Research

JAMA Neurology | Original Investigation

Sublingual Edaravone Dexborneol for the Treatment

of Acute Ischemic Stroke
The TASTE-SL Randomized Clinical Trial
Yu Fu, MD; Anxin Wang, PhD; Renhong Tang, MD; Shuya Li, MD; Xue Tian, PhD; Xue Xia, PhD; Jinsheng Ren, MD; Shibao Yang, MD;
Rong Chen, MD; Shunwei Zhu, MD; Xiaofei Feng, MD; Jinliang Yao, MD; Yan Wei, MD; Xueshuang Dong, MD; Yun Ling, MD;
Fei Yi, MD; Qian Deng, MD; Cunju Guo, MD; Yi Sui, MD; Shugen Han, MD; Guoqiang Wen, MD; Chuanling Li, MD; Aiqin Dong, MD;
Xin Sun, MD; Zhimin Wang, MD; Xueying Shi, MD; Bo Liu, MD; Dongsheng Fan, MD

Visual Abstract
IMPORTANCE Sublingual edaravone dexborneol, which can rapidly diffuse and be absorbed Editorial
through the oral mucosa after sublingual exposure, is a multitarget brain cytoprotection
composed of antioxidant and anti-inflammatory ingredients edaravone and dexborneol. Supplemental content

OBJECTIVE To investigate the efficacy and safety of sublingual edaravone dexborneol on

90-day functional outcome in patients with acute ischemic stroke (AIS).

DESIGN, SETTING, AND PARTICIPANTS This was a double-blind, placebo-controlled,

multicenter, parallel-group, phase 3 randomized clinical trial conducted from June 28, 2021,
to August 10, 2022, with 90-day follow-up. Participants were recruited from 33 centers in
China. Patients randomly assigned to treatment groups were aged 18 to 80 years and had a
National Institutes of Health Stroke Scale score between 6 and 20, a total motor deficit score
of the upper and lower limbs of 2 or greater, a clinically diagnosed AIS symptom within 48
hours, and a modified Rankin Scale (mRS) score of 1 or less before stroke. Patients who did
not meet the eligibility criteria or declined to participate were excluded.

INTERVENTION Patients were assigned, in a 1:1 ratio, to receive sublingual edaravone

dexborneol (edaravone, 30 mg; dexborneol, 6 mg) or placebo (edaravone, 0 mg; dexborneol,
60 μg) twice daily for 14 days and were followed up until 90 days.

MAIN OUTCOMES AND MEASURES The primary efficacy outcome was the proportion of
patients with mRS score of 1 or less on day 90 after randomization.

RESULTS Of 956 patients, 42 were excluded. A total of 914 patients (median [IQR] age, 64.0
[56.0-70.0] years; 608 male [66.5%]) were randomly allocated to the edaravone dexborneol
group (450 [49.2%]) or placebo group (464 [50.8%]). The edaravone dexborneol group
showed a significantly higher proportion of patients experiencing good functional outcomes
on day 90 after randomization compared with the placebo group (290 [64.4%] vs 254
[54.7%]; risk difference, 9.70%; 95% CI, 3.37%-16.03%; odds ratio, 1.50; 95% CI, 1.15-1.95,
P = .003). The rate of adverse events was similar between the 2 groups (89.8% [405 of 450]
vs 90.1% [418 of 464]).

CONCLUSION AND RELEVANCE Among patients with AIS within 48 hours, sublingual
edaravone dexborneol could improve the proportion of those achieving a favorable
functional outcome at 90 days compared with placebo.

TRIAL REGISTRATION ClinicalTrials.gov Identifier: NCT04950920

Author Affiliations: Author

affiliations are listed at the end of this
article.
Corresponding Author: Dongsheng
Fan, MD, Department of Neurology,
Peking University Third Hospital,
JAMA Neurol. doi:10.1001/jamaneurol.2023.5716 Xinhua Dao, Beijing 100191, China
Published online February 19, 2024. (dsfan@sina.com).

(Reprinted) E1

Downloaded from jamanetwork.com by guest on 02/24/2024

 Research Original Investigation
T
he major goal of intervention in acute ischemic stroke (AIS)
is to salvage the ischemic penumbra.1 Brain cytoprotec-
tion, as proposed by Stroke Treatment Academic Indus-
try Roundtable (STAIR), has the capability of reducing ischemic
brain injury by antagonizing detrimental molecular events in all
brain components.2,3 Given the intended goal for broad protec-
tion, cytoprotective approaches that exert pleiotropic effects
on multiple targets of the ischemic cascade, including excito-
toxicity, oxidative stress, inflammation, apoptosis, etc, are the
priority recommendation.2 Recently, the Efficacy and Safety of
Nerinetide for the Treatment of Acute Ischemic Stroke (ESCAPE-
NA1) trial reported that eicosapeptide nerinetide had a potentially
cytoprotective effect via inhibition of neuronal excitotoxicity and
reduction of the production of nitric oxide among patients with
AIS who were not treated with alteplase after endovascular
thrombectomy. The findings suggested that brain cytoprotection
in human stroke might be possible and promising, although
further confirmations are still required.4
Edaravone dexborneol is another intravenously administered
multitarget brain cytoprotective agent composed of edaravone
and dexborneol.5 The 2 components of edaravone dexborneol
were shown to be effective in the prevention and treatment of
AIS.6,7 The Treatment of Acute Ischemic Stroke with Edaravone
Dexborneol(TASTE)trialhasindicatedthat,comparedwithedara-
vone alone, intravenous edaravone dexborneol could improve
90-day functional outcomes in patients with AIS.8,9 However,
intravenous drug administration largely depends on health
care resources, which may be delayed or limited during busy pe-
riods (ie, times during which there are many hospital admissions)
and thereby cause a significant barrier in terms of time delay to
protect neuronal function. Sublingual edaravone dexborneol,
an innovative, multitarget drug composed of edaravone and
Figure 1. Enrollment and Randomization of the Patients
956 Patients assessed for eligibility
42 Excluded
914 Randomized
450 Received sublingual edaravone dexborneol and
were included in the intention-to-treat population
60 Excluded
22 Discontinued trial treatment
prematurely
10 Had adverse events or serious adverse
events
11 Withdrew by patient decision
1 Withdrawn by physicians
20 Lost to 90-d follow-up
16 Needed prohibited concomitant
medications
2 Enrolled inappropriately
390 Included in the per-protocol population
E2 JAMA Neurology Published online February 19, 2024 (Reprinted)
Downloaded from jamanetwork.com by guest on 02/24/2024
Sublingual Edaravone Dexborneol in Acute Ischemic Stroke
Key Points
Question Does sublingual edaravone dexborneol improve
functional outcome in patients with acute ischemic stroke?
Findings In this randomized clinical trial including 914 patients
who received sublingual edaravone dexborneol or placebo, the
proportion of patients achieving a favorable outcome defined as a
90-day modified Rankin Scale score of 1 or less was 64.4% in the
sublingual edaravone dexborneol group, which was significantly
higher than 54.7% in the placebo group. The rate of adverse
events was similar between the 2 groups.
Meaning Sublingual edaravone dexborneol, as a fast-acting and
convenient agent, could improve the proportion of patients who
achieve good clinical outcomes at 90 days compared with placebo
among patients with acute ischemic stroke.
dexborneol, can rapidly diffuse and be absorbed through the oral
mucosa after sublingual exposure, without interfering with dis-
position and elimination properties and significantly alter the total
bioavailabilityofedaravoneanddexborneol.Whetheritcouldpro-
vide a rapid brain cytoprotective effect for patients with AIS re-
mains unclear. Therefore, the Treatment of Acute Ischemic Stroke
with Sublingual Edaravone Dexborneol (TASTE-SL) trial was de-
signed to investigate the effects of sublingual edaravone dexbor-
neol on 90-day functional outcome in patients with AIS.
Methods
Study Design and Patients
This was a phase 3, double-blind, placebo-controlled, multi-
center, parallel-group, randomized clinical trial conducted in
38 Did not meet inclusion criteria or met
exclusion criteria
4 Declined to participate
464 Received placebo and were included in the
intention-to-treat population
61 Excluded
18 Discontinued trial treatment
prematurely
8 Had adverse events or serious adverse
events
9 Withdrew by patient decision
1 Withdrawn by physicians
15 Lost to 90-d follow-up
23 Needed prohibited concomitant
medications
5 Were outside the study window The prohibited concomitant
medications included drugs with
indications of neuroprotection or
403 Included in the per-protocol population
cerebral infarction and unmarketed
drugs or other drugs for clinical trials.
jamaneurology.com
 Sublingual Edaravone Dexborneol in Acute Ischemic Stroke Original Investigation Research

protocol (Supplement 1).10 Patients were eligible if they were

Table 1. Baseline Characteristics
aged from 18 to 80 years old, had a National Institutes of Health
Edaravone Stroke Scale (NIHSS) score between 6 and 20, had a total mo-
dexborneol group Placebo group
Characteristics (n = 450) (n = 464) tor deficit score of the upper and lower limbs of 2 or greater,
Age, median (IQR), y 64.1 (56.0-69.8) 64.6 (57.1-71.4) had a clinically diagnosed AIS symptom within 48 hours, and
Sex, No. (%) had a modified Rankin Scale (mRS) score of 1 or less before the
Female 141 (31.3) 165 (35.6) stroke. All patients included in the study were of Chinese Han
Male 309 (68.7) 299 (64.4) ethnicity. Coordinating centers and clinical site information are
BMI, median (IQR)a,b 24.6 (22.5-26.6) 24.7 (22.5-26.8) available in eAppendix 1 and 2 in Supplement 2. Detailed ex-
Blood pressure, median (IQR), clusion criteria are shown in eTable 1 in Supplement 2. The
mm Hg study was conducted in accordance with the principles of the
Systolic 147 (135-162) 147 (136-160) Declaration of Helsinki and followed the Consolidated Stan-
Diastolic 86 (79-95) 86 (80-94) dards of Reporting Trials (CONSORT) reporting guidelines.
Medical history, No. (%)
Stroke 124 (27.6) 143 (30.8) Randomization and Masking
Hypertension 367 (81.6) 366 (78.9) Within 48 hours after symptom onset, eligible patients were
Diabetes 184 (40.9) 187 (40.3) randomly assigned in a 1:1 ratio to receive sublingual edara-
Dyslipidemia 217 (48.2) 221 (47.6) vone dexborneol or placebo. The randomization was strati-
Coronary heart disease 132 (29.3) 136 (29.3) fied by clinical centers and time onset of AIS (≤24 hours and
NIHSS score, median (IQR) 7 (7-8) 7 (7-8) >24 hours). Randomization was implemented via a central-
mRS score before onset, No. (%) ized, interactive web-based response system.
0 415 (92.2) 423 (91.2) The packaged, labeled, and assigned progressions were
1 35 (7.8) 41 (8.8) conducted by an independent clinical research organization.
Time to randomization, h The 2 drugs were identical in color and shape. All drugs were
≤24 196 (43.6) 203 (43.8)
concealed in uniform packages with the same label, which did
not indicate the contents of the package, and 1 single-use drug
>24 254 (56.4) 261 (56.3)
tablet was packed in a small box, with each large box contain-
TOAST subtype, No. (%)a
ing 30 small boxes (including 2 small boxes of backup drugs).
Large-artery atherosclerosis 247 (55.8) 245 (53.5)
Drug management personnel distributed the study drugs ac-
Cardioembolism 10 (2.3) 16 (3.5)
cording to the requirements of the clinical trial protocol and
Small-vessel occlusion 178 (40.2) 185 (40.4)
recorded in a timely fashion the distribution and recovery of
Other determined etiology 7 (1.6) 11 (2.4)
each study drug on a special record sheet. The blinding and
Undetermined etiology 1 (0.2) 1 (0.2)
allocation concealment were guaranteed.
Kidney functionc
Normal 399 (88.7) 414 (89.2)
Treatment
Mildly impaired 47 (10.4) 48 (10.3) Patients in the sublingual edaravone dexborneol group re-
Moderately to severely 4 (0.9) 2 (0.4) ceived a sublingual dose of edaravone dexborneol, 36 mg
impaired
(edaravone, 30 mg; dexborneol, 6 mg), twice a day for 14 con-
Abbreviations: BMI, body mass index; mRS, modified Rankin Scale;
secutive days. Patients in the placebo group received a sub-
NIHSS, National Institutes of Health Stroke Scale; TOAST, Trial of Org 10172 in
Acute Stroke Treatment. lingual placebo drug (edaravone, 0 mg; dexborneol, 60 μg
a
The number of patients with missing data on BMI was 20 in the edaravone [simulated the cool taste of sublingual edaravone dexbor-
dexborneol group and 17 in the placebo group; missing data on TOAST neol]) twice a day for 14 consecutive days. All patients were
subtypes was 7 in the edaravone dexborneol group and 6 in the placebo followed up to day 90 after randomization.
group.
b
Calculated as weight in kilograms divided by height in meters squared.
c
Outcomes
Kidney function was defined according to estimated glomerular filtration rate
(eGFR, calculated as milliliters per minute per 1.73 m2) as follows: normal
The primary efficacy outcome was the proportion of patients with
kidney function (baseline eGFR ⱖ90), mild kidney impairment (60 ⱕ baseline an mRS score of 1 or less on day 90 after randomization. The sec-
eGFR <90), and moderate to severe kidney impairment (baseline eGFR <60). ondary outcomes included mRS score on day 90, the proportion
of patients with an mRS score of 2 or less on day 90, the change
33 centers in China between June 28, 2021, and August 10, in NIHSS score from baseline to day 14, and the proportion of
2022. The ethics committee from each study center ap- patients with an NIHSS score of 1 or less on day 14, 30, and 90 af-
proved this study, and all patients or their legally acceptable ter randomization. Safety outcomes included adverse events,
surrogates had given informed consent before they were as- treatment-related adverse events within 90 days, and changes
signed to a treatment group. The study was registered with in vital signs and laboratory data before and after treatment.
ClinicalTrials.gov. Written informed consent for participa-
tion in the trial was provided by the patients or their legal rep- Statistical Analysis
resentative. Details of the trial rationale, design, and meth- Assuming that the proportion of patients with an mRS score
ods have been described previously and are provided in the of 1 or less on day 90 was 50% in the sublingual edaravone

jamaneurology.com (Reprinted) JAMA Neurology Published online February 19, 2024 E3

Downloaded from jamanetwork.com by guest on 02/24/2024

 Research Original Investigation Sublingual Edaravone Dexborneol in Acute Ischemic Stroke

Table 2. Efficacy and Safety Outcomes

Edaravone
dexborneol group Placebo group
Outcomes (n = 450) (n = 464) Effect size (95% CI) P value
Primary outcome
mRS score ≤1 on day 90, No. (%)a,b 290 (64.4) 254 (54.7) RD 9.70 (3.37 to 16.03) .003
OR 1.50 (1.15 to 1.95)
Secondary outcomes
mRS score on day 90, median (IQR) 1 (0-2) 1 (1-3) Common OR 1.33 (1.05 to 1.68) .02
mRS score ≤2 on day 90, No. (%) 342 (76.0) 337 (72.6) RD 3.37 (−2.29 to 9.03) .24
OR 1.19 (0.89 to 1.61)
Changes of NIHSS score from baseline −3 (−4 to −3) −3 (−4 to −3) MD −0.11 (−0.49 to 0.27) .58
to day 14, mean (95% CI)a
NIHSS score ≤1 on day 14, No. (%)a 63 (15.0) 71 (16.2) RD −1.25 (−6.10 to 3.60) .62
OR 0.91 (0.63 to 1.32)
NIHSS score ≤1 on day 30, No. (%)a 146 (35.9) 136 (32.2) RD 3.64 (−2.80 to 10.09) .27
OR 1.18 (0.88 to 1.57)
a
NIHSS score ≤1 on day 90, No. (%) 230 (57.2) 215 (51.6) RD 5.66 (−1.16 to 12.47) .11
OR 1.26 (0.95 to 1.65)
Safety outcomes
AE within 90 d, No. (%) 405 (89.8) 418 (90.1) RD −0.31 (−4.21 to 3.59) .88
OR 0.97 (0.63 to 1.49)
TRAE within 90 d, No. (%) 61 (13.6) 50 (10.8) RD 2.78 (−1.46 to 7.02) .20
OR 1.30 (0.87 to 1.93)
SAE within 90 d, No. (%) 42 (9.3) 47 (10.1) −0.80 (−4.64 to 3.05) .69
0.91 (0.59 to 1.42)
Abbreviations: AE, adverse event; MD, mean difference; mRS, modified Rankin measurement (24 and 10). The number of patients with missing data on NIHSS
Scale; NIHSS, National Institutes of Health Stroke Scale; OR, odds ratio; RD, risk score on day 14 was 29 in the edaravone dexborneol group and 26 in the
difference; SAE, severe adverse event; TRAE, treatment-related adverse event. placebo group; missing data on NIHSS score on day 30 was 43 in the
a
Missing data on mRS was considered to be 6 for patients without at least 1 edaravone dexborneol group and 42 in the placebo group; missing data on
postbaseline measurement of mRS (8 patients in the edaravone dexborneol NIHSS score on day 90 was 48 in the edaravone dexborneol group and 47 in
group and 8 in the placebo group) or discontinued treatment or follow-up due the placebo group.
b
to adverse events (8 and 14), and the last observation carried forward A total of 20 patients had telephone assessment of the mRS, and 64 had video
approach was used for patients who had at least 1 valid postbaseline assessment of the mRS.

dexborneol group and 40% in the placebo group, a 2-sided α
of .05, power of 80%, and dropout rate of 15% were used to
estimate the total sample size. One prespecified interim analy-
sis for sample size re-estimation by an independent data moni-
toring committee was conducted according to conditional
power calculations as indicated by the promising zone method
when 50% of patients were enrolled, and no need for an adap-
tive increase in sample size was found.
All the analyses were performed in the intention-to-treat
population. Baseline characteristics were presented as median
with IQR for continuous variables and frequency with propor-
tion for categorical variables. Missing data on the primary out-
come were handled with treatment policy strategy, composite
variable strategies, and while on treatment strategies.11 Group
difference in the primary efficacy outcome was calculated, and
the corresponding 95% CIs of the difference between proportions
were estimated based on normal approximation. Odds ratios
(ORs) with 95% CIs were calculated using logistic regression.
Similar approaches were used for binary secondary out-
comes, including mRS score of 2 or less on day 90; NIHSS score
of 1 or less on day 14, 30, and 90; and safety outcomes on ad-
verse events and treatment-related adverse events. For mRS
score on day 90, an ordinal logistic regression analysis was per-
formed, with the results presented as common OR and 95%
CI, where a common OR in favor of sublingual edaravone dex-
borneol was greater than 1.0. For changes in NIHSS score from
baseline to day 14, mean values with 95% CIs were calculated
for each group, and mean differences with 95% CI between the
groups were estimated by analysis of covariance. In addition,
a post hoc sensitivity analysis was performed using different
approaches to impute missing data on the primary efficacy out-
come and complete case analysis. Finally, the heterogeneity
of treatment effects on the primary outcome among several
prespecified subgroups was evaluated by including the inter-
action term between treatment and subgroup effect into the
logistic regression model.
All tests were 2-sided, and P < .05 was considered statis-
tically significant. Statistical analyses were performed with SAS
software, version 9.4 (SAS Institute).
Results
Baseline Characteristics
Among 956 patients screened, 914 patients (median [IQR] age,
64.0 [56.0-70.0] years; 608 male [66.5%]; 306 female [33.5%])
were randomized, among whom 450 patients (49.2%) received
sublingual edaravone dexborneol, and 464 patients (50.8%) re-
ceived placebo (Figure 1). Baseline characteristics of the patients
were similar in the 2 groups (Table 1). A total of 838 patients
(91.7%) had an mRS score of 0. Concomitant treatment taken dur-
ing the treatment period is reported in eTable 2 in Supplement 2.
Among the treated patients, 793 patients (390 [49.2%] in the
edaravone dexborneol group and 403 [50.8%] in the placebo
group) did not have major violations of the study protocol (eg,
inappropriate enrollment, premature discontinuation of trial

E4 JAMA Neurology Published online February 19, 2024 (Reprinted) jamaneurology.com

Downloaded from jamanetwork.com by guest on 02/24/2024

 Sublingual Edaravone Dexborneol in Acute Ischemic Stroke
Figure 2. Distribution of Functional Outcomes at 90 Days
in the Intention-to-Treat Population
Modified Rankin Scale score at 90 d
0 1 2 3 4 5
Edaravone
28 37 12 12
dexborneol
Placebo 24 30 18 11
0 10 20 30 40 50 60 70 80
Patients, %
Shown are scores on the modified Rankin Scale (mRS) for the patients in the 2
treatment groups. Scores range from 0 to 6, with 0 indicating no symptoms;
1, no clinically significant disability; 2, slight disability (patients are able to look
after their own affairs without assistance but are unable to carry out all previous
activities); 3, moderate disability (patients require some help but are able to
walk unassisted); 4, moderately severe disability (patients are unable to attend
to bodily needs without assistance and are unable to walk unassisted); 5, severe
disability (patients require constant nursing care and attention); and 6, death.
Missing data on mRS was considered to be 6 for patients without at least 1
postbaseline measurement of mRS (8 patients in the edaravone dexborneol
group, and 8 in the placebo group), or discontinued treatment or follow-up due
to adverse events (8 and 14), and the last observation carried forward approach
was used for patients who had at least 1 valid postbaseline measurement
(24 and 10).
treatment, concomitant use of prohibited medications [drugs
that may interact with the trial drug as described in the proto-
col], or were outside of the study window or lost to 90-day follow-
up) and therefore were included in the per-protocol analysis
(Figure 1).
Efficacy Outcomes
In the primary analysis population, missing data on mRS was
considered to be 6 for patients without at least 1 postbaseline
measurement of mRS (8 patients in the edaravone dexbor-
neol group and 8 in the placebo group) or discontinued treat-
ment or follow-up due to adverse events (8 and 14), and the
last observation carried forward approach was used for pa-
tients who had at least 1 valid postbaseline measurement (24
and 10). A favorable outcome of an mRS score of 1 or less on
day 90 occurred in 290 of 450 patients (64.4%) in the edara-
vone dexborneol group and in 254 of 464 patients (54.7%) in
the placebo group (risk difference, 9.70%; 95% CI, 3.37%-
16.03%; OR, 1.50; 95% CI, 1.15-1.95, P = .003) (Table 2). Sen-
sitivity analysis with different approaches to imputation on
missing primary data (eTable 3 in Supplement 2) and the per-
protocol analysis (eTable 4 in Supplement 2) showed similar
results.
For the secondary outcomes, an ordinal comparison of the
distribution of patients across mRS categories showed that the
good function outcome favored the edaravone dexborneol
group (common OR, 1.33; 95% CI, 1.05-1.68; P = .02) (Table 2
and Figure 2). However, edaravone dexborneol had no effect
on other prespecified secondary outcomes, including the pro-
portion of patients with an mRS score of 2 or less, the differ-
ence between the groups in the change of NIHSS score from
jamaneurology.com
Downloaded from jamanetwork.com by guest on 02/24/2024
Original Investigation Research
baseline to day 14, and the proportion of NIHSS score of 1 or
less on day 14, 30, and 90 (Table 2 and eTable 3 in Supple-
ment 2). The results of the subgroup analyses for the primary
outcome suggest that the effect of sublingual edaravone dex-
6 borneol was consistent across multiple subgroups (Figure 3).
1
Safety Outcomes
6 4
Adverse events occurred in 405 patients (89.8%) in the edara-
vone dexborneol group and in 418 patients (90.1%) in the pla-
9 7
cebo group (risk difference, −0.31%; 95% CI, −4.21% to 3.59%;
1 OR, 0.99; 95% CI, 0.64-1.53) (Table 2). Treatment-related ad-
90 100 verse events occurred in 61 patients (13.6%) in the edaravone
dexborneol group and in 50 patients (10.8%) in the placebo
group (risk difference, 2.78%; 95% CI, −1.46% to 7.02%; OR,
1.30; 95% CI, 0.87-1.93) (Table 2). Detailed information on ad-
verse events is presented in eTables 5 to 12 in Supplement 2.
Discussion
In this double-blind, placebo-controlled, randomized clini-
cal trial involving patients with AIS within 48 hours after symp-
tom onset, sublingual edaravone dexborneol could improve
the proportion of patients achieving a good functional out-
come on day 90 after randomization without increasing risks
of any adverse events.
Despite compelling evidence for the potential benefit of
brain cytoprotection observed in the preclinical animal mod-
els, most of the previously published clinical trials targeting
brain cytoprotection failed to show significant clinical effi-
cacy for patients with AIS during the last decade. The Stroke-
Acute Ischemic NXY Treatment (SAINT) I and II trials, for in-
stance, suggested that NXY-059, a free radical–trapping agent,
was ineffective for the treatment of AIS within 6 hours of symp-
tom onset.12 Similar nonsignificant efficacy was also found for
uric acid in the Safety and Efficacy of Uric Acid in Patients with
Acute Stroke (URICO-ICTUS) trials,13 for magnesium sulfate in
the Field Administration of Stroke Therapy-Magnesium (FAST-
MAG) trial,14 for intravenous albumin in the Albumin in Acute
Ischemic Stroke (ALIAS) trials,15 and for natalizumab in the
Safety and Efficacy of Natalizumab in Patients with Acute
Ischemic Stroke (ACTION) trial,16 respectively. However, the
ESCAPE-NA14 and TASTE trials reported a better 90-day good
functional outcome for patients treated with brain cytopro-
tective agents.8 These trials indicated that although no sig-
nificant clinical benefits were found for most brain cytopro-
tective agents developed based on the single-pathway strategy,
treatments targeting several pathways of ischemic injury
seemed to be more promising due to the complicated mecha-
nism of injury in ischemic stroke.
Sublingual edaravone dexborneol is a brain cytopro-
tective agent composed of edaravone and dexborneol, 2
active ingredients with synergistic antioxidant and anti-
inflammatory effects. Edaravone is an effective oxygen radi-
cal scavenger that has long been recommended for AIS
therapy in China and Japan.17-19 Laboratory and clinical evi-
dence have indicated that edaravone could lessen neuronal
damage and endothelial cell injury through inhibition of
(Reprinted) JAMA Neurology Published online February 19, 2024 E5
 Research Original Investigation Sublingual Edaravone Dexborneol in Acute Ischemic Stroke

Figure 3. Subgroup Analysis for the Primary Analysis

Edaravone Placebo Edaravone P for

Subgroup dexborneol Placebo OR (95% CI) better dexborneol better interaction
Overall 290 (64.4) 254 (54.7) 1.50 (1.15-1.95)
Age, y
≤65 165 (67.4) 142 (59.4) 1.41 (0.97-2.04)
.68
>65 125 (60.9) 112 (49.8) 1.58 (1.07-2.31)
Sex
Male 89 (63.1) 86 (52.1) 1.57 (0.99-2.49)
.78
Female 201 (65.1) 168 (56.2) 1.45 (1.05-2.01)
Time from onset to treatment, h
≤24 125 (63.8) 105 (51.7) 1.64 (1.10-2.45)
.55
>24 165 (65.9) 149 (57.1) 1.39 (0.98-1.99)
mRS score prior to onset
0 273 (65.8) 236 (55.8) 1.52 (1.15-2.01)
.63
1 17 (48.6) 18 (43.9) 1.21 (0.49-2.98)
NIHSS score
≥7 202 (76.5) 183 (70.7) 1.35 (0.92-2.00)
.43
>7 88 (47.3) 71 (34.6) 1.69 (1.13-2.55)
History of stroke
No 221 (67.8) 187 (58.3) 1.51 (1.09-2.08)
.84
Yes 69 (55.6) 67 (46.8) 1.42 (0.88-2.31)
History of hypertension
No 63 (75.9) 59 (60.2) 2.08 (1.08-3.97)
.29
Yes 227 (61.9) 195 (53.3) 1.42 (1.06-1.91)
History of hyperlipidemia
No 176 (66.2) 161 (58.1) 1.41 (0.99-2.00)
.46
Yes 114 (61.9) 93 (49.7) 1.65 (1.09-2.49)
History of diabetes
No 154 (66.1) 143 (58.9) 1.36 (0.94-1.98)
.57
Yes 136 (62.7) 111 (50.2) 1.66 (1.14-2.44)
History of heart disease
No 209 (65.7) 188 (57.3) 1.43 (1.04-1.96)
.58
Yes 81 (61.4) 66 (48.5) 1.68 (1.04-2.74)
TOAST classification of stroke
Large-artery atherosclerosis 153 (61.9) 126 (51.4) 1.54 (1.07-2.20)
Cardioembolism 6 (60.0) 9 (56.3) 1.17 (0.23-5.81)
Small-vessel occlusion 122 (68.5) 110 (59.5) 1.49 (0.96-2.29) .99
Other determined etiology 5 (71.4) 7 (63.6) 1.43 (0.18-11.09)
Undetermined etiology 1 (100.0) 1 (100.0) 1.43 (0.18-11.09)
Kidney function
Normal 258 (64.7) 235 (56.8) 1.39 (1.05-1.85)
.44
Mildly impaired 29 (61.7) 19 (39.6) 2.46 (1.08-5.61)
Moderately to severely impaired 3 (75.0) 0 (0.0)
0 1 2 3 4 5 6
OR (95% CI)

mRS indicates modified Rankin Scale; NIHSS, National Institutes of Health Stroke Scale; OR, odds ratio; TOAST, Trial of Org 10172 in Acute Stroke Treatment.

neurotoxicity, reduction of chronic inflammation, and regu-
lation of the expression of endothelial and neuronal protein
in the ischemic cascade progress. 6,20-25 Dexborneol, the
other ingredient of sublingual edaravone dexborneol, serves
as an important component of proprietary Chinese medicine
for the treatment of stroke. It was proved to prevent neuronal
injury after cerebral ischemia via multiple action mecha-
nisms, including improvement of cerebral blood flow, inhibi-
tion of neuronal excitotoxicity, resistance to reactive oxygen
species injury, and inhibition of inflammatory processes
and caspase-related apoptosis. On the other hand, the high
permeability of dexborneol could promote other agents to
pass through the blood-brain barrier to exert synergistic
therapeutic effects.26 Accordingly, the preclinical trial has
shown that edaravone dexborneol was more effective in pro-
tecting the brain from ischemic and/or ischemic reperfusion
injury, thereby avoiding the potential deficiency of those
single-target agents.5 Through these mechanisms, our study
showed that sublingual edaravone dexborneol improved the
favorable outcome, defined as an mRS score of 0 to 1 after 90
days of randomization, for patients with AIS.
Sublingual agents can quickly disintegrate once in con-
tact with the saliva and before being swallowed, which leads
to a high drug concentration in situ and rapid absorption
through the highly permeable sublingual mucosa.27,28 Phar-
macokinetic studies have shown that sublingual formulation

E6 JAMA Neurology Published online February 19, 2024 (Reprinted) jamaneurology.com

Downloaded from jamanetwork.com by guest on 02/24/2024

 Sublingual Edaravone Dexborneol in Acute Ischemic Stroke Original Investigation Research

rapidly diffused and was absorbed across the buccal mem-
brane without interfering with disposition and elimination
properties and without significantly altering the total bioavail-
ability of edaravone and dexborneol. Consequently, the time
to peak plasma concentration and the area under the curve re-
main largely unchanged. Sublingual edaravone dexborneol is
the first, to our knowledge, sublingual brain cytoprotective
agent administered to patients with AIS and provides several
clinical advantages compared with intravenous drugs, includ-
ing faster onset of action, lower dose requirement, better pa-
tient compliance and convenience, and increased bioavail-
ability. As acknowledged, time to treatment is a criteria factor
in the odds of achieving good outcomes after stroke; acute-
stage therapies should be administered immediately once a
stroke has been identified. Sublingual edaravone dexbor-
neol, as a rapidly acting, patient-friendly brain cytoprotec-
tion, showed good efficacy on functional outcomes in pa-
tients with AIS in this trial. Taken together, sublingual
edaravone dexborneol could be administrated earlier to pa-
tients with AIS in future clinical practice regardless of health
care resources, even for those who are in a coma, experience
dysphagia, are at home, or are en route to the hospital in the
ambulance.
Limitations
There were several limitations in our study. First, we did not
compare the effect of sublingual edaravone dexborneol and
intravenous edaravone dexborneol. However, the total
bioavailability of sublingual edaravone dexborneol and the
edaravone dexborneol injection was similar according to
pharmacokinetic data. In addition, considering the same
components of the drugs and the blinding settings, sublin-
gual placebo as the control group may be a better choice.
Second, patients who received endovascular therapy were
excluded from this study. In recent years, given the rapid
development progress in the field of endovascular therapy,
the effect of sublingual edaravone dexborneol in patients
receiving endovascular therapy should be investigated in
future investigations. Third, our study enrolled a substantial
number of patients with a relatively low NIHSS score (mild
stroke), which resulted in nonsignificant results for the sec-
ondary outcomes. Fourth, some information was not avail-
able in our trial, such as results of the EuroQol Health Ques-
tionnaire and hemorrhagic event data; therefore, further
investigations are warranted to investigate the effect of sub-
lingual edaravone dexborneol on these outcomes. Finally,
this trial was conducted among Chinese patients, thus the
findings should be validated in other ethnic groups.
Conclusions
In this randomized clinical trial involving patients with AIS
treated within 48 hours after symptom onset, sublingual edara-
vone dexborneol provided brain cytoprotection by improv-
ing functional outcomes after 90 days of randomization.

ARTICLE INFORMATION
Accepted for Publication: November 11, 2023.
Published Online: February 19, 2024.
doi:10.1001/jamaneurol.2023.5716
Open Access: This is an open access article
distributed under the terms of the CC-BY License.
© 2024 Fu Y et al. JAMA Neurology.
Author Affiliations: Department of Neurology,
Peking University Third Hospital, Beijing, China (Fu,
Fan); Beijing Neurosurgical Institute, Capital
Medical University, Beijing, China (A. Wang); State
Key Laboratory of Neurology and Oncology Drug
Development, Nanjing, China (Tang, Ren, Zhu,
Feng); Department of Neurology, Beijing Tiantan
Hospital, Capital Medical University, Beijing, China
(S. Li, Xia); China National Clinical Research Center
for Neurological Diseases, Beijing Tiantan Hospital,
Capital Medical University, Beijing, China (S. Li, Xia);
Department of Clinical Epidemiology and Clinical
Trial, Capital Medical University, Beijing, China
(Tian); Beijing Municipal Key Laboratory of Clinical
Epidemiology, Beijing, China (Tian); Simcere
Pharmaceutical Group Limited, Nanjing, China (Ren,
Zhu, Feng); Neurodawn Pharmaceutical Co Ltd,
Nanjing, China (Yang, Chen, Yao); Harrision
International Peace Hospital, Hengshui, China
(Wei); Daqing Oilfied General Hospital, Daqing,
China (X. Dong); Nanshi Hospital of Nanyang,
Nanyang, China (Ling); Pingxiang People’s Hospital,
Pingxiang, China (Yi); The First Affiliated Hospital of
Nanyang Medical College, Nanyang, China (Deng);
Liaocheng People’s Hospital, Liaocheng, China
(Guo); The First People’s Hospital of Shenyang,
Shenyang, China (Sui); Mei He Kou Central Hospital,
Jilin, China (Han); Hainan General Hospital, Hainan,
China (Wen); Xuzhou Central Hospital, Jiangsu,
China (C. Li); Cangzhou Central Hospital, Hebei,
China (A. Dong); The First Hospital of Jilin
University, Jilin, China (Sun); Taizhou First People’s
Hospital, Zhejiang, China (Z. Wang); Anqing
Municipal Hospital, Anhui, China (Shi); The First
Affiliated Hospital Baotou Medical College, Baotou,
China (Liu).
Author Contributions: Dr Fan had full access to all
of the data in the study and takes responsibility for
the integrity of the data and the accuracy of the
data analysis. Drs Fu, A. Wang, and Tang
contributed equally to this work.
Concept and design: Fu, A. Wang, Tang, Yang, Chen,
Zhu, Feng, Yao, Wen, C. Li, Sun, Z. Wang, Liu, Fan.
Acquisition, analysis, or interpretation of data: Fu,
A. Wang, S. Li, Tian, Xia, Ren, Yang, Chen, Zhu, Yao,
Wei, X. Dong, Ling, Yi, Deng, Guo, Sui, Han,
A. Dong, Shi, Fan.
Drafting of the manuscript: Fu, A. Wang, Xia, Fan.
Critical review of the manuscript for important
intellectual content: Fu, A. Wang, Tang, S. Li, Tian,
Ren, Yang, Chen, Zhu, Feng, Yao, Wei, X. Dong, Ling,
Yi, Deng, Guo, Sui, Han, Wen, C. Li, A. Dong, Sun,
Z. Wang, Shi, Liu, Fan.
Statistical analysis: Fu, A. Wang, Tian, Xia, Wei.
Obtained funding: Yang, Chen, Yao.
Administrative, technical, or material support: Fu,
A. Wang, Tang, S. Li, Ren, Yang, Chen, Zhu, Feng,
Yao, X. Dong, Ling, Yi, Guo, Sui, Han, Wen, C. Li, A.
Dong, Sun, Z. Wang, Shi, Liu, Fan.
Supervision: Fu, A. Wang, Zhu, Deng, Fan.
Conflict of Interest Disclosures: Drs Tang, Ren,
Zhu, and Feng reported being employees of
Simcere Pharmaceutical Group. Drs Yang and Chen
reported being employees of Neurodawn
Pharmaceutical. No other disclosures were
reported.
Funding/Support: This trial was sponsored and
funded by grant SIM1911 from Simcere
Pharmaceutical and grant 2022YFA1303000 from
National Key R&D Program of China.
Role of the Funder/Sponsor: The funders had
no role in the design and conduct of the study;
collection, management, analysis, and
interpretation of the data; preparation, review, or
approval of the manuscript; and decision to submit
the manuscript for publication.
Data Sharing Statement: See Supplement 3.
Additional Contributions: We thank all study
participants, their relatives, and the members of
the survey teams at the study centers of the trial.
No one received financial compensation for their
contributions.
REFERENCES
1. Ramos-Cabrer P, Campos F, Sobrino T, Castillo J.
Targeting the ischemic penumbra. Stroke. 2011;42
(1)(suppl):S7-S11.
2. Savitz SI, Baron JC, Fisher M; STAIR X
Consortium. Stroke treatment academic industry
roundtable X: brain cytoprotection therapies in the
reperfusion era. Stroke. 2019;50(4):1026-1031.
doi:10.1161/STROKEAHA.118.023927
3. Lyden P, Buchan A, Boltze J, Fisher M; STAIR XI
Consortium*. Top priorities for cerebroprotective
studies—a paradigm shift: report from STAIR XI.

jamaneurology.com (Reprinted) JAMA Neurology Published online February 19, 2024 E7

Downloaded from jamanetwork.com by guest on 02/24/2024

 Research Original Investigation
Stroke. 2021;52(9):3063-3071. doi:10.1161/
STROKEAHA.121.034947
4. Hill MD, Goyal M, Menon BK, et al; ESCAPE-NA1
Investigators. Efficacy and safety of nerinetide for
the treatment of acute ischaemic stroke
(ESCAPE-NA1): a multicentre, double-blind,
randomised controlled trial. Lancet. 2020;395
(10227):878-887. doi:10.1016/S0140-6736(20)
30258-0
5. Wu HY, Tang Y, Gao LY, et al. The synergetic
effect of edaravone and borneol in the rat model of
ischemic stroke. Eur J Pharmacol. 2014;740:522-531.
doi:10.1016/j.ejphar.2014.06.035
6. Zhang N, Komine-Kobayashi M, Tanaka R, Liu M,
Mizuno Y, Urabe T. Edaravone reduces early
accumulation of oxidative products and sequential
inflammatory responses after transient focal
ischemia in mice brain. Stroke. 2005;36(10):2220-
2225. doi:10.1161/01.STR.0000182241.07096.06
7. Liu R, Zhang L, Lan X, et al. Protection by
borneol on cortical neurons against oxygen-glucose
deprivation/reperfusion: involvement of
anti-oxidation and anti-inflammation through
nuclear transcription factor κappaB signaling
pathway. Neuroscience. 2011;176:408-419.
doi:10.1016/j.neuroscience.2010.11.029
8. Xu J, Wang A, Meng X, et al; TASTE Trial
Investigators. Edaravone dexborneol vs edaravone
alone for the treatment of acute ischemic stroke:
a phase 3, randomized, double-blind, comparative
trial. Stroke. 2021;52(3):772-780. doi:10.1161/
STROKEAHA.120.031197
9. Xu J, Wang Y, Wang A, et al. Safety and efficacy
of edaravone dexborneol vs edaravone for patients
with acute ischaemic stroke: a phase 2, multicentre,
randomised, double-blind, multiple-dose,
active-controlled clinical trial. Stroke Vasc Neurol.
2019;4(3):109-114. doi:10.1136/svn-2018-000221
10. Fu Y, Tang R, Chen R, et al. Efficacy and safety
of Y-2 sublingual tablet for patients with acute
ischaemic stroke: protocol of a phase 3 randomised
double-blind placebo-controlled multicentre trial.
Stroke Vasc Neurol. Published online June 12, 2023.
doi:10.1136/svn-2022-002014
11. European Medicines Agency. ICH E9 (R1)
addendum on estimands and sensitivity analysis in
clinical trials to the guideline on statistical principles
for clinical trials. Accessed September 7, 2022.
E8 JAMA Neurology Published online February 19, 2024 (Reprinted)
Downloaded from jamanetwork.com by guest on 02/24/2024
Sublingual Edaravone Dexborneol in Acute Ischemic Stroke
https://www.ema.europa.eu/en/documents/
scientific-guideline/ich-e9-r1-addendum-
estimands-sensitivity-analysis-clinical-trials-
guideline-statistical-principles_en.pdf.
12. Diener HC, Lees KR, Lyden P, et al; SAINT I and II
Investigators. NXY-059 for the treatment of acute
stroke: pooled analysis of the SAINT I and II Trials.
Stroke. 2008;39(6):1751-1758. doi:10.1161/
STROKEAHA.107.503334
13. Chamorro A, Amaro S, Castellanos M, et al;
URICO-ICTUS Investigators. Safety and efficacy of
uric acid in patients with acute stroke
(URICO-ICTUS): a randomised, double-blind phase
2b/3 trial. Lancet Neurol. 2014;13(5):453-460.
doi:10.1016/S1474-4422(14)70054-7
14. Saver JL, Starkman S, Eckstein M, et al;
FAST-MAG Investigators and Coordinators.
Prehospital use of magnesium sulfate as
neuroprotection in acute stroke. N Engl J Med.
2015;372(6):528-536. doi:10.1056/NEJMoa1408827
15. Martin RH, Yeatts SD, Hill MD, Moy CS, Ginsberg
MD, Palesch YY; ALIAS Parts 1 and 2 and NETT
Investigators. ALIAS (albumin in acute ischemic
stroke) trials: analysis of the combined data from
parts 1 and 2. Stroke. 2016;47(9):2355-2359.
doi:10.1161/STROKEAHA.116.012825
16. Elkins J, Veltkamp R, Montaner J, et al. Safety
and efficacy of natalizumab in patients with acute
ischaemic stroke (ACTION): a randomised,
placebo-controlled, double-blind phase 2 trial.
Lancet Neurol. 2017;16(3):217-226. doi:10.1016/
S1474-4422(16)30357-X
17. Wang YJ, Zhang SM, Zhang L, et al. Chinese
guidelines for the secondary prevention of ischemic
stroke and transient ischemic attack 2010. CNS
Neurosci Ther. 2012;18(2):93-101. doi:10.1111/j.1755-
5949.2011.00290.x
18. Wang Y, Liu M, Pu C. 2014 Chinese guidelines
for secondary prevention of ischemic stroke and
transient ischemic attack. Int J Stroke. 2017;12(3):
302-320. doi:10.1177/1747493017694391
19. Ishihara H, Suzuki M. Japanese guidelines for
the management of stroke 2015: overview of the
chapter on subarachnoid hemorrhage. Article in
Japanese. Nihon Rinsho. 2016;74(4):677-680.
20. Edaravone Acute Infarction Study Group.
Effect of a novel free radical scavenger, edaravone
(MCI-186), on acute brain infarction: randomized,
placebo-controlled, double-blind study at
multicenters. Cerebrovasc Dis. 2003;15(3):222-229.
doi:10.1159/000069318
21. Cheng B, Guo Y, Li C, et al. Edaravone protected
PC12 cells against MPP(+)-cytoxicity via inhibiting
oxidative stress and up-regulating heme
oxygenase-1 expression. J Neurol Sci. 2014;343(1-2):
115-119. doi:10.1016/j.jns.2014.05.051
22. Matsumoto S, Murozono M, Kanazawa M, Nara
T, Ozawa T, Watanabe Y. Edaravone and
cyclosporine A as neuroprotective agents for acute
ischemic stroke. Acute Med Surg. 2018;5(3):213-221.
doi:10.1002/ams2.343
23. Mizuno A, Umemura K, Nakashima M.
Inhibitory effect of MCI-186, a free radical
scavenger, on cerebral ischemia following rat
middle cerebral artery occlusion. Gen Pharmacol.
1998;30(4):575-578. doi:10.1016/S0306-3623(97)
00311-X
24. Wada T, Yasunaga H, Inokuchi R, et al. Effects
of edaravone on early outcomes in acute ischemic
stroke patients treated with recombinant tissue
plasminogen activator. J Neurol Sci. 2014;345(1-2):
106-111. doi:10.1016/j.jns.2014.07.018
25. Watanabe T, Tanaka M, Watanabe K, Takamatsu
Y, Tobe A. Research and development of the free
radical scavenger edaravone as a neuroprotectant.
Article in Japanese. Yakugaku Zasshi. 2004;124
(3):99-111. doi:10.1248/yakushi.124.99
26. Li Y, Ren M, Wang J, et al. Progress in borneol
intervention for ischemic stroke: a systematic
review. Front Pharmacol. 2021;12:606682.
doi:10.3389/fphar.2021.606682
27. Okuda Y, Irisawa Y, Okimoto K, Osawa T,
Yamashita S. A new formulation for orally
disintegrating tablets using a suspension
spray-coating method. Int J Pharm. 2009;382(1-2):
80-87. doi:10.1016/j.ijpharm.2009.08.010
28. AlAli AS, Aldawsari MF, Alalaiwe A, et al.
Exploitation of design-of-experiment approach for
design and optimization of fast-disintegrating
tablets for sublingual delivery of sildenafil citrate
with enhanced bioavailability using fluid-bed
granulation technique. Pharmaceutics. 2021;13(6):
870. doi:10.3390/pharmaceutics13060870
jamaneurology.com