Assessment of the Nicotine Pharmacokinetics When Using Two Types of E-Cigarettes in Healthy Adults Who Smoke: Results From Two Randomized, Crossover Studies

This was a randomized, open-label, six-period, six-sequence, crossover study conducted at a single clinic in Florida, USA, with 7 days of confinement. The study was approved by the Institutional Review Board (IRB) responsible for review and approval (Midlands IRB, Kansas, USA) and adhered to the ethical standards of the Declaration of Helsinki, applicable sections of the U.S. Code of Federal Regulations, and ICH E6 Good Clinical Practices. All participants gave written informed consent to participate in the study.

This was a randomized, open-label, seven-period, 14-sequence, crossover study conducted at a single clinic in Florida, USA, with 8 days of confinement. The study was approved by the IRB responsible for review and approval (Midlands IRB, Kansas, USA) and adhered to the ethical standards of the Declaration of Helsinki, applicable sections of the U.S. Code of Federal Regulations, and ICH E6 Good Clinical Practices. All participants gave written informed consent to participate in the study.

A commercially available cig-a-like type of e-cigarette device (eDNC1.0a, Logic Power manufactured by Logic Technology Development, LLC, Princeton, NJ, USA) with three flavor variants containing freebase nicotine was evaluated as the test product in this study. The eDNC1.0a consists of a replaceable cartomizer (cartridge) assembly containing e-liquid and an atomizer unit, and a rechargeable battery unit. Three cartridges with different flavors were tested; the cartridges were Product A: tobacco flavor, 2.4% nicotine (w/w); Product B: menthol flavor, 2.4% nicotine (w/w); and Product C: cherry flavor, 2.4% nicotine (w/w). The subjects’ own brand of commercial filter cigarettes (Product D) was used as a comparator. The U.S. FDA-approved nicotine replacement products, Nicorette® Gum, White Ice Mint, 2 mg (GlaxoSmithKline Consumer Healthcare, L.P., Brentford, UK: Product E) and a reference e-cigarette similar in composition and freebase nicotine content to the eDNC1.0a (Blu PLUS+ Tanks, Classic Tobacco (2.4% nicotine (w/w)) manufactured by Fontem US, LLC, Charlotte, NC, USA: Product F) were also evaluated.

A commercially available closed-tank e-cigarette device (eDNC2.0a, Logic Pro manufactured by Logic Technology Development, LLC) with four flavor variants containing freebase nicotine was evaluated as the test product in this study. The eDNC2.0a is a three-piece product consisting of a battery unit, a replaceable e-liquid capsule, and a capsule case. Four different flavored capsules were tested; the capsules were Product A: tobacco flavored, 1.8% nicotine (w/w); Product B: menthol flavored, 1.8% nicotine (w/w); Product C: cherry flavored, 1.4% nicotine (w/w); and Product D: berry mint flavored, 1.49% nicotine (w/w). The subjects’ own brand of commercially available filter cigarettes (Product E) was used as a comparator. The U.S. FDA-approved nicotine replacement products, Nicotrol® Inhaler, 10 mg per cartridge (Pfizer Inc., New York, NY, USA: Product F) and a reference e-cigarette similar in composition to the eDNC2.0a (Vuse Vise, Original (3.0% nicotine (freebase nicotine)(w/w)) manufactured by R.J. Reynolds Vapor Company, Winston-Salem, NC, USA: Product G) were also evaluated.

Subjects were randomly assigned to one of six product-use sequences based on a Williams design by using a computer-generated pseudorandom permutation procedure. On each assessment day (Days 1 to 6), subjects received a study product according to the randomized product-use sequence and used it once daily. On the morning of each day, subjects were asked to smoke a single cigarette, use either the test or reference e-cigarette ad libitum for 5 min, or use nicotine gum for 30 min according to package instructions. Follow-up was conducted in the week following discharge.

Subjects were randomly assigned to one of 14 product-use sequences based on a Williams design by using a computer-generated pseudorandom permutation procedure. On each assessment day (Days 1 to 7), subjects received a study product according to the randomized product-use sequence and used it once daily. In the morning of each day, subjects were asked to smoke a single cigarette, use either the test or reference e-cigarette ad libitum for 5 min, and use the nicotine inhaler for 20 min according to package instructions. Follow-up was conducted in the week following discharge.

Subjects’ baseline characteristics, including sex, age, race, BMI, daily cigarette consumption, and smoking history, the Fagerström Test for Cigarette Dependence (28, 29), were recorded at screening.

Before and after use of each test and reference e-cigarette, the weight of the e-cigarette (device and cartridge or capsule in situ) was measured to determine mass loss as the amount of e-liquid consumed by each subject. The theoretical nicotine consumption was estimated based on the amount of e-liquid used and the nicotine content (w/w%) of the e-liquid.

On each study day (Days 1 to 6 in Study 1, Days 1 to 7 in Study 2), before, during, and after product use, 14 blood samples were collected either by direct venipuncture or from a cannula placed in a forearm vein, for measurement of plasma nicotine concentrations, at the following times: 5 min before product use (at t = 0 min, time of start product use), and then at 1, 3, 5, 7, 10, 15, 30, 45, 60, 90, 120, 180, and 360 min. Blood samples were collected into K₂EDTA tubes, and the tubes were gently mixed by inverting 8 times. Within a maximum of 60 min from sample collection, tubes were centrifuged at approximately 1000–1300 relative centrifugal force at 5 °C for approximately 10 min. Plasma nicotine was analyzed by liquid chromatography-tandem mass spectrometry using validated analytical methods with appropriate quality controls at Celerion (Lincoln, NE, USA) (30). The lower limit of quantification (LLOQ) was determined to be 0.200 ng/mL. The following PK parameters were calculated from the individual plasma nicotine concentrations applying non-compartmental methods using Phoenix WinNonlin version 6.3 (Certara LP, Princeton, NJ, USA): the maximum observed plasma concentration (C_max), the area under the plasma concentration versus time curve from time zero to the last time point with a measurable concentration (AUC_0–last), and the time of maximum observed plasma concentration (T_max). When a baseline nicotine concentration above the LLOQ was observed, the PK parameters (C_max and AUC_0–last) were adjusted by subtracting the portion attributable to the baseline nicotine concentration by using the baseline concentration and primary terminal elimination rate constant obtained from each subject. Baseline-adjusted PK parameters were calculated according to the following equations: Cmax⁡=Cmax⁡,unadjusted−C0×eKel⋅Tmax⁡); and AUC0-last=AUC0-last,unadjusted−C0/Kel, \begin{array}{l} {\rm{C}}_{\max } = {\rm{C}}_{\max ,{\rm{unadjusted}}} - {\rm{C}}_0 \times {\rm{e}}^{{\rm{K}}_{{\rm{el}}} \cdot {\rm{T}}_{\max } } );\,{\rm{and}}\, \\ {\rm{AUC}}_{{\rm{0 - last}}} = {\rm{AUC}}_{{\rm{0 - last,unadjusted}}} - {\rm{C}}_0 /{\rm{K}}_{{\rm{el}}} {\rm{,}} \\ \end{array} where C₀ is the concentration before product use and K_el is the terminal elimination rate constant.

Three subjective effects questionnaires were administered on each study day (Days 1 to 6 in Study 1, Days 1 to 7 in Study 2). The product liking question (“Do you like the product effect?”) was administered as a 100-mm visual analog scale (VAS) ranging from “not at all (0 mm)” to “extremely (100 mm)” at 5, 15, 30, 45, 60, and 120 min after product use, and the percentage of subjects likely to prefer the product (VAS score > 50 to 100 mm) was evaluated from the observed maximum VAS score (E_max). The intent to use again question (“If given the opportunity, would you use this product again?”) was administered as a 100-mm line VAS rating scale with “Definitely would not (0 mm),” “Don’t care (50 mm),” to “Definitely would (100 mm)” at 60 min post product use, and the percentage of subjects likely to use the product again (VAS score > 50 to 100 mm) was evaluated. The modified Product Evaluation Scale (mPES) questionnaire, consisting of 12 questions, was administered at the end of product use with scoring on a 7-point scale (ranging from 1 (not at all) to 7 (extremely)) at completion of product use, and the following subscales scores were evaluated: Satisfaction (average score of 4 questions); Psychological reward (average score of 5 questions); Aversion (average score of 2 questions); and Relief (score of 1 question). The mPES questionnaire was adapted from the modified Cigarette Evaluation Scale (31) to evaluate all study products by replacing the words “smoking” and “cigarette” with “it”.

Safety outcomes, including an assessment of adverse events (AEs), were recorded. These included any abnormal clinical findings from vital signs, clinical laboratory tests, and physical examinations throughout the study period after randomization. A product-emergent AE (PEAE) was also identified as an AE that occurred after the first product use or that was present before the first product use and became more severe after the first product use.

Fifty-eight subjects were screened and randomized to one of six product-use sequences, with 52 subjects completing the study in accordance with the protocol (Table 2). Of the six subjects who withdrew from the study, two were physician-decided withdrawals and four withdrew from the study themselves. The physician-withdrawn subjects were withdrawn because of poor venous access or because of disruptive and argumentative behavior.

The mean plasma nicotine - concentration-time curve following single use of the study products is shown in Figure 1. The profile of the plasma nicotine concentration-time curves obtained after use of eDNC1.0a with three flavor variants (Products A, B, and C) was similar. Following the use of eDNC1.0a with three flavor variants (products A, B, and C, respectively), analysis of the geometric LS means showed that C_max values of 3.71, 3.44, and 4.18 ng/mL, were reached after 15.0 min of product use, and similar AUC_0–last values of 6.24, 5.62, and 7.04 hr × ng/mL were observed (Table 3). In comparison, the profile of the nicotine concentration-time curves obtained after the use of the usual brand of CC (Product D) showed higher concentrations throughout the 360-minute sampling period, with a higher C_max value (15.32 ng/mL) that was reached earlier (7 min), and a higher AUC_0–last value (21.09 hr × ng/mL) observed. The nicotine gum (Product E) differed from that of the test e-cigarettes in that the C_max value of 3.26 ng/mL was reached later (45 min), and a slightly higher AUC_0–last value (8.21 hr × ng/mL) was observed. The reference e-cigarette (Product F) was found to be similar to the test e-cigarettes in terms of the profile of the nicotine concentration-time curve and PK parameters. Nicotine consumption from a single use was estimated to be 0.68–0.84 mg by measuring the weight of e-liquid used, with no marked differences between the test and reference e-cigarettes (Products A, B, C, and F, shown in Table 3).

Figure 1.

Overall product liking, intent to use the product again, and subscales of mPES were markedly lower when using eDNC1.0a with three flavor variants (Products A, B, and C) compared to the usual brand of CC (Product D, shown in Table 4). These subjective measures were generally similar between the test e-cigarettes, nicotine gum (Product E), and reference e-cigarette (Product F).

Thirty AEs were reported by 22 of the 58 subjects. All AEs were mild or moderate in severity; 14 PEAEs were considered to be possibly related or related to study product use. Nervous system disorders (e.g., headache) were the most common PEAEs reported during this study, with 14 events reported by 11 subjects, and there were no trends in PEAEs related to product use. All AEs were resolved without treatment, and there were no serious AEs or severe PEAEs reported during the study.

Fifty-five subjects were screened and randomized to one of 14 product-use sequences, with 51 subjects completing the study without any major protocol deviations (Table 5). Two subjects withdrew from the study themselves and two were excluded from the analysis because of multiple major protocol deviations due to poor venous access.

The mean plasma nicotine concentration-time curves following single use of the study products are shown in Figure 2. The profile of the plasma nicotine concentration-time curves obtained after use of eDNC2.0a with four flavor variants (Products A, B, C, and D) was similar. Following the use of eDNC2.0a with four flavor variants (products A, B, C, and D, respectively), analysis of the geometric LS means showed that C_max values of 4.22, 4.11, 3.52, and 3.99 ng/mL, were reached after 10.0 or 15.0 min of product use, and similar AUC_0–last values of 6.71, 6.62, 5.28, and 6.17 hr × ng/mL were observed (Table 6). In comparison, the profile of the nicotine concentration-time curves obtained after the use of the usual brand of CC (Product E) showed higher concentrations throughout the 360-minute sampling period, with a higher C_max value (12.30 ng/mL) that was reached earlier (7 min), and a higher AUC_0–last value (19.02 hr × ng/mL) observed. The nicotine inhaler (Product F) differed from the test e-cigarettes in that the C_max value of 1.95 ng/mL was reached later (40 min), and a slightly lower AUC_0–last value (3.93 hr × ng/mL) was observed. The reference e-cigarette (Product G) showed higher nicotine concentrations throughout the 360-minute sampling period, with a higher C_max value (5.74 ng/mL) that was reached at a similar time (7 min) and a slightly higher AUC_0–last value (7.69 hr × ng/mL) compared to the test e-cigarettes. Nicotine consumption from a single use of the test e-cigarettes (Products A, B, C, and D) was estimated to be 0.93 to 1.09 mg by measuring the weight of e-liquid used, which was less than that observed for the reference e-cigarette (Product G, 1.45 mg, shown in Table 6).

Figure 2.

Overall product liking, intent to use the product again, and subscales of mPES were markedly lower with the use of eDNC2.0a with four flavor variants (Products A, B, C, and D) compared to the usual brand of CC (Product E, shown in Table 7). These subjective measures of the test e-cigarettes were generally higher than the nicotine inhaler (Product F) and slightly lower or comparable to the reference e-cigarette (Product G), except for the mPES subscale “Aversion”.

Nineteen AEs were reported by 12 of the 55 subjects. All AEs were mild or moderate in severity; 11 PEAEs were considered to be possibly related or related to study product use. Nervous system disorders (e.g., headache) was the most common PEAEs reported during this study, with 12 events reported by 10 subjects, and there were no trends in PEAEs related to product use. All AEs were resolved without treatment, and there were no serious AEs or severe PEAEs reported during the study.