Influence on Cough and Airway Symptoms by Oral Capsaicin

This is a phase 2 clinical study in humans for therapeutic use of Capsicum oleoresin - (capsaicin) in chronic idiopathic cough (CIC). The study has a randomised, double-blind and cross-over design. During 4 weeks the participants take either active capsules (Capsicum oleoresin), or matching placebo capsules. This period follows by 2 weeks of "wash out" and then another 4 weeks with active capsules or placebo.

The IMP

The soft gel capsules used in the study are easily digested in the stomach and contain standardised doses of Capsicum oleoresin with the main component capsaicin (8-methyl-N-vanillyl-6-nonenamide) (from chili extract). The placebo capsules are filled with sorbitol and colorant.

Capsicum oleoresin and capsaicin is found naturally in a variety of chili fruits and the use of chili peppers in food varies greatly between different parts of the world. The maximum daily intake of capsaicin from Capsicum oleoresin in the current study could be exceeded many times in countries such as Mexico and Thailand, but it does not reach that level in most Western countries.

In concordance with the known effect of topical capsaicin for neuropathic pain, the expected effect of orally taken capsaicin is desensitisation. Capsaicin desensitise the cough-sensitive transient receptor potential vanilloids one receptors (TRPV1) and other TRP receptors followed by decreased cough reflex sensitivity and coughing in patients with CIC. Capsaicin causes depletion of neuropeptides leading to "exhausting" of the receptor with ameliorated symptoms in the patients.

In the application to the Swedish Medical Products Agency the capsules, the active product ingredient (API) and the formulation are extensively described in the investigational medical product dossier (IMPD).

In summary

• Supplier of raw material for the Capsicum oleoresin product/IMP: RANSOM; Ltd, London, UK
• The IMP is developed by RISE, Sodertalje, Sweden
• Filling of capsules with Capsicum oleoresin and placebo: Catalent Pharma Solutions, St Petersburg, Florida, USA
• Packing in bottles of capsules with IMP and placebo: by Apotek Produktion \& Laboratorier AB (APL), Stockholm, Sweden
• Labelling: Tamro, Gothenburg, Sweden (an independent pharmaceuticals logistics service provider). Each package will be labelled with study code, randomisation number, dosage form, quantity, dosage instructions, name of investigator, expiry date, storage instruction.
• Storage: in room temperature at the allergy clinic at Sahlgrenska University Hospital and at Tamro, Gothenburg, Sweden
• Independent monitoring of the clinical study by qualified staff from Gothia Forum, Sahlgrenska University Hospital, Gothenburg, Sweden
• A safety report will be completed after the study end and sent to the relevant Competent Authority and the Ethics Committee

Study population

60 patients with CIC from the outpatient clinic at the asthma and allergy clinic, Sahlgrenska University Hospital, Gothenburg, Sweden are included.

Study design

The total study time for each patient is 24 hours + 10 weeks and includes 6 visits. At the first visit tests and questionnaires are assessed as described below (outcome measures) and the patients collect a cough monitor to record cough during 24 hours. After returning the cough monitor the patients start with 4 weeks of active treatment or placebo. This is followed by 2 weeks of wash out and then the patients start with another 4 weeks of active treatment or placebo. After this the study ends. Estimated time for each visit is 2 hours.

During the first 2 weeks of each arm (active or placebo) of 4 weeks the patients take one capsule morning and evening. During the following 2 weeks they take 2 capsules morning and evening.

At each visit:

• The patients answer the questionnaires LCQ-S, the HARQ-S and also score their cough severity during the past two weeks using a VAS-scale (0-100 mm)
• A cough sensitivity test is performed with a standardised capsaicin inhalation test, assessing concentration of inhaled capsaicin causing 2 coughs (C2), 5 coughs (C5) and 10 coughs (C10)
• Lung function measured with spirometry and impulse oscillometry (IOS)
• Pain sensitivity is assessed with pressure algometry at 4 pre-defined points
• Levels of capsaicin/dihydrocapsaicin in sera are measured

At 4 opportunities:

1: 24 hours before any treatment. 2: at the end of 4 weeks of active treatment or placebo respectively. 3: at the end of 2 weeks wash out. 4: at the end of 4 weeks of active treatment or placebo respectively:

• Fraction of exhaled nitric oxide (FeNO) is measured
• Particles in exhaled air (PExA) are measured
• The cough frequency during 24h is assessed using the LCM cough monitor

Statistical Analysis Plan (SAP):

All detailed statistical analyses will be specified in a separate study SAP that is developed and finalized.

Sample size calculation

The sample size calculations for the primary variables LogC2 and LogC5 given the results obtained in an earlier pilot study are described in the following:

In order to reach the power of 0.80 with mean difference in LogC2 between the Active-Placebo treatment of 0.590 and standard deviation for the difference of 0.885, with two-tailed Fisher's non-parametric permutation test, alpha = 0.025, 24 patients are needed to be included in the study.

In order to reach the power of 0.80 with mean difference between in LogC5 the Active-Placebo treatment of 0.425 and standard deviation for mean difference of 0.911, with two-tailed Fisher's non-parametric permutation test, alpha = 0.025, 47 patients are needed to be included in the study.

In order to adjust for drop-outs, 60 patients will be included in this study. Statistical methodology The main analyses will be on the ITT population and complementary analyses will be performed on the PP population. In this cross over study all main analysis will be adjusted for period effects.

For non-normal continuous variables, ordered categorical variables and dichotomous variables the following non-parametric analyses of cross-over design for test between Capsaicin and Placebo will be used, adjusting for period effects. The difference between period 2 and period 1 values will be calculated and this difference will be tested between the treatment sequences. For continuous variables this difference will be continuous and analysed with Fisher's non-parametric permutation test. For ordered categorical variables and dichotomous variables this difference will be an ordered categorical variable and analysed with Mantel-Haenszel chi-square test.

The test regarding the effect of Capsaicin vs Placebo on normally distributed efficacy variables will be performed by using generalised linear models for cross-over design, with sequence, period and treatment group as fixed effects. SAS code for the analysis of normally distributed data:

PROC MIXED; CLASS SEQ PATIENT PERIOD TRT; MODEL Y=SEQ PERIOD TRT; REPEATED / TYPE=CS SUB=PATIENT(SEQ); RUN; The most appropriate variance-covariance structure will be used, that might by other than Compound Symmetry (CS) specified in the code above.

For continuous variables the distribution of the variables will be given at baseline, visit 2 and visit 4 and change from baseline to visit 2 and 4 by Capsaicin treatment and placebo.

95% confidence interval for the mean difference between Capsaicin and placebo treatment adjusted for period effect will be given. The results will be presented in both tables and figures.

All categorical variables will be described by number and percentage, and all continuous variables by mean, standard deviation, median, minimum and maximum.

The carry-over effect for the primary variables will be analysed by testing the mean of period 1 and period 2 values between the two treatment sequences.

For test of change within groups Sign test will be used for ordered categorical variables and dichotomous variables and Fisher's non-parametric permutation test for paired observations for continuous variables.

The missing values will be imputed by using stochastic imputation (seed=99786) using all available baseline and relevant follow-up data for all efficacy analyses.

The main analyses will be performed on stochastic imputed values and sensitivity analyses on complete case and on last observation carried forward for primary efficacy variable. Baseline and demographic variables will be described for the total sample.

The study will be considered positive in case any of the two primary variables reach the significance level below 0.025. For all other tests the significance level of 0.05 will be applied. All significance tests will be two-tailed and all analyses will be performed by using SAS® v9.4 or later (SAS Institute Inc., Cary, NC, USA).