This 56-week, multiregional, multicenter, ambulatory study was conducted at 42 sites in the United States, Japan, and Europe, and it is registered at ClinicalTrials.gov (identifier NCT02432729). The institutional review boards or independent ethics committees for each participating institution granted ethical approval. The study followed the principles defined by the International Conference on Harmonization Good Clinical Practice, the Declaration of Helsinki, and other applicable regulations [17,18]. No smoking control arm was included, as the endpoints were analyzed before and after SC and numerous studies have already documented such aspects in smokers [19-21]. The first subject was enrolled in the study on May 5, 2015, and the last subject completed the study on May 30, 2017. The study design is illustrated in Figure 1. Samples were collected during the baseline visit for baseline biomarker analysis. During this visit, participants were asked to define their target quit date, the date from which they would stop smoking. The target quit date had to be within 14 days after the baseline visit. The next visit, scheduled 24 to 48 hours after the target quit date, was used to determine whether the subject had stopped smoking and to provide SC support. A grace period of up to 14 days was allowed after the target quit date, during which occasional use of nicotine and/or tobacco-containing products was tolerated. Starting from the actual quit date, participants had to abstain completely from smoking and from using any nicotine or tobacco-based products other than nicotine replacement therapy (NRT; allowed for up to 3 months and 2 weeks). The period from baseline visit to the actual quit date has been established to identify participants with a higher likelihood of successful SA for 1 year.

Participants recorded their own actual quit date and provided the information to their study clinic. Visits were scheduled on a monthly basis for the duration of the 52-week study (1 year), with biomarker sample collections scheduled for the Month 3, Month 6, and Month 12 visits. The study ended after a 28-day safety follow-up period. SC support, including counseling and behavioral support, was provided throughout the study at scheduled visits and between visits as requested by the subject. Participants were allowed to use NRT to support SC if requested by the subjects. NRT was started any time between the target quit date and 1 week after the actual quit date and was permitted for up to 3 months and 2 weeks. Adverse events were collected at each visit. Participants who smoked after the actual quit date were discontinued from the study.

All participants provided written informed consent. The study enrolled healthy adult smokers who were motivated to quit smoking within the next 30 days. Their motivation was assessed using a questionnaire based on the Prochaska stages of change [22]. The main inclusion and exclusion criteria are summarized in Textbox 1. Participants had at least 10 years of smoking history and had smoked at least 10 cigarettes per day over the last 12 months. Participants with FEV₁ and forced vital capacity (FVC) <0.7 and FEV₁ <80% predicted value at postbronchodilator spirometry and those with FEV₁ and FVC <0.75 (postbronchodilator) and reversibility in FEV₁ (both >12% and >200 mL from pre- to postbronchodilator values) were excluded from the study. There were no limitations on race or ethnicity. Stratified sampling was used to ensure adequate representation of genders (at least 40% of each sex at enrollment).

A total of 1035 participants who successfully abstained from smoking for at least 2 weeks after the actual quit date were remaining in the study, meaning the study was completed with at least 190 successful quitters.

The main objective of the study was to describe the biological and functional changes in smokers who are continuously abstinent from smoking. The biomarkers of effect, including those associated with CVD, respiratory diseases, xenobiotic metabolism, and genotoxicity, are provided in Table 1. This broad range of biomarkers of effect were selected based on the predefined criteria according to the epidemiological evidence that the biomarkers of effect were associated with smoking-related diseases, sensitive to smoking status, and reversible upon SC over a period of time that was compatible with the study duration.

The biomarkers of exposure to HPHCs in smokers who continuously abstained from smoking (Table 2) were derived from published guidelines from the WHO and the US Food and Drug Administration [23,24] and according to predefined criteria, as reported previously [25].

The rate of continuous SA was determined at each visit following the actual quit date.

Safety was established by monitoring adverse events, body weight, vital signs, spirometry, electrocardiogram, hematology and clinical chemistry marker panels, urine analysis, physical examination, and concomitant medications. Adverse events were coded according to MedDRA terminology.

Full lung function assessments, blood and urine samples for biomarkers of effect, and biomarkers of exposure analyses were conducted at baseline (Visit 2) and at 3 time points during the study (Month 3, Month 6, and Month 12 visits). Cough assessment by visual analog scale and Likert scales (intensity of cough, frequency of cough, and amount of sputum collection) were conducted at baseline (Visit 2) and at 3 time points during the study (Month 3, Month 6, and Month 12 visits). Noncompliance with continuous SA was verified as follows:

Socioeconomic status, lifestyle, stage of change [22], and nicotine dependence [26] were assessed as baseline characteristics. One central laboratory was used for sample management and several laboratories were used for the blood and urine analyses (Covance Central Laboratory Services Inc; Celerion, Lincoln, United States; and Celerion Switzerland AG, Switzerland) using validated and fit-for-purpose methods.

On the basis of the results of the Lung Health Study and on the 1-year abstinence rates from English smoking treatment services [27], at least 950 smokers were required to have approximately 190 participants who successfully abstain from smoking for the duration of the study [28]. Sample size was driven by FEV₁, which is expected to have the lowest effect size. Approximately 190 participants were needed to estimate the mean increase from baseline of 1.98 (% predicted) in FEV₁ at Month 12, with a 90% probability of obtaining a margin of error (95% CI) of at most ±1 (% predicted). Enrolled participants who failed to abstain for at least 2 weeks were discontinued from the study.

Changes from baseline were summarized for the main analysis population, defined by quitters with no major protocol deviations impacting subject availability. Data collected after evidence of noncompliance with continuous abstinence were not included in the analysis. For analysis purposes, the concentrations of free cotinine (≥50 ng/mL) [29] and total NNAL (≥75.9 pg/mL) [30] in 24-hour urine collected at Visit 11 were included in the list of tools for continuous SA verification. This study had no formal prespecified hypotheses to be tested for statistical significance. However, a 95% CI accompanied all effect estimates.

The approach of our study is unlike other SC studies whose primary objective was to test the efficacy of an SC treatment (ie, drug or behavioral cessation support) [31-33]. This dataset will supplement the existing literature data on the effects of SC while providing prospective and comprehensive perspective on favorable health effects that occur following SC. Although several papers report favorable changes in biomarkers of effect with SC [34,35], there are very little data on 1-year observations and none with such an extensive set of endpoints for clinical risk and exposure. Changes in inflammatory markers with SC have been evaluated in previous studies. SC is reported to reduce oxidant stress and inflammation. This is evidenced by an improved urinary F₂ isoprostane:creatinine [F₂:Cr] ratio and decreases in WBC counts [36]. Time-dependent changes in LDL levels have been reported with cessation. LDL levels were measured in 50 smokers before cessation and at 3 months and 1 year after cessation. At 1 year after cessation, LDL levels were reported by Komiyama et al to decrease significantly, although levels did not change at 3 months [37]. For a long time, it has also been known that smokers have lower HDL levels than nonsmokers. Past studies have shown that HDL levels increase following cessation, and that this increase occurs rapidly, in less than 3 weeks, but with no clear pattern of change thereafter [6]. SC is also reported to be accompanied by a rapid reduction in tobacco smoke carcinogen and toxicant biomarkers. After 3 days of cessation, an >80% reduction has been reported for monohydroxybutyl mercapturic acid, 3-hydroxypropyl mercapturic acid, 4-hydroxybut-2-yl mercapturic acid, S-phenyl mercapturic acid, and 2-hydroxyethyl mercapturic acid. Gradual reduction has been reported for some biomarkers, including a 92% reduction in total NNAL after 42 days [38]. Variable dose-response change in biomarkers of exposure following a reduction in cigarettes smoked per day has also been reported. Some biomarkers, such as plasma nicotine, show a strong dose-response reduction, whereas others, such as plasma thiocyanate, show weaker dose-response reductions [39].

In this study, the list of biological and functional parameters and biomarkers of exposure to be tested was extensive. This list was carefully selected to provide high-quality evidence for the effect of SA on a variety of HPHCs and the health-related effects of smoking that are reversible upon SC. These endpoints will provide further insights into the risk profile of a smoker following abstention from smoking within a short time frame by examining a collection of logical, empirically coherent, and mutually supportive data from multiple clinical risk components across several biological processes, physiological systems, and mechanisms that are known to contribute to the pathogenesis of smoking-related diseases.

The results of this study may offer a valuable point of reference for future assessments of alternative products in the context of tobacco harm reduction, providing scientific evidence of the potential of alternative products to reduce the risk of harm in smokers within a 1-year time frame when compared with continuing to smoke cigarettes. With this aim, cross-study analyses need to be designed carefully to ensure baseline comparability of quitters with the population of smokers switching to the alternative product. Such a comparative approach is valuable for obtaining data on the risk reduction potential of a product alternative to cigarettes earlier than epidemiological studies, which require long-term assessment to provide data.

The design and approach used in the present study should be considered in light of its limitations. One limitation was related to the difficulty of recruiting smokers who were both willing and motivated to quit smoking, completely and continuously, for 1 year. In addition, the uncontrolled before and after study design and the lack of a control arm with no intervention might require careful interpretation of the outcomes of this study. The results of this study should be interpreted with caution owing to the lack of control of potential confounding factors. Factors such as concomitant medications, lifestyle, and weight increase can cause changes in lipid metabolism and other biochemical processes following SC [40].

This study was designed to provide an extensive dataset on changes in biomarkers of effect and biomarkers of exposure after 12 months of continuous SA. Therefore, it will provide a comprehensive overview of the beneficial short-term health effects that occur over the course of 1 year in a smoker who ceases smoking. The results from this study will complement existing evidence for the benefits of cessation. In the context of smoking harm reduction, these results may be used as a benchmark for the future evaluation of alternative products to cigarettes and to supplement the existing literature on the biological and functional health effects of SC.