Effect of Atorvastatin on Serum Levels of Periostin in Asthma - A Placebo Controlled Randomized Clinical Trial

Background: Asthma is a common pulmonary disease with a high morbidity rate. Considering the role of inammation in the disease induction and severity, and the anti–inammatory effect of atorvastatin, we aimed to investigate the effect of atorvastatin on serum level of periostin and eosinophil in patients with asthma, in order to introduce the usefulness of this drug on reducing the inammation in patients with asthma. Methods: In a blinded randomized control trial, 80 patients with denite diagnosis of asthma were enrolled into the intervention or placebo group, who received 40 mg/day atorvastatin and similar placebo every day for 8 weeks. Spirometric measurements were performed and asthma control test was completed by the patients at baseline, after 4 weeks, and at the end of study (after 8 weeks). The primary outcomes of the study were serum eosinophil count and periostin levels. Results: 40 patients were analyzed in each group and their data were analyzed. Mean asthma control scores in the intervention and placebo groups were similar at baseline (17.95±3.75 vs. 17.98±3.77; P>0.05) and improved in the intervention group (19.88±3.28; P<0.001), but remained unchanged in the placebo group (18.6±3.26; P=0.09). Neither the spirometric changes, nor the serum levels of eosinophil count and periostin were signicantly different between the groups (P>0.05). Conclusions: Although the intervention group had improved asthma control score after the intervention, this effect could be because of the indirect effect of atorvastatin, as the spirometric measurements and inammatory markers did not change signicantly and were not different between the placebo and intervention groups. Further studies with longer follow-ups may show changes in spirometric parameters by atorvastatin or its effect on other inammatory markers.

release toxic granule proteins and reactive oxygen species (ROS) for defending pathogens, as well as chemokines, cytokines, lipid mediators, and other growth factors, and their increase in bloodeosinophilia-can cause airway damage and remodeling and is associated with severe asthma [6,7].
However, eosinophilia cannot be used as a diagnostic marker of asthma, as it can be observed in several in ammatory conditions [8]; accordingly, research has continued on more speci c markers for diagnosis of asthma.
Periostin, an extracellular matrix protein of fasciclin family, is identi ed as a downstream molecule of cytokines (IL-4 and IL-13), attributed to subepithelial brosis in bronchial asthma [9]. It has been introduced as a systemic marker of eosinophilic airway in ammation [10] and is overexpressed in the lungs of patients with respiratory disorders, speci cally asthma [11]. Sputum and serum periostin levels have been associated with sputum eosinophil proportions, but it may have insu cient ability in prediction of eosinophilic asthma [12]. Research on the role of periostin in progression of asthma and using its serum levels as a marker of asthma severity or treatment e cacy is still active [13].
The complex etiology of asthma has resulted in di culty of treatment and continuing search for a more appropriate treatment, targeting in ammation [14]. Atorvastatin has shown bene ciary effects on reduction of in ammatory cell in ltration and airway hyperresponsiveness, through reducing the serum levels of in ammatory markers, such as transforming growth factor (TGF)-β1 and matrix metalloproteinases (MMP)-9, as well as immunoglobulins, such as IgE and IgG in animal studies [15,16]. However, human studies have shown dissimilar results in view of its e cacy on pulmonary function [17][18][19][20]. Therefore, in the present study, we aimed to investigate the effect of atorvastatin on serum level of in ammatory markers, spirometric parameters, and asthma control in patients with asthma, in order to determine the usefulness of this drug in treatment of asthma.

Study design
The present study was a triple-blind placebo-controlled randomized clinical trial (RCT), conducted in Shiraz, Iran, from January to July 2018. The study protocol was approved by the Ethics Committee of Shiraz University of Medical Sciences (Ethics code: IR.SUMS.MED.REC.1396.111), and was recorded in IRCT website by the code: IRCT2017102216367N2. The trial protocol can be accessed from https://www.irct.ir/user/trial/15332/view. The study adheres to CONSORT guidelines. The study sample size was calculated at 34 in each group considering standardized effect size of 0.65, using the results of the study by Hothersall and colleagues [17], and considering 95% con dence interval (95% CI), study power of 80%, and the chance of lost to follow-up, 40 patients were included in each group.
Patients who referred to the Lung Clinic of Shahid Faghihi Hospital and Motahari Clinic during the study period were considered for inclusion into the study, if they aged 18-60 years, had a history of asthma for more than 1 year, no history of smoking during the past 3 months or < 10 pack/year, no history of lower respiratory tract infection during the past 4 weeks, or using oral steroids or statins during the past 3 months. Also, the patients using drugs interacting with statins, such as antifungal drugs, macrolide antibiotics, cyclosporine, gem brozil, verapamil, and amiodarone, patients with hepatitis or active hepatic disease, myopathy, or myositis, and lactating, or pregnant women were not included into the study. Diagnosis of asthma was made by the pulmonology specialist according to history taking, physical examination, pulmonary tests, and imaging. The design and objectives of the study were explained to all participants and written informed consent was obtained from those who were willing to participate in the study and they were clari ed that they were free to leave the study whenever they wished to.
A statistician, who was not involved in other steps of the study, generated "A" and "B" codes for 80 patients using random allocation software. The researcher collecting the data, the statistician who analyzed the data, and the nurse who gave the drugs to the patients were not aware of the group allocations and the included patients were enrolled into the intervention and placebo groups based on the codes. The intervention group received 40 mg atorvastatin (Sobhan Iran Pharma Co., Iran) every day for 8 weeks and the placebo group received placebo with the same pattern. The placebo was prepared by the Schools of Pharmacy of Shiraz University of Medical Sciences and was completely similar in shape and color to atorvastatin, but only contained neutral ingredients without pharmacological effect. The patients were excluded from the study if they did not adhere the prescribed drugs routinely or did not refer for any of the follow-up examinations, or developed any complications related to the administered drug, or refused to continue the study. The patient recruitment stages are shown in Fig. 1.
All the patients underwent spirometric examination at baseline, after 4, and 8 weeks, and forced expiratory volume in rst second (FEV 1 ), forced vital capacity (FVC), FEV 1 /FVC, total lung capacity (TLC), residual volume (RV), RV/TLC, total airway resistance (Raw), and speci c airway resistance (sRaw) were recorded for each patient. Before and after the intervention (at baseline and after 8 weeks), the patients completed the asthma control test (ACT) [21]. This test is a simple method for assessing adequacy of asthma control by ve items, which provides a numerical score for assessment of asthma control during the previous 4 weeks. The validity of the Persian version of this test, used in this study, has been con rmed previously [22]. Furthermore, at baseline and the end of the study (after 8 weeks), one venous blood sample was taken from the patients, centrifuged at 300 G after 10 minutes, and its serum was separated. Then, the samples were kept at -70 degrees and serum parameters, including white blood cells (WBC), serum eosinophil percentage, polymorphonuclear percentage, hemoglobin, total cholesterol, low density lipoprotein (LDL), high density lipoprotein (HDL), triglyceride, creatinine, alanine transaminase (ALT), aspartate transaminase (AST), alkaline phosphatase (ALP), and serum periostin were measured using Thermoscientist USA LOT 0070122017 REF: EHPOSTIN and AWARENESS ELISA reader device. After drawing the standard gures and de ning cut-offs, the values were reported in ng/mL.
The primary outcomes of the study were considered as serum eosinophil count and periostin levels and the secondary outcomes of the study were considered as spirometric changes and the scores of ACT. Also, the patients' demographics, including age, sex, weight, height, and smoking status were extracted from the medical records. Participants' ages were categorized into below and above 50, and smoking status was recorded as positive or negative, while former and current smoking were considered as positive. Body mass index (BMI) was calculated based on the recorded weight and height.

Statistical analysis
The collected data were rst input into the statistical software IBM SPSS Statistics for Windows version 23.0 (IBM Corp. 2014. Armonk, NY: IBM Corp.), used for analysis. The descriptive results were presented by frequency (percentage) and mean ± standard deviation (SD). One-sample Kolmogorov-Smirnov test was used to assess the normal distribution of the data. Continuous variables were compared between the groups using t test and between the intervals using paired t test or repeated measures ANOVA. Categorical variables were compared using Chi-square test. P values of 0.05 or less were considered as statistically signi cant.

Results
The data of 40 patients were assigned and analyzed in each group. The demographic characteristics of the groups, including age categories, sex distribution, BMI, and smoking status were not statistically different between the intervention and placebo groups (P > 0.05, Table 1). According to the results of independent samples t test, mean ± SD of ACT score was not statistically different between the groups at baseline: 17.95 ± 3.75 in the intervention group and 17.98 ± 3.77 in placebo group (P = 0.98). At the end of the study, the results of paired t test showed that mean ± SD of ACT score improved signi cantly in the intervention group (19.88 ± 3.28; P < 0.001), but remained unchanged in the placebo group (18.6 ± 3.26; P = 0.09). Although the mean scores after intervention were not different between the groups (P = 0.09; based on the results of independent samples t test), the change in the scores was signi cantly higher in the intervention group compared with the placebo group (13.25 ± 2.96 vs. 5.19 ± 2.41, respectively; P = 0.038). Comparison of asthma control scores between the groups is shown in Fig. 2.
Comparison of the spirometric parameters between the groups showed that only TLC and RV showed statistically signi cant reduction in the intervention group (both P = 0.02), while they remained unchanged in the placebo group (P > 0.05); meanwhile, neither mean TLC and RV nor other parameters were different between the groups (P > 0.05, Table 2). Considering serum parameters, mean WBC count, total cholesterol, HDL, LDL, and TG showed statistically signi cant change after intervention compared with baseline values in the intervention group (P = 0.016, 0.005, 0.023, 0.015, 0.043, respectively), but not in the placebo group (P > 0.05); however, the difference between the groups was not statistically signi cant in none of serum parameters (P > 0.05, Table 3).

Discussion
In the present RCT, we compared the results of 8-week administration of 40 mg/day atorvastatin on asthma control and spirometric and serum parameters of patients with asthma. The results showed that asthmatic control score signi cantly improved in the intervention group, but did not change in placebo group, while neither spirometric parameters, nor the serum level of periostin and eosinophil counts were different between the groups. Serum lipid pro le of the patients was also evaluated before and after the intervention in order to determine patients' adherence to treatment, con rmed by the signi cant reduction in serum levels of total cholesterol, HDL, LDL, and triglyceride in the intervention group.
Atorvastatin is a cholesterol-lowering drug with anti-in ammatory effects and animal models have shown its e cacy on reduction of in ammatory cell in ltration and airway hyper-responsiveness [15,16].
However, human studies have shown controversial results considering its e cacy in treatment of asthma. Some have reported better asthma control in patients with severe asthma using statins [23,24], while other studies, like ours, have reported no signi cant difference [25][26][27]. In the study by Fahimi and colleagues, 17 adult patients with moderate to severe asthma received 10 mg/day atorvastatin or placebo for 4 weeks in a cross-over design study [25]. Similar to the results of the present study, they also reported no difference between spirometric parameters of the intervention or placebo group; however, there were some differences between the studies in the improvement of spirometric parameters by time in the intervention group, as they reported signi cant increase in the FVC (P = 0.01), while we detected signi cant reduction in TLC and RV in the intervention group, but not in FVC. These differences between the studies could be because of the different dose and duration of administered atorvastatin from our study. Alavi and others investigated the effects of 40 mg/day atorvastatin on patients with moderate to severe asthma compared with placebo, and reported changes in spirometric parameters in each of the groups without signi cant difference between the groups [26], which is consistent with the results of the present study. Similar to the methods of the present study, in a previous study on 62 patients, the researchers randomized patients with asthma to receive 40 mg/day atorvastatin or placebo for 8 weeks.
The results showed elevated FEV 1 and FVC in the placebo group after 8 weeks and elevated FEV 1 in the intervention group after 8 weeks, without signi cant difference between the groups in spirometric parameters [27]. The dose/duration of administration of atorvastatin in this study was similar to ours and no signi cant difference in spirometric parameters between the groups was also consistent with the results of our study; however, they only evaluated FEV 1 and FVC, while we reported complete spirometric results and showed signi cant changes in TLC and RV in the intervention group. Meanwhile, as TLC and RV were both lower in the intervention group of our study, it could not show any improvements in this group, which is consistent with the results of previous studies [25,27].
In the present study, the results of ACT showed signi cant improvement in asthma control scores of the intervention group (from 17.95 ± 3.75 to 19.88 ± 3.28; P < 0.001), in addition to greater change in the scores in the intervention group compared with the placebo group (13.25 ± 2.96 vs. 5.19 ± 2.41, respectively; P = 0.038). These results con rm the bene ciary effect of the administered atorvastatin on clinical outcomes of the patients with asthma. In the study by Moini and colleagues, 62 patients with mild to moderate asthma were randomized to receive 40 mg/day atorvastatin or placebo for 8 weeks.
The results showed signi cant reduction in asthma control scores both in intervention and placebo groups, while there was no difference between the groups in this regard [27]. They reported mean asthma control score of the placebo group increased from 15.27 ± 4.34 to 17.83 ± 4.47 (P < 0.001) and that of the intervention group from 16.78 ± 4.11 to 19.85 ± 3.20 (P < 0.001) [27]. Comparing the mean values of their study with our study shows that the baseline scores of their patients were lower than ours, while the mean asthma control score after 8 weeks in the intervention group of their study (19.85 ± 3.20) was close to that reported in our study (19.88 ± 3.28). Accordingly, these results con rm the signi cant effect of 8 weeks administration of 40 mg/day atorvastatin on asthma control, although this effect might not be the direct effect of the drug.
Atorvastatin has shown bene ciary effects on reducing the in ammatory markers such as tumor necrosis factor-α (TNF-α), C-reactive protein (CRP), and interleukins in other diseases, such as ischemic stroke [28], rheumatoid arthritis [29], and diabetes [30]. Accordingly, in the present study, we examined its e cacy on eosinophil count and periostin levels of patients' sera, as these are considered as important in ammatory markers in asthma [12]. Nevertheless, the results of our study showed no signi cant changes in eosinophil count and periostin in the groups and no difference between the intervention and placebo group in this regard. The serum eosinophil count of the placebo group in the present study was 8.11 ± 4.63 at baseline and 8.06 ± 5.37 after 8 weeks, while in the intervention group, it was17.98 ± 9.9 at baseline and 8.3 ± 4.21 after the intervention; meanwhile, there was no signi cant change in either groups, or signi cant difference between the groups at baseline or after intervention. In the study by Moini and colleagues, eosinophil count signi cantly reduced in the placebo group (from 221.31 ± 134.01 to 204.62 ± 134.64) and in the intervention group (from 244.35 ± 154.4 to 179.14 ± 117.43; both P < 0.001), which is contrary to the results of the present study. But they also reported no signi cant difference between the groups, which is in line with the results of the present study. In the study by Hothersall and colleagues, patients' outcomes were compared between the groups receiving 40 mg/day atorvastatin for 8 weeks and placebo in a cross-over design study with 6-week washout period and the results showed signi cant reduction in absolute sputum macrophage count in atorvastatin group, while morning peak expiratory ow was not different between the groups [17]. The results of studies could also depend on the main treatment of the studied patients, because of the possibility of drug interactions, as in the study by Hothersall and colleagues, inhaled corticosteroids were used as the background treatment.
As described above, the few studies addressing the e cacy of atorvastatin on asthma control and in ammatory markers in the patients with asthma have reported different results. The same discrepancy in the results of studies is observed regarding the effect of other statins, like simvastatin, on asthma control and reducing in ammation in such patients [31,32]. However, a wide variety is observed among studies in terms of disease severity, parameters, and main treatments used to control asthma. These suggest that further studies are required to determine the exact e cacy of statins on asthma control.
Current review studies indicated that atorvastatin reduced airway in ammation in patients with asthma, but could not improve lung function; however, the in ammatory markers studied vary among studies [20].
As studies that measured general in ammatory markers, like eosinophil count [27] and estimated sedimentation rate (ESR) [26], have declared the necessity to use more speci c markers. In this study, we measured the change in serum levels of periostin on patients with asthma receiving atorvastatin for the rst time, introduced as a novel biomarker in such patients [33]; but, the results of our study showed no signi cant difference between the groups in serum levels of periostin.
One of the limitations of the present study was that we did not examine the changes in airway tissue, endothelial, or cellular levels for a more accurate assessment of any change in patients' airway hyperresponsiveness and in ammation, and only investigated clinical and serological parameters, while histopathological assessment of the patients could guide us better in this regard. Furthermore, follow-up period of the present study was short and the sample size was limited, although they were acceptable compared with similar studies.
As the intervention was on both sexes, young and middle age patients with varying degrees of body weights, this suggested that the results of the present study could be attributed to a wide range of asthmatic patients.

Conclusion
The results of the present RCT showed no signi cant difference in spirometric parameters of patients with asthma receiving 40 mg/day atorvastatin or placebo for 8 weeks. The serological results also determined neither signi cant change in eosinophil count or periostin levels, nor any difference between the groups in terms of these in ammatory biomarkers. The only signi cant difference between the groups in this study was the greater change in asthma control scores in the intervention group compared with the placebo group, as well as signi cant improvement in the asthma control scores in the intervention group. However, as far as this clinical nding was not supported by lung function test or in ammatory markers, we cannot conclude signi cant e cacy of atorvastatin on asthma control, and this effect was presumably because of other effects, such as increased patients' awareness on controlling asthma. Accordingly, further studies with larger sample size and longer follow-up periods are required to determine the exact effect of statins, such as atorvastatin, on asthma control. We suggest measurement of more accurate parameters, like histopathological changes in future studies.  Flow diagram for study enrollment Figure 2 Comparison of mean asthma control scores between the groups

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download. CONSORT2010Checklist.doc