Predictors of clinical response to extrafine and non-extrafine particle inhaled corticosteroids in smokers and ex-smokers with asthma

Asthma patients who smoke experience more severe symptoms as well as airflow limitation, and benefit less from treatment with inhaled corticosteroids (ICS) compared with non-smoking asthmatics [1]. Additionally, smoking is associated with small airways dysfunction (SAD) [2]. The small airways, defined by a diameter ≤ 2 mm contribute to the resistance in the airways of patients with obstructive airways disease [3]. This has a clinical impact since small airways can be inflamed in asthma and hence narrowed [4]. We hypothesized that smokers and ex-smokers with asthma would benefit more from extrafine than non-extrafine particle ICS. However, our OLiVIA-study showed that extrafine and non-extrafine particle ICS were equally effective in improving small airways function in current and ex-smokers with asthma [5]. This outcome might be ascribed to the fact that the presence of SAD was not an inclusion criterion in our study. Therefore, we performed a post-hoc analysis to investigate whether current and ex-smoking asthmatics with SAD show a better clinical response to extrafine compared to non-extrafine particle ICS. Next, we investigated which clinical parameters, apart from the presence of SAD, predict a favorable response to extrafine and non-extrafine particle ICS.

The OLiVIA-study was an open-label, randomized, three-way crossover, two-center study, comparing two-week treatment with extrafine hydrofluoroalkane (HFA)-beclomethasone 200 μg b.i.d. (QVAR) to non-extrafine HFA-beclomethasone 400 μg b.i.d. (Clenil) and non-extrafine HFA-fluticasone 250 μg b.i.d. (Flixotide) [5]. The primary outcome was the change in airway hyperresponsiveness (AHR) to small particle adenosine, expressed as the provocative dose of adenosine causing a 20% drop in FEV₁ (forced expiratory volume at 1 s) from baseline to post-treatment (ΔPD₂₀). Small particle adenosine is a provocative agent that acts indirectly via the release of mediators from inflammatory cells. It was chosen rather than methacholine as it is a more sensitive measurement to detect improvement in AHR after treatment with ICS in patients with asthma [6].

Since there is no clearly defined golden standard for SAD, we applied various parameters next to the PD₂₀ adenosine, i.e. parameters of spirometry, body plethysmography, impulse oscillometry (IOS) and multiple breath nitrogen washout (MBNW). We used cut-off values for SAD parameters as follows: forced expiratory flow between 25 and 75% of forced vital capacity [FEF_25–75] < lower limit of normal (LLN) from spirometry; ratio of residual volume to total lung capacity [RV/TLC] > upper limit of normal (ULN) from body plethysmography; difference between resistance at 5 Hz and 20 Hz [R₅-R₂₀] > 0.1 kPa sL^− 1 from IOS [7]; and ventilation heterogeneity of the acinar structures [S_acin] and conductive airways [S_cond] > ULN from MBNW [8]. These measurements were taken at the baseline visit after an ICS washout period of four to six weeks. The presence of SAD was defined as: at least 3 out of 5 criteria fulfilled when all lung function measurements were performed, or with 2 out of 3 criteria if MBNW was not carried out, which was the case in 8 patients.

We performed a two-sided paired t-test or Wilcoxon test to assess the difference in ΔPD₂₀ adenosine after treatment with extrafine compared to non-extrafine particle ICS in patients with SAD and patients without SAD. PD₂₀ values were log₂-transformed prior to analyses. To investigate the difference in ΔPD₂₀ between patients with SAD and without SAD within one treatment group, we performed a student t-test or Mann-Whitney test. Next, we explored which baseline variables (i.e. demographics, small and large function measures, presence of atopy and blood leukocyte counts) were associated with ΔPD₂₀ adenosine after extrafine and non-extrafine particle ICS treatment using univariate linear regression analysis. We finally performed multivariate linear regression analyses for each treatment type to identify independent clinical predictors for ΔPD₂₀ including age, sex, smoking status, numbers of neutrophils and eosinophils, and completed with those variables showing a trend towards association with ΔPD₂₀ (p < 0.1) in the univariate analysis. We allowed a maximum of 7 variables to be included in the multivariate model. The variables in this final model were selected based on the following criteria: (1) the variable changed regression coefficients (B) > 10%, (2) the model had highest total explained variance (R²), and (3) the model preferably included the highest number of subjects.

We analyzed 43 asthma patients (22 smokers and 21 ex-smokers), 42% being male, with mean (standard deviation (SD)) age 45 (12.6) years, FEV₁ 83 (14.5) % predicted, and PD₂₀ 2.69 (3.41) mg. Baseline characteristics of the study population are presented in Additional file 1: Table S1. At baseline, we found that SAD was present in 18 (42%) patients (see Additional file 1: Table S2). Asthma patients with and without SAD had similar response in PD₂₀ adenosine to extrafine (QVAR) and non-extrafine particle ICS treatments (Clenil and Flixotide) (Fig. 1; and Additional file 1: Table S3).

Results of the univariate linear regression analyses are presented in Additional file 1: Table S4. Multivariate linear regression analyses (Table 1) showed that no parameters of SAD were associated with treatment response to either small or large particle ICS. Lower numbers of blood neutrophils were associated with a larger increase in PD₂₀ adenosine, independently from the level of blood eosinophils, in patients treated with QVAR, Clenil and Flixotide (all p < 0.05). Higher blood eosinophils were also associated with a larger increase in PD₂₀ adenosine in patients treated with Flixotide (p < 0.05). Finally, younger age tended to be associated with a larger increase in PD₂₀ adenosine after using Clenil (p = 0.05).

We did not confirm our hypothesis that current and ex-smoking asthmatics with SAD had a better treatment response to QVAR compared to Clenil and Flixotide. In line with previous reports on better clinical effects of ICS in asthmatics with eosinophilia [9‐11], we show that higher blood eosinophils are associated with a less severe AHR after treatment with non-extrafine particle ICS (Flixotide). Of interest, our study shows that higher blood neutrophils are associated with more severe AHR after treatment with both extrafine and non-extrafine particle ICS in smokers and ex-smokers with asthma, independently from the level of blood eosinophils. Findings by Telenga et al. support that lower blood neutrophils are associated with an increase in FEV₁ after 2-week ICS-therapy [11]. Taken together, we find that higher blood neutrophils are associated with less clinical ICS-response in smokers and ex-smokers with asthma. However, our data (Additional file 1: Figure S1) do not clearly show a cut-off value for eosinophilia or neutrophilia that can be used in clinical practice to define a positive response. Future studies have to assess whether this change in inflammation in peripheral blood is reflected by changes in the small and larger airways.

A strength of the study is that we defined SAD by cut-off levels applying multiple measurement techniques including spirometry, body plethysmography, impulse oscillometry and multiple breath nitrogen washout. A limitation is the relatively small sample size of patients with SAD and non-SAD which limits the power of the study and raises the risk of type I error. However, a post-hoc power analysis indicated that we had a sufficient number of patients to detect a difference of one doubling dose increase in PD₂₀ adenosine between patients with and without SAD. With a standard deviation of one doubling dose, β = 0.8 and α = 0.05, we would need 16 patients per group while in our study, 18 patients with and 23 patients without SAD were investigated. Further studies with larger sample sizes would provide a clearer picture of SAD in smokers and ex-smokers with asthma. Another possible limitation is that we treated for two weeks which may have been too short to improve SAD.

In conclusion, we show that smoking and ex-smoking asthmatics with and without SAD have a similar response to small particle adenosine after treatment with either extrafine or non-extrafine particle ICS. These findings suggest that clinicians may not need to consider SAD in order to decide whether current- and ex-smoking asthmatics would benefit preferentially from treatment with extrafine rather than non-extrafine particle ICS. Of importance, we find that lower blood neutrophils is a favorable predictor of ICS response, independent from the level of blood eosinophils.