Myriam Chartoire (2), Leon Aarons (1), Jacklyn Smith (2)(3), Kayode Ogungbenro (1)
(1) Centre for Applied Pharmacokinetics Research, Manchester Pharmacy School, The University of Manchester, Oxford Road, Manchester M13 9PT, United Kingdom, (2) Division of Infection Immunity and Respiratory Medicine, School of Biological Sciences, University of Manchester, Manchester, UK, (3) University Hospital of South Manchester
Introduction: Coughing is a worldwide health problem and is associated with comorbidities which have a significant impact on the quality of life of patients [1]. The mechanism of cough hypersensitivity remains unclear, leading to poorly effective antitussive therapies and difficulties in establishing new effective treatments. This lack of efficient antitussive therapies as well as the absence of validated tests, that are able to discriminate various cough conditions in clinical practice, make the diagnosis and treatment of cough challenging. In order to elucidate the pathophysiological mechanisms and evaluate the efficacy of antitussive therapies, the capsaicin cough challenge test has been developed.
Objective: The aim of this analysis was to use a population pharmacodynamic model to analyse and interpret data from the capsaicin cough challenge test of a group of healthy volunteers and chronic cough patients.
Methods: Within the capsaicin cough challenge test, healthy volunteers and chronic cough patients received four consecutive inhalations of capsaicin at increasing doses. The cough responses induced by capsaicin were analysed using a population PD model where the number of coughs was assumed to follow a Poisson distribution and the mean count response was described by a sigmoid Emax model [2]. The population dose-response model was implemented in NONMEM (Version 7.3) and used to estimate the capsaicin dose inducing half-maximal response (ED50) and the maximum cough response evoked by capsaicin (Emax).
The tachyphylaxis effect between consecutive inhalations of the same capsaicin dose was investigated by adding a tachyphylaxis parameter (K) to the model following the same methodology as Satia et al., 2017.
The effect of sex, age, body mass index, disease state (chronic cough patients or healthy volunteers), 24h cough frequency, day cough frequency, night cough frequency, predicted FEV1, FVC, smoking level, and the score obtained by a visual analog scale during day and night on the cough response data were evaluated. The covariates were choosen by investigating the relationships between the posterior parameter estimates obtained from the model without covariates and potential covariates. The covariate model building was performed using forward inclusion-backward deletion procedures in NONMEM.
Results: The population PD model was stable and the prediction of the model showed a good fit with the observed cough response data at a population level and for different sub-populations.
The population PD model also fit the cough response data at an individual level. The observed cough responses to capsaicin were highly variable between subjects and the model was flexible enough to adapt to these different individual profiles.
Chronic cough, sex and 24h cough frequency were found to have impact on the cough responses. Chronic cough patients showed elevated capsaicin-evoked cough responses as chronic cough patients had a 113% higher Emax value compared to the healthy volunteers. Females had a 72 % lower ED50 compared to males which means that females started to cough at a lower capsaicin dose. Finally, higher 24h cough frequency were associated with lower cough thresholds as for every unit that 24h cough frequency increased, ED50 decreased by 4%.
The introduction of a tachyphylaxis parameter improved the fit of the model by lowering the objective function value by 186 units. A tachyphylaxis effect was found, leading to a reduction of approximatively 13% of the parameter Emax between consecutive inhalations of the same capsaicin dose.
Conclusions: The population PD model was successfully applied to the capsaicin cough challenge test data of a group of healthy volunteers and chronic cough patients. Furthermore, the model adequately described the observed cough response to capsaicin. The population PD model was able to discriminate healthy volunteers from chronic cough patients and the cough response was correlated with the 24h cough frequency before the test.
References:
[1] Pandya, A.K., K.K. Lee, and S.S. Birring, Cough. Medicine, 2016. 44(4): p. 213-216.
[2] Satia, et al., Capsaicin-evoked cough responses in asthmatic patients: Evidence for airway neuronal dysfunction. J Allergy Clin Immunol, 2017. 139(3): p. 771-779 e10.
Reference: PAGE 29 (2021) Abstr 9783 [www.page-meeting.org/?abstract=9783]
Poster: Drug/Disease Modelling - Other Topics