Department of Pulmonology, University Hospital of Zurich, Raemistrasse 100, 8091 Zurich, SwitzerlandSwitzerland
Also available as PDF (figures and tables appear in the PDF).
Background: Periostin, an extracellular matrix protein, is involved in inflammatory processes of the lung. To date, most studies have focused on periostin in asthma patients, its role in chronic obstructive pulmonary disease (COPD) is less clear and no information has been reported on blood levels of periostin in COPD patients in the context of exacerbation rates. As such, this exploratory study aimed to investigate whether periostin is helpful to distinguish between COPD patients with frequent and infrequent exacerbations. Methods: We performed an examination of patients with COPD participating in a COPD cohort study in Switzerland. Periostin levels were determined in serum samples by using a commer-cially available enzyme-linked immunosorbent assay (ELISA) kit. Patients underwent eval-uation of clinical symptoms including exacerbation rate (exacerbation defined by requiring oral corticosteroids and/or antibiotics) and lung function. In a subgroup of patients an annual follow-up was available that was considered in an additional analysis. Results: Twenty six patients (global initiative of obstructive lung disease (GOLD) stage 1 none, 31% stage 2, 38% stage 3, 31% stage 4) were included in the analysis. The mean±standard deviation (SD) age of the patients was 63±5.9 years, 16 were males, 24 were smokers or ex-smokers. The median (quartiles) post-bronchodilator FEV1% predicted was 36(27/57). There was no significant difference in periostin levels between patients with frequent and infrequent exacerbations. The follow-up data revealed no evidence that periostin is helpful in distinguish-ing frequent from infrequent exacerbators. Conclusion: Our analysis performed in a small group of carefully matched COPD patients demonstrates that there is no significant relationship between exacerbation rate and periostin levels in blood.
Keywords: COPD; Periostin; Eosinophilic inflamation; Exacerbations
ABBREVIATIONS: BMI: Body Mass Index; COPD: Chronic Obstructive Pulmonary Disease; ELISA: Enzyme-Linked Immunosorbent Assay; FEV1: Forced expiratory volume in one sec- ond; GOLD: Global Initiative of Obstructive Lung Disease; SD: Standard Deviation; WIS- DOM: Withdrawal of Inhaled Steroids during Optimised bronchodilator Management; QoL: Quality of Life; CGPS: Copenhagen General Population Study. Trial Registration: https://clinicaltrials.gov/, NCT01527773. Registered 18.01.2012 (retrospec- Copyright tively registered).
©2016 Clarenbach CF. This is an open access article distributed un- To the Editor, der the Creative Commons Attribu- tion 4.0 International License (CC BY 4.0), which permits unrestricted Periostin, a multifunctional matricelluar protein, has been reported to be an excellent predictor use, distribution, and reproduction of airway eosinophilia.1 Eosinophilic inflammation within the airways is usually considered as in any medium, provided the origi- a hallmark of patients with asthma and it was found that the exacerbation rate was significantly nal work is properly cited.
reduced in patients with high levels of periostin. While, to date, most studies have focused on periostin in asthma, its role in COPD is less clear and no infor- compared to their matched peers with infrequent exacerbations, mation has been reported on blood levels of periostin in COPD no signifcant difference in periostin blood levels was detected. patients in the context of exacerbation rates. However, recent Since in a subgroup of patients annual measurements were avail- publications focused on the important role of eosinophils and able, we also analyzed patients that changed their exacerbation 3,4 their association with exacerbation rate.3,4 Moreover, COPD pa- frequency over time, e.g. from frequent to infrequent or vice ver- tients with elevated eosinophil counts also had an increased risk sa. This additional analysis provided no evidence to support that of exacerbations.5 A recent publication implied that high blood periostin is helpful in segregating both groups. No correlation eosinophils and high plasma periostin were associated with im- was found between the level of serum eosinophils and periostin proved lung function three months after starting treatment with in our cohort (Table 1).
inhaled corticosteroids in combination with a long acting bron- chodilator.5 Frequent exacerbations, predominantly found in pa- tients with severe COPD, result in accelerated disease progres- These data suggest that periostin may serve as a bio- sion and mortality.7 These patients usually have a more rapid de- marker in COPD patients. As such, this exploratory study aimed cline in lung function, worse quality of life (QoL) and decreased to gain insight whether periostin is helpful to distinguish be- exercise performance. In a stepwise regression performed in tween COPD patients with frequent and infrequent exacerba- more than 2000 COPD patients the history of previous exac- tions.
erbations was found to be the strongest predictor of subsequent COPD exacerbations.8 We performed an examination of patients with COPD participating in a COPD cohort study from seven pulmonary In addition, a number of previous studies were under- clinics in Switzerland (NCT01527773). The cantonal ethics taken to identify biomarkers that may help to predict exacerba- committee of Zurich approved the study and patients provided tion frequency.9 To date, however, no serum biomarkers reliably written informed consent (KEK-ZH-Nr. 2011-0106). Perios- predict exacerbation frequency in COPD patients.
tin levels were determined in serum samples by using a com- mercially available ELISA kit (Thermo Fisher Scientific, Zug, Blood eosinophils in COPD have been shown to pre- Switzerland). COPD severity was defined according to GOLD dict corticosteroid responsiveness during acute exacerbation. guidelines. Patients underwent evaluation of clinical symptoms Patients with a peripheral blood eosinophil count of ≥2% at the including exacerbation rate (exacerbation defined by requiring onset of an outpatient managed exacerbation responded prompt- oral corticosteroids and/or antibiotics) and lung function. All pa- ly to prednisolone, whereas those with a count of <2% had a tients were examined in a stable phase of the disease and, in case higher rate of treatment failure compared with placebo.10 In a of an acute exacerbation, the evaluation was postponed by at post-hoc analysis of the withdrawal of inhaled steroids during least 6 weeks.
optimised bronchodilator management (WISDOM)-trial Watz and colleagues4 demonstrated that withdrawal of inhaled corti- Comparisons among groups were performed with costeroids in COPD patients with an elevated baseline serum paired t-test, Wilcoxon signed ranks test and chi-square test. eosinophil count led to a higher risk of moderate to severe exac- erbations. The authors concluded that an eosinophil count of 4% Since the number of patients with frequent exacerba- or greater 300 cells per µl may predispose to a deleterious effect tions (≥2 per year) was considerably lower than the number of of inhaled corticosteroid withdrawal.4 A recent analysis of the patients with infrequent exacerbations, we matched every pa- data from the Copenhagen General Population Study (CGPS) tient from the frequent exacerbator group to a patient with infre- revealed that in COPD patients an increased blood eosinophil quent exacerbations. Matching criteria included age, body mass level above 0.34 (gram/liter) g/l was associated with a 1.76-fold index (BMI) and disease severity according to FEV1. Smoking increased risk of severe exacerbations.3 was considered as additional confounder in the analysis because it has been reported recently that periostin may be repressed by While these publications highlight the role of eosino- smoking.6 phils in COPD, in particular in the context of acute exacerba- tions, to the best of our knowledge our study is the first report on Twenty six patients (GOLD stage 1 none, 31% stage the levels of periostin in COPD and exacerbation rate. 2, 38% stage 3, 31% stage 4) were included in the analysis. The mean±SD age of the patients was 63±5.9 years, 16 were males, The current analysis focusing on the role of periostinin 24 were smokers or ex-smokers. The median (quartiles) post- stable patients with COPD however did not help to stratify pa- bronchodilator FEV1% predicted was 36(27/57). There was no tients into frequent or infrequent exacerbators.
significant difference among the two groups for demographic data. In a subgroup of patients an annual follow-up was avail- Smoking was recently postulated to repress periostin able that was considered in an additional analysis. at the individual gene level.6 However, in our study only 1 pa- tient in the infrequent exacerbation group and two patients of the When COPD patients with frequent exacerbations were frequent exacerbators were active smokers and the association All COPD patients (n=26) Infrequent exacerbators (n=13) Frequent exacerbators (n=13) p value Clinical characteristics Age, years 63(5.9) 64(6.2) 61(5.6) 0.059 Male/Female 16/10 8/5 8/5 1.000 Body mass index, kg/m 25.7(23.5/28.1) 26.6(24.7/28.1) 24.5(22.9/26.5) 0.422 Pack years of smoking 38(26.4) 46(29.6) 29(20.7) 0.148 Current smokers, N(%) 3(12) 1(8) 2(15) LAMA+LABA, N(%) 2(8) 2(15) 0(0) 0.157 LAMA+LABA+GC, N(%) 20(77) 10(77) 10(77) 1.000 Lung function FEV1, % pred. 36(27/57) 41(23/57) 33(28/57) 0.779 FVC, % pred. 77(15.7) 78(15.8) 76(16.2) 0.672 TLC, % pred. 115(21.5) 117(21.6) 113(21.9) 0.497 RV/TLC-ratio 138(45.8) 133(55.9) 142(35.1) 0.394 DLCO, % pred. 36(32/47) 36(33/47) 37(32/60) 0.136 Blood gas analysis PaO2, kpa 9.4(8.1/11.0) 8.8(8.1/9.9) 9.5(8.1/11.0) 0.753 PaCO2, kpa 5.0(0.7) 5.2(0.8) 4.8(0.6) 0.113 Laboratory parameters Periostin, ng/ml 9.39(6.79/17.93) 7.85(6.40/9.04) 12.32(9.73/20.21) 0.133 Hemoglobin, g/dl 14.6(1.2) 14.6(1.4) 14.6(1.1) 0.865 Leukocytes, G/l 8.6(2.08) 8.1(2.13) 9.1(1.99) 0.215 Lymphocytes, G/l 1.56(1.27/2.12) 1.60(1.46/1.89) 1.42(1.20/2.42) 0.534 Neutrophils, G/l 6.15(1.93) 5.67(1.72) 6.72(2.08) 0.144 Eosinophils, G/l 0.08(0.05/0.16) 0.08(0.08/0.16) 0.08(0.03/0.16) 0.267 CRP, mg/l 2.5(1.1/4.0) 2.6(1.2/4.0) 1.5(1.0/4.0) 0.382 Values are mean(SD) or median(quartiles) unless otherwise stated; LAMA: long-acting muscarinergic antagonist; LABA: long-acting beta-agonist; GC: glucocorticoids;FEV1: forced expiratory volume in one second; FVC: forced vital capacity; TLC: total lung capacity; RV: residual volume; DLCO: Diffusing capacity of the lung for carbon mon- oxide. *p<0.05 for comparison of differences between the two groups.
between smoking status and periostin level was not significant. CONSENT Moreover, recent findings from a large cross-sectional study sug- gested that serum levels of periostin do not have to be adjusted The Cantonal Ethics Committee of Zurich approved the study for smoking history, age and gender.11 As such, it is unlikely that and patients provided written informed consent (KEK-ZH-Nr. the levels of periostin in our cohort have been under estimated. 2011-0106). The study was performed in accordance with the Declaration of Helsinki.
In summary, our observation, although limited by the small size, suggests that serum periostin levels do not help dis-