Eva Germovsek (1), Anna Hansson (2), Mats O Karlsson (1), Åke Westin (2), Paul A Soons (3), An Vermeulen (3), Maria C Kjellsson (1)
(1) Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden; (2) McNeil AB, Helsingborg, Sweden; (3) Janssen R&D, a division of Janssen Pharmaceutica NV, Beerse, Belgium
Objectives: Tobacco use is a major public and individual health problem, with recent estimations showing it causes over 7 million avoidable deaths yearly [1]. This is mainly due to the fact that smoking increases the risk of developing cancer, but it also causes chronic obstructive pulmonary disease, accelerates cardiovascular disease development, increases the risk of infectious diseases and insulin resistance, etc. To reduce tobacco use, nicotine replacement therapy (NRT) formulations are commonly used since they reduce individual’s craving for nicotine and thus increase the probability of stopping smoking [2, 3]. The aim of this work was to quantitatively investigate how a pharmacodynamic marker of long-term withdrawal symptom response (i.e. ‘integrated’ craving, meaning craving over a period of time) is linked to smoking cessation.
Methods: Retrospective data that were available for analysis were collected in 19 separate studies, and included three NRT formulations, specifically: inhaler, mouth spray, or patch, and also data from the combination use of inhaler and patch. Smokers, who were instructed not to smoke, were included in either the active (where they received NRT) or the placebo treatment arm. If a subject was not present on a visit, this was (per definition in the studies) conservatively defined as non-abstinence. Integrated craving was assessed with three craving scales: a 4-category scale, a 5-category scale, and a 100 mm visual analogue scale (VAS). Abstinence status was a binary response, and was modelled using a time-to-event (TTE) model, where the event was non-abstinence. Since events were only assessed at study visits, the exact time of an event was unknown, and interval-censoring was used. Several hazard functions were explored, where hazard was time-constant, time-varying (i.e. Gompertz, Weibull distributions were used), or a combination of both. Integrated craving at previous and current (i.e. when an event was recorded) visit was investigated as a predictor on the hazard. NONMEM 7.4 (ICON Development Solutions, Ellicott City, Maryland) with the Laplace approximation was used to obtain the likelihood.
Results: The data included 9,323 adult subjects with median (range) 2 (0-99) previous attempts to quit smoking, and median (range) age 42 (17-81) years and body mass index 25 (13-64) kg/m2. The largest proportion of subjects (30%) smoked their first cigarette 6-30 minutes after waking up, and the smallest proportion of subjects (6%) >60 minutes after waking up. In total, 9% of subjects remained smoking abstinent until the end of the studies, with study lengths ranging from 3 weeks to 2 years. A combination of time-constant and the Gompertz hazard proved best to describe the data, with hazard decreasing over time. Current integrated craving provided a better fit than previous craving. Integrated craving was positively related with the hazard of having an event, and was included in the model using a sigmoidal Emax function as it showed a better fit to the data (dOFV=-2015), compared to a linear relationship between craving and hazard. The integrated craving giving 50% of maximal hazard due to craving (Crav50) was 0.81 (relative standard error (RSE) 5%) for the 4-category scale, and 24.5 (3% RSE) for the VAS. For craving measured with the 5-category scale Crav50 was 0.98 (1% RSE) and 1.24 (7% RSE) for single NRT formulations and for combination of inhaler and patch, respectively. Hill exponent was estimated as 12 (17% RSE). Hazard at 1 year was 0.13/month for maximal craving from different scales, and 0.014/month for craving of zero (i.e. lowest craving).
Conclusions: A time-to-event model was developed, where integrated craving was related to hazard in a sigmoidal Emax fashion, and showed that higher craving is related to lower probability of remaining abstinent. Future work will involve developing a model to connect all craving scales (similarly as done in [4]) and investigating other predictors, such as markers of smoking addiction.
Disclosures: EG, MCK and MOK declare no conflicts of interest; AH, ÅW, PS and AV are (former) employees of subsidiary companies of Johnson & Johnson.
References:
[1] GBD 2017 Risk Factor Collaborators. Global, regional, and national comparative risk assessment of 84 behavioural, environmental and occupational, and metabolic risks or clusters of risks for 195 countries and territories, 1990-2017: a systematic analysis for the Global Burden of Disease Study 2017. The Lancet, 2018; 392: 1923-1994.
[2] Benowitz, N.L., Clinical pharmacology of nicotine: implications for understanding, preventing, and treating tobacco addiction. Clin Pharmacol Ther, 2008; 83(4): 531-541.
[3] Stead, L.F., R. Perera, C. Bullen, D. Mant, J. Hartmann-Boyce, K. Cahill, and T. Lancaster, Nicotine replacement therapy for smoking cessation. Cochrane Database Syst Rev, 2012; 11: CD000146.
[4] Germovsek, E., Hansson, A., Kjellsson, M.C., Perez Ruixo, J.J., Westin, Å., Soons, P.A., Vermeulen, A., Karlsson, M.O., An exposure-response model relating nicotine plasma concentration to momentary craving across different nicotine replacement therapy formulations, PAGE 27 (2018) Abstr 8649 [www.page-meeting.org/?abstract=8649]
Reference: PAGE 28 (2019) Abstr 9074 [www.page-meeting.org/?abstract=9074]
Poster: Drug/Disease Modelling - Other Topics