Evidence of Improvements to Arterial Stiffness Among Regular Users of Combustible Cigarettes – Effect of Inhalation of β-Caryophyllene: A Randomized, Double-Blind, Placebo-Controlled Study

This was a randomized, double-blind, parallel-group study with a 1:1 allocation ratio between the BCP and placebo groups. The primary endpoint was the difference in baPWV change at 12 weeks between the BCP with placebo. Secondary endpoints included the safety of BCP capsules, among other items (Table S1).

The recruitment process (November 2, 2022, to May 29, 2023) was outsourced to a specialized company to enroll Japanese residents based on the criteria listed in Table S2. An additional table lists eligible cigarette brands whose tobacco was confirmed to not contain BCP (Table S3). After obtaining informed consent, screening tests (December 10, 2022, to May 29, 2023) (Table S4) were conducted at the Three-One Clinic and Watanabe Hospital in Tokyo, Japan. Participants who met the inclusion (Table S5) but not the exclusion (Table S6) criteria were enrolled from December 10, 2022, and followed up from March 4, 2022, to September 30, 2023. The inclusion and exclusion criteria, except for baPWV, were designed to select healthy volunteers. The lower limit of baPWV, 1,300 cm/s, was set based on the results of previous studies (20) to select subjects with moderately hardened arteries. Discontinuation criteria were also established (Table S7).

For eligible participants, a drug number that did not contain information regarding the treatment group was issued on the registration system built on an electronic data capture system, and the allocator was notified. Allocation to groups was performed using a stratified substitution block method with a BCP:Placebo ratio of 1:1 and a random block size of 2 or 4. Two allocation factors were used: ⚬

baPWV < 1,500 cm/s vs. ≥ 1,500 cm/s and

The allocation staff assigned drug numbers to capsules according to an allocation table that was inaccessible to the principal investigator and study participants. The allocation was outsourced to Stat Academy LLC (Kobe, Japan) and incorporated by the CMIC HealthCare Institute Co., Ltd. (Tokyo, Japan).

After the assignment, the participants and intervention providers remained blinded by presenting only randomized codes. These codes were managed electronically on electronic data capture and kept strictly confidential. At the end of the study, the principal investigator asked Satt Co., Ltd. (Tokyo, Japan), the institution in charge of data management, to open key and output codes.

BCP was supplied by Inabata Koryo Co., Ltd. (Osaka, Japan). Medium-chain triglycerides (MCT) were purchased from KAO (Tokyo, Japan). The capsules (3.4 mm diameter) were prepared by Sunsho Pharmaceutical Co., Ltd. (Fuji, Japan) using the dropping method. The liquid compositions for the placebo and BCP capsules were 19.30 mg MCT and 2.90 mg BCP with 16.40 mg MCT (15% BCP), respectively. The BCP amount per capsule was determined as described in Supplementary method 1.

Four weeks before the initiation date (day 0), both the capsule-attaching device and capsules (Figure S1) were provided to the participants, along with a written manual, during a visit to the study site. After breaking the capsule in the filter prior to smoking, participants in both groups smoked one cigarette with a capsule attached to a facility adjacent to the research institute, as described previously (20). Blood samples were collected in the research facility four times (5 mL each): before smoking and 10, 20, and 40 min after starting smoking. Blood plasma was obtained by centrifugation, frozen, and stored at −80 °C until analysis. BCP and nicotine in the blood were quantified as described in the Supplementary method 2.

Participants were asked to visit the clinic on the initiation date (day 0) and at weeks 4, 8, and 12. During each visit, including day 0, the tests listed in Table S8 were performed, including the original questionnaire survey (Table S9). In contrast, the SF-36 questionnaire (40, 41) was administered only on day 0 and week 12. Capsules and capsule-attaching devices were again provided to the participants in both groups on day 0. They prepared cigarettes they normally smoked and attached them to the respective capsules as required. The smoking of capsule cigarettes was discretionary. The number of cigarette boxes opened each day, the number of cigarettes remaining in these boxes, and the number of cigarettes smoked other than the designated cigarettes were recorded in a diary distributed to the participants.

Blood pressure, pulse pressure, heart rate, baPWV, and ankle-brachial index were measured using FORM-5 (Fukuda Colin Co., Ltd., Tokyo, Japan) or BP-203RPEIII (Omron Corporation, Kyoto, Japan). The variables % vital capacity and %FEV1.0 were measured using the SP-3700 COPD Lung Per (Fukuda Denshi Co., Ltd., Tokyo, Japan) or HI-205 T (Chest M.I., INC., Tokyo, Japan) spirometer. Height and weight were measured using the AD-6351 or AD-6228AP (both A&D Company, Limited, Tokyo, Japan) scales. Carotid ultrasonography was performed using Viamo SSA-640 A (Toshiba Corporation, Tokyo, Japan) or HD3 (Philips Japan, Ltd., Tokyo, Japan). Chest radiographs were captured using the NAOMI-2002 (RF Co., Ltd., Nagano, Japan) or KXO-32 S (Canon Medical Systems Corporation, Otawara, Japan) systems. Electrocardiograms were measured using Vasera VS-1500ATW or ECG-1350 (both Fukuda Denshi Co., Ltd.). Pregnancy tests were performed using an HCG Quick Checker Dip (Mizuho Medy Co., Ltd., Tosu, Japan). Blood and urine tests were outsourced to the LSI Medience Corporation (Tokyo, Japan), while the salivary cotinine test was outsourced to Hoken Kagaku, Inc. (Yokohama, Japan).

Statistical significance and power were set at 5% bilaterally and 80%, respectively. Based on the results of a pilot study (20), the baPWV reduction in the BCP group was 73 cm/s, whereas the increase in the placebo group was 3 cm/s, with a standard deviation of 117 cm/s. Assuming a 10% dropout rate, the target population was calculated as 90 participants. Interim analyses were not performed.

Comparisons between intervention groups for baPWV change up to week 12 were examined using a two-way repeated-measures analysis of covariance with group, time point, their interaction term, and baPWV on day 0 as independent variables. The SAS statistical analysis software suite (version 9.4_M7, SAS Institute Japan Ltd., Tokyo, Japan) was used to compare the above. Means and 95% confidence intervals were calculated. Within-group comparisons were performed on day 0 and week 12 for each group. Two-sided significance was set at 5%. Statistical analyses of the secondary endpoints and stratified analyses other than the questionnaire were performed similarly to those for the primary endpoint. The questionnaire, SF-36, and number of cigarettes smoked per week were tabulated at each time point. The Mann-Whitney U test was used for between-group comparisons, and Wilcoxon’s signed-rank test was used for before-after comparisons within each group. Spearman’s rank correlation test was used for correlation analysis. Multiple comparisons were corrected using Bonferroni correction. Statistical analyses were performed by Satt Co., Ltd.

The capsules contained 19.3 mg of liquid, with 2.54 mg of BCP per capsule in the BCP group, and no BCP was detected in the placebo group.

Of 292 applicants whose baPWV was larger than 1,300 cm/s, 208 were excluded because they did not meet all eligibility criteria, such as respiratory dysfunction, carotid artery intimal thickening, and hypertension. The number of people excluded by reason for screening is described in Table S10. Subjects were excluded from the study based on screening tests; therefore, on the day 0 test, which was conducted eight weeks after the screening test, a few people met the exclusion criteria. The remaining 84 participants were randomized to BCP (assigned: n = 42, intervention: n = 36, analyzed: n = 33) or placebo (as signed: n = 42, intervention: n = 39, analyzed: n = 36) groups. Only individuals not subjected to capsule cigarette smoking intervention were excluded from the analysis; all data from participants whose medical data could be measured were included in the analysis. After obtaining fixed data and during statistical analysis, no participant was further excluded from the analysis in either the BCP or the placebo group. Missing values were not imputed. The flowchart of the study procedure is shown in Figure 1.

Figure 1.

Flowchart of participant recruitment.

The baseline demographic and clinical characteristics of each group are presented in Table S11. A total of 208 individuals were excluded based on the exclusion criteria, leaving 84 participants (74 males and 10 females) with an average age of 48.9 ± 7.7 years old. Of these, 69 people were randomly assigned to either the BCP group (n = 36, 49.4 ± 8.7 years old) or the placebo group (n = 33, 49.9 ± 8.7 years old) for the commencement of a prospective trial. The major background factors did not differ significantly between the treatment groups. The cigarette brands smoked by individual participants with the nicotine and ‘tar’ amounts of these brands are listed in Table S3.

Figure 2 illustrates the blood concentrations of BCP and nicotine in each study group. In both groups, BCP concentrations were approximately 1 ng/mL before smoking. However, the BCP blood concentration reached a maximum concentration (Cmax) of 4.42 ng/mL at time-tomaximum (Tmax) of 10 min after starting to smoke in the BCP group. In contrast, no discernible increase in BCP blood concentration was detected in the placebo group (Table S12). The nicotine concentrations before smoking were 8.4 ng/mL in the BCP group and 5.2 ng/mL in the placebo group. The nicotine blood concentration curves at a Tmax of 10 min after starting smoking showed Cmax values of 15.64 ng/mL and 10.22 ng/mL in the BCP and placebo groups, respectively.

Figure 2.

Changes in BCP (a) and nicotine (b) blood concentrations in the placebo and BCP groups before and 10, 20, and 40 min after smoking a capsule-containing cigarette. Error bars represent the standard deviation. BCP = β-caryophyllene.

A significant increase in nicotine blood concentration was observed in both groups (Table S13). The average nicotine amount per cigarette was 0.55 mg in the placebo group and 0.59 mg in the BCP group. Furthermore, the number of cigarettes smoked per day was 32.42–33.31 in the placebo group and 33.54–35.06 in the BCP group (Supplementary data 1). No significant difference in the number of cigarettes smoked per day or the rate of change in the number of cigarettes smoked was observed between the two study groups at any time point.

The mean absolute baPWV values on day 0 and at weeks 4, 8, and 12 (Table S14), the mean changes in baPWV (Table S15), and the before-after and between-group comparisons for baPWV changes through week 12 (Figure 3a, Table S16) were analyzed. The least-squares means of baPWV changes were 9.0 cm/s and −3.8 cm/s in the placebo and BCP groups, respectively. In the analysis of all subjects, the within-group and between-group comparison of baPWV change from day 0 to 4, 8, and 12 weeks showed no significant change in either group. The results related to secondary endpoints are described in the additional results (Supplementary data 2).

Figure 3.

Changes in baPWV from day 0. (a) All participants. (b) Stratified analysis of participants with baPWV < 1,400 cm/s on day 0. (c) Stratified analysis of participants with baPWV ≥ 1,400 cm/s on day 0. Error bars represent standard deviations. *P < 0.05, within-group comparison vs. day 0; †P < 0.05: between-group comparison; baPWV: brachial-ankle pulse wave velocity; BCP: β-caryophyllene.

Although not included in the prior analysis plan, we performed exploratory stratified analyses of baPWV changes.

First, a stratified analysis was performed with a baPWV cutoff of 1,400 cm/s on day 0. Based on this, the absolute baPWV values on day 0 and at weeks 4, 8, and 12 (Table S17) and the before-after and between-group comparisons (Figures 3b, 3c, Table S18) were analyzed. In the group with a baPWV ≥ 1,400 cm/s on day 0, the within-group change in baPWV on day 0 did not differ significantly from that in the placebo group. However, significant differences were observed at weeks 4 (P = 0.003) and 8 (P = 0.038) in the BCP group (Figure 3c). The change in baPWV was also significantly different between the placebo and BCP groups at week 4 (P = 0.041; Figure 3c). Stratified analysis based on blood BCP levels was also performed, but no significant differences were found. Moreover, correlation analyses were performed between the initial baPWV value and the baPWV change at 4, 8, and 12 weeks in the BCP and placebo groups (Figure 4, Table S19, Figure S2). No significant correlation was observed in the placebo group. However, the correlation coefficients in the BCP group were −0.502 (P = 0.002) at week 4, −0.615 (P < 0.001) at week 8, and −0.513 (P = 0.002) at week 12, indicating significant negative correlations.

Figure 4.

Correlations between baPWV on day 0 and the 12-week change in baPWV in the BCP (a) and placebo (b) groups. baPWV: brachial-ankle pulse wave velocity; BCP: β-caryophyllene.

Additional tables show the rates of adverse events in the placebo (n = 36) and BCP (n = 39) groups (Table S20, Table S21).

No serious adverse events were observed in either group. The incidence of adverse reactions was 2.8% (1/36 participants) for “increased sputum” and “oropharyngeal discomfort” in the placebo group and 2.6% (1/39 participants) for “increased sputum” in the BCP group, with no significant difference in the incidence of adverse reactions between the placebo and BCP groups. All adverse reactions resolved without treatment.