrelor vs placebo among pa- 1.06), compared with 143 patients (4.3%) in the ticagrelor group tients with stroke with a baseline NIHSS score of 4 to 5 and 0 vs 182 patients (5.5%) in the placebo group among those with to 3. In patients with moderate stroke, the observed primary less severe stroke (HR, 0.77; 95% CI, 0.62 to 0.96) (P for inter- end point event rate in patients taking ticagrelor was numeri- action = .64; Table 2). Similar findings were observed when the cally lower than that in patients taking placebo (7.6% vs 9.1%; entire population regardless of qualifying event was in- HR, 0.84; 95% CI, 0.66 to 1.06; risk difference, −1.54%; 95% cluded (eTable 2 and eFigure 1 in Supplement 2). CI, −3.43 to 0.35), with a number needed to treat of 65 to avoid The proportion of overall disability (mRS score greater than 1 stroke or death at day 30 (Table 2). This effect size was simi- 1) at day 30 was greater in patients with moderate stroke com- lar to that in patients with less severe stroke (4.7% vs 5.7%; HR, pared with those with less severe stroke (37.2% [612 of 1643] 0.82; 95% CI, 0.66 to 1.01; risk difference, −1.00%; 95% CI, vs 18.8% [620 of 3299] in the ticagrelor group; 37.6% [609 of −2.06 to 0.06), with a number needed to treat of 100. Kaplan- 1619] vs 19.4% [627 of 3236] in the placebo group; eTable 3 in Meier event curves for the primary efficacy end point by treat- Supplement 2). Efficacy of ticagrelor on overall disability at day ment assignment and baseline NIHSS group are shown in 30 in patients with stroke with a baseline NIHSS score of 4 to Figure 2. Ischemic stroke occurred in 121 patients (7.2%) in the 5 (37.2% vs 37.6%; OR, 0.98; 95% CI, 0.85 to 1.13) was similar ticagrelor group vs 141 patients (8.6%) in the placebo group to that in patients with a stroke with an NIHSS score of 0 to 3 jamaneurology.com (Reprinted) JAMA Neurology September 2021 Volume 78, Number 9 1095 Downloaded From: https://jamanetwork.com/ on 03/22/2023 Research Original Investigation Efficacy and Safety of Ticagrelor and Aspirin in Patients With Moderate Ischemic Stroke Figure 2. Kaplan-Meier Event Curves for the Primary Efficacy End Point (Stroke or Death) by Treatment Assignment and Baseline National Institutes of Health Stroke Scale (NIHSS) Group 10 Placebo NIHSS 4-5 group 8 Ticagrelor NIHSS 4-5 group Cumulative probability of primary end point, % 6 Placebo NIHSS 0-3 group Ticagrelor NIHSS 0-3 group 4 2 0 0 5 10 15 20 25 30 34 Time from randomization, d No. at risk Ticagrelor NIHSS 0-3 3359 3252 3221 3211 3203 3197 3195 718 Placebo NIHSS 0-3 3312 3190 3150 3139 3131 3125 3117 718 Ticagrelor NIHSS 4-5 1671 1578 1557 1552 1547 1543 1540 273 Placebo NIHSS 4-5 1641 1535 1508 1500 1495 1494 1490 303 Safety Outcomes Figure 3. Modified Rankin Scale (mRS) Score Distribution at Day 30 in Patients With an Ischemic Stroke by Baseline National Institutes Among those with stroke with a baseline NIHSS score of 4 to of Health Stroke Scale (NIHSS) Group 5, the primary safety end point, GUSTO severe bleeding, oc- curred in 8 patients (0.5%) in the ticagrelor group and in 4 pa- mRS score tients (0.2%) in the placebo group (HR, 1.97; 95% CI, 0.59 to 6 5 4 3 2 1 0 6.53; risk difference, 0.24%; 95% CI, −0.18 to 0.65), with a num- ber needed to harm of 425 to produce 1 severe bleeding within 200 30 days, while the primary safety end point in those with an NIHSS score of 0 to 3 occurred in 16 patients (0.5%) in the ti- cagrelor group and 3 patients (0.1%) in the placebo group (risk 150 difference, 0.39%; 95% CI, 0.13 to 0.64), with a number needed to harm of 259 (P for interaction = .26) (Table 2). In patients Patients, No. 100 with moderate stroke, GUSTO moderate or severe bleeding oc- curred in 12 patients (0.7%) in the ticagrelor group and in 6 pa- tients (0.4%) in the placebo group (HR, 1.98; 95% CI, 0.74 to 50 5.28). Similar to that in patients with less severe stroke, per- manent discontinuation of study drug in patients with mod- erate stroke was more common in the ticagrelor group than 0 Ticagrelor Placebo Ticagrelor Placebo in the placebo group. Similar findings were observed when the entire population regardless of qualifying event was in- NIHSS 4-5 NIHSS 0-3 cluded (eTable 2 in Supplement 2). (18.8% vs 19.4%; OR, 0.96; 95% CI, 0.85 to 1.09) (P for inter- action = .86). Distributions of mRS scores at day 30 with and Discussion without a subsequent ischemic stroke are shown in eTable 4 in Supplement 2. These data suggest a greater disability in pa- In this exploratory analysis of the THALES trial, we found that tients with a subsequent stroke compared with those with- although the risk of subsequent stroke or death was numeri- out a subsequent stroke. In patients with moderate stroke, a cally higher in patients presenting with moderate ischemic subsequent disabling stroke up to day 30 occurred in 99 of 1667 stroke, the relative efficacy and safety of ticagrelor added to patients (5.9%) in the ticagrelor group and in 115 of 1637 pa- aspirin were similar to those presenting with less severe is- tients (7.0%) in the placebo group (OR, 0.83; 95% CI, 0.63 to chemic events. Ticagrelor plus aspirin treatment appears 1.10), with a number needed to treat of 91 to avoid 1 disabling beneficial and safe in both the less severe and moderate stroke stroke at day 30; the effect size was similar in those with less subgroups. severe stroke (OR, 0.78; 95% CI, 0.59 to 1.03) (P for interac- Prognosis and clinical management of patients with is- tion = .74; Figure 3; eTable 3 in Supplement 2). Similar find- chemic stroke of moderate severity has drawn physicians’ at- ings were observed when the entire population regardless of tention in recent years. Consistent with our results, previous qualifying event was included (eTables 5 and eTable 6 and eFig- studies showed that patients with strokes of moderate sever- ure 2 in Supplement 2). ity have a higher risk of deterioration and poor outcome com- 1096 JAMA Neurology September 2021 Volume 78, Number 9 (Reprinted) jamaneurology.com Downloaded From: https://jamanetwork.com/ on 03/22/2023 Efficacy and Safety of Ticagrelor and Aspirin in Patients With Moderate Ischemic Stroke Original Investigation Research pared with those with low severity.10,11 The efficacy of DAPT event, a subsequent stroke can have a substantial impact on with clopidogrel and aspirin in patients with high-risk TIA or outcomes. A previous secondary analysis of the THALES trial minor ischemic stroke with a baseline NIHSS score of 0 to 3 was showed that ticagrelor plus aspirin treatment was superior to evaluated in the CHANCE and POINT trials.5,6 Pooled analy- aspirin alone in preventing disabling stroke or death at 30 days sis of the POINT and CHANCE trials demonstrated that 21- and reduced the total burden of disability owing to ischemic day clopidogrel plus aspirin therapy reduced the risk of ma- stroke recurrence in the entire population of the trial.24 The jor ischemic events at 90 days by 34% without any significant present study indicates that ticagrelor plus aspirin treatment increase in major hemorrhages (0.1% vs 0.3%; nominal P = .10) prevents disabling stroke in the subgroups of patients present- for patients with high-risk TIA or minor ischemic stroke with ing with minor and with moderate stroke. Given the high risk a baseline NIHSS score of 0 to 3.12 Patients with stroke of higher of new stroke and developing disabling stroke, intensive an- severity may respond differently than patients with mild stroke tiplatelet therapy should also be started as a matter of ur- severity to DAPT; however, to our knowledge, no clinical trials gency to prevent new or worsening stroke events in patients have evaluated the efficacy of DAPT in patients with stroke with with ischemic stroke with a baseline NIHSS score of 4 to 5. an NIHSS score of 4 to 5. In addition, the risk of bleeding, es- pecially severe bleedings events, such as intracranial hemor- Limitations rhage, was the major concern when intensive antiplatelet This study has several limitations. This is an exploratory sub- therapy was administrated for patients with moderate stroke group analysis of a large-scale randomized clinical trial, and in clinical practice.13 Prior DAPT trials in patients with acute type I errors can be introduced in such analyses. Although the ischemic stroke have suggested an increase in minor extracra- THALES trial is the largest acute DAPT stroke trial to date, the nial bleedings with higher NIHSS scores, but this was not ob- number of patients with stroke with an NIHSS score of 4 to 5 served for intracranial hemorrhage.14,15 The present study pro- was 3312 of 11 016 patients (30.1%), with modest power. There- vided additional evidence that DAPT with ticagrelor and aspirin fore, caution is warranted when interpreting the present find- had consistent benefit and risk of severe bleeding events, such ings, especially for the safety outcome due to the small num- as intracranial hemorrhage, between patients with baseline ber of bleeding events. Most patients with acute ischemic stroke NIHSS scores of 4 to 5 and those with baseline NIHSS scores present with an NIHSS score of 0 to 525; however, the efficacy of 0 to 3. Additionally, the absolute increase in risk of intra- and safety of ticagrelor in patients with NIHSS scores greater cranial bleedings or other severe bleeding events for ticagre- than 5 has not been studied. lor plus aspirin treatment was low (risk difference, 0.24%; 95% CI, −0.18 to 0.65), with a number needed to harm of 425 to pro- duce 1 severe bleeding event within 30 days. A major sequela of stroke is disability, which in turn is as- Conclusions sociated with poor long-term prognosis and health-related The present exploratory analysis showed that patients with an quality of life as well as high societal cost.16-18 Although pa- acute ischemic stroke with a baseline NIHSS score of 4 to 5 have tients with minor stroke or TIA had mild and nondisabling a consistent benefit from receiving treatment with ticagrelor symptoms, these patients have a high risk of new stroke dur- plus aspirin vs aspirin alone as patients with baseline NIHSS ing the short-term period (5% to 12% within 3 months), most score of 0 to 3, with no further increase in the risk of intracra- of which are disabling.5,19,20 Because of greater severity of the nial bleedings or other severe bleeding events. Although the index stroke, patients with baseline NIHSS scores of 4 to 5 may risk of disability is higher overall in patients with stroke with have more overall disability and risk of developing disabling a baseline NIHSS score of 4 to 5 owing to greater severity of stroke than those with baseline NIHSS scores of 0 to 3.21-23 How- the index event, DAPT with ticagrelor and aspirin reduced the ever, even if there is a high risk of disability from the index number of patients with subsequent disabling stroke. ARTICLE INFORMATION Washington, DC (Evans); Department of Medical Critical revision of the manuscript for important Accepted for Publication: June 11, 2021. Sciences, Uppsala University, Uppsala, Sweden intellectual content: Li, Amarenco, Denison, Evans, (James); Stroke Unit, Vall d’Hebron Hospital, Himmelmann, James, Birve, Ladenvall, Molina, Published Online: July 9, 2021. Barcelona, Spain (Molina); Dean’s Office, Dell Johnston. doi:10.1001/jamaneurol.2021.2440 Medical School, University of Texas at Austin Statistical analysis: Birve. Open Access: This is an open access article (Johnston). Obtained funding: Himmelmann. distributed under the terms of the CC-BY-NC-ND Author Contributions: Drs Wang and Johnston had Administrative, technical, or material support: License. © 2021 Wang Y et al. JAMA Neurology. full access to all of the data in the study and take Wang, Pan, Denison, Himmelmann, Ladenvall, Author Affiliations: Department of Neurology, responsibility for the integrity of the data and the Johnston. Beijing Tiantan Hospital, Capital Medical University, accuracy of the data analysis. Drs Wang and Pan Study supervision: Li, Amarenco, Denison, Beijing, China (Wang, Pan, Li); China National contributed equally to this work as co–first authors. Himmelmann, Ladenvall, Molina, Johnston. Clinical Research Center for Neurological Diseases, Study concept and design: Wang, Pan, Li, Amarenco, Conflict of Interest Disclosures: Dr Wang has Beijing, China (Wang, Pan, Li); Department of Denison, Evans, Himmelmann, James, Ladenvall, received grants from AstraZeneca, Sanofi, and Neurology and Stroke Center, Bichat–Claude Molina, Johnston. Amgen as well as consulting fees from Sanofi. Bernard Hospital, University of Paris, Paris, France Acquisition, analysis, or interpretation of data: Dr Amarenco has received grants and personal fees (Amarenco); Biopharmaceuticals Research and Wang, Amarenco, Denison, Himmelmann, James, from AstraZeneca, Bristol Myers Squibb, Pfizer, and Development, AstraZeneca, Gothenburg, Sweden Birve, Ladenvall, Johnston. Boston Scientific; grants from AstraZeneca and (Denison, Himmelmann, Birve, Ladenvall); Drafting of the manuscript: Wang, Pan. Merck; and personal fees from Sanofi, Janssen, Biostatistics Center, George Washington University, GlaxoSmithKline, FibroGen, Shinpoong, Amgen, jamaneurology.com (Reprinted) JAMA Neurology September 2021 Volume 78, Number 9 1097 Downloaded From: https://jamanetwork.com/ on 03/22/2023 Research Original Investigation Efficacy and Safety of Ticagrelor and Aspirin in Patients With Moderate Ischemic Stroke Bayer, Kowa, Viatris, Novo Nordisk, and Portola high-risk TIA. Eur Stroke J. Published online March haemorrhage in patients with minor strokes: Pharmaceuticals. Dr Evans has received personal 11, 2021. doi:10.1177/23969873211000877 a subgroup analysis of the CHANCE trial. Stroke fees from AstraZeneca and Johnson & Johnson. 5. Wang Y, Wang Y, Zhao X, et al; CHANCE Vasc Neurol. 2016;1(2):29-36. doi:10.1136/svn-2016- Dr James has received grants paid to his institution Investigators. Clopidogrel with aspirin in acute 000008 from AstraZeneca, Novartis, Jansen, and Bayer. minor stroke or transient ischemic attack. N Engl J 16. Park JH, Ovbiagele B. Relationship of functional Dr Johnston has received grants from AstraZeneca Med. 2013;369(1):11-19. doi:10.1056/NEJMoa1215340 disability after a recent stroke with recurrent stroke and Sanofi. No other disclosures were reported. risk. Eur J Neurol. 2016;23(2):361-367. doi:10.1111/ 6. Johnston SC, Easton JD, Farrant M, et al; Clinical Funding/Support: The study was funded by Research Collaboration, Neurological Emergencies ene.12837 AstraZeneca. Treatment Trials Network, and the POINT 17. Wang YL, Pan YS, Zhao XQ, et al; CHANCE Role of the Funder/Sponsor: The funding source Investigators. Clopidogrel and aspirin in acute investigators. Recurrent stroke was associated with had a role in the design and conduct of the study; ischemic stroke and high-risk TIA. N Engl J Med. poor quality of life in patients with transient collection, management, analysis, and 2018;379(3):215-225. doi:10.1056/NEJMoa1800410 ischemic attack or minor stroke: finding from the interpretation of the data; preparation, review, or 7. Johnston SC, Amarenco P, Denison H, et al; CHANCE trial. CNS Neurosci Ther. 2014;20(12): approval of the manuscript; and decision to submit THALES Investigators. Ticagrelor and aspirin or 1029-1035. doi:10.1111/cns.12329 the manuscript for publication. aspirin alone in acute ischemic stroke or TIA. N Engl 18. Ganesh A, Luengo-Fernandez R, Wharton RM, The THALES Steering Committee and J Med. 2020;383(3):207-217. doi:10.1056/ Rothwell PM; Oxford Vascular Study. Ordinal vs Investigators: The THALES Steering Committee NEJMoa1916870 dichotomous analyses of modified Rankin Scale, and Investigators are listed in Supplement 3. 8. Johnston SC, Amarenco P, Denison H, et al; 5-year outcome, and cost of stroke. Neurology. Meeting Presentation: This paper was presented THALES Investigators. The Acute Stroke or 2018;91(21):e1951-e1960. doi:10.1212/WNL. at the 7th Annual Scientific Session of the Chinese Transient Ischemic Attack Treated with Ticagrelor 0000000000006554 Stroke Association & Tiantan International Stroke and Aspirin for Prevention of Stroke and Death 19. Wu CM, McLaughlin K, Lorenzetti DL, Hill MD, Conference; July 10, 2021; Bejing, China. (THALES) trial: rationale and design. Int J Stroke. Manns BJ, Ghali WA. Early risk of stroke after Data Sharing Statement: See Supplement 4. 2019;14(7):745-751. doi:10.1177/1747493019830307 transient ischemic attack: a systematic review and 9. GUSTO Investigators. An international meta-analysis. Arch Intern Med. 2007;167(22):2417- Additional Contributions: Editing support at the 2422. doi:10.1001/archinte.167.22.2417 manuscript revision stage, under the direction of randomized trial comparing four thrombolytic the authors, was provided by Catherine Crookes, strategies for acute myocardial infarction. N Engl J 20. Amarenco P, Lavallée PC, Labreuche J, et al; BSc (Ashfield Medical Communications, Med. 1993;329(10):673-682. doi:10.1056/ TIAregistry.org Investigators. One-year risk of Maidenhead, United Kingdom), which was funded NEJM199309023291001 stroke after transient ischemic attack or minor by AstraZeneca. 10. Fischer U, Baumgartner A, Arnold M, et al. stroke. N Engl J Med. 2016;374(16):1533-1542. What is a minor stroke? Stroke. 2010;41(4):661-666. doi:10.1056/NEJMoa1412981 REFERENCES doi:10.1161/STROKEAHA.109.572883 21. Amarenco P, Denison H, Evans SR, et al; 1. Powers WJ, Rabinstein AA, Ackerson T, et al. 11. Xiong Y, Gu H, Zhao XQ, et al. Clinical THALES Steering Committee and Investigators. Guidelines for the early management of patients characteristics and in-hospital outcomes of varying Ticagrelor added to aspirin in acute nonsevere with acute ischemic stroke: 2019 update to the definitions of minor stroke: from a large-scale ischemic stroke or transient ischemic attack of 2018 guidelines for the early management of acute nation-wide longitudinal registry. Stroke. 2021;52 atherosclerotic origin. Stroke. 2020;51(12):3504-3513. ischemic stroke: a guideline for healthcare (4):1253-1258. doi:10.1161/STROKEAHA.120.031329 doi:10.1161/STROKEAHA.120.032239 professionals from the American Heart 12. Pan Y, Elm JJ, Li H, et al. Outcomes associated 22. Cucchiara B, Elm J, Easton JD, et al. Disability Association/American Stroke Association. Stroke. with clopidogrel-aspirin use in minor stroke or after minor stroke and transient ischemic attack in 2019;50(12):e344-e418. doi:10.1161/STR. transient ischemic attack: a pooled analysis of the POINT trial. Stroke. 2020;51(3):792-799. 0000000000000211 Clopidogrel in High-Risk Patients With Acute doi:10.1161/STROKEAHA.119.027465 2. Wang Y, Liu M, Pu C. 2014 Chinese guidelines for Non-Disabling Cerebrovascular Events (CHANCE) 23. Cucchiara B, George DK, Kasner SE, et al. secondary prevention of ischemic stroke and and Platelet-Oriented Inhibition in New TIA and Disability after minor stroke and TIA: a secondary transient ischemic attack. Int J Stroke. 2017;12(3): Minor Ischemic Stroke (POINT) trials. JAMA Neurol. analysis of the SOCRATES trial. Neurology. 2019;93 302-320. doi:10.1177/1747493017694391 2019;76(12):1466-1473. doi:10.1001/jamaneurol.2019. (7):e708-e716. doi:10.1212/WNL. 3. Liu L, Chen W, Zhou H, et al; Chinese Stroke 2531 0000000000007936 Association Stroke Council Guideline Writing 13. Ferreiro JL, Sibbing D, Angiolillo DJ. Platelet 24. Amarenco P, Denison H, Evans SR, et al; Committee. Chinese Stroke Association guidelines function testing and risk of bleeding complications. THALES Steering Committee and Investigators. for clinical management of cerebrovascular Thromb Haemost. 2010;103(6):1128-1135. Ticagrelor added to aspirin in acute ischemic stroke disorders: executive summary and 2019 update of doi:10.1160/TH09-11-0799 or transient ischemic attack in prevention of clinical management of ischaemic cerebrovascular 14. Tillman H, Johnston SC, Farrant M, et al. Risk for disabling stroke: a randomized clinical trial. JAMA diseases. Stroke Vasc Neurol. 2020;5(2):159-176. major hemorrhages in patients receiving Neurol. 2020;78(2):1-9. doi:10.1001/jamaneurol. doi:10.1136/svn-2020-000378 clopidogrel and aspirin compared with aspirin alone 2020.4396 4. Dawson J, Merwick Á, Webb A, Dennis M, Ferrari after transient ischemic attack or minor ischemic 25. Reeves M, Khoury J, Alwell K, et al. Distribution J, Fonseca AC; European Stroke Organisation. stroke: a secondary analysis of the POINT of National Institutes of Health Stroke Scale in the European Stroke Organisation expedited randomized clinical trial. JAMA Neurol. 2019;76(7): Cincinnati/Northern Kentucky Stroke Study. Stroke. recommendation for the use of short-term dual 774-782. doi:10.1001/jamaneurol.2019.0932 2013;44(11):3211-3213. doi:10.1161/STROKEAHA.113. antiplatelet therapy early after minor stroke and 15. Wang D, Gui L, Dong Y, et al. Dual antiplatelet 002881 therapy may increase the risk of non-intracranial 1098 JAMA Neurology September 2021 Volume 78, Number 9 (Reprinted) jamaneurology.com Downloaded From: https://jamanetwork.com/ on 03/22/2023