Periostin and closing volume in combined pulmonary fibrosis and emphysema

Background Combined pulmonary fibrosis and emphysema (CPFE) is an entity characterized by the presence of emphysema in upper lobes and fibrosis in lower lobes. Due to the presence of the two diseases concomitantly, it may be difficult to diagnose. This study aims at a better understanding of this entity and proposes biological markers (functional and biochemical) that help in this characterization. A prospective, observational, cross-sectional study was carried out at a reference center. Pulmonary function tests (spirometry, CO-diffusion capacity, plethysmography and single-maneuver nitrogen washout test - SBWN 2 ) and biochemical markers (periostin, mucin-16, PDGF-BB and TGF-β 1 ) were measured in groups of patients: idiopathic pulmonary fibrosis, CPFE and chronic obstructive pulmonary disease (COPD).


Introduction
Idiopathic pulmonary fibrosis (IPF) is a chronic lung disease characterized by progressive interstitial fibrosis of unknown cause. 1 Some patients, typically those with a high smoking load, develop study.
Patients with IPF, CPFE, and chronic obstructive pulmonary disease (COPD) were selected. The diagnosis of IPF was based on the criteria of the American Thoracic Society, European Respiratory Society, Japanese Respiratory Society, and Latin American Thoracic Association 2011. 14 All selected patients showed consistent clinical presentation and a tomographic pattern of usual interstitial pneumonia (UIP). The diagnosis of CPFE was based on the original report by Cottin et al. 2 as a syndrome characterized by upper lobe emphysema (well-demarcated areas of decreased attenuation with very thin wall (<1 mm) or no wall) and pulmonary fibrosis of the lower lobes (presence of features consistent with usual interstitial pneumonia) according to high resolution chest CT (HRCT).
All cases of interstitial lung disease (IPF and CPFE) were confirmed by a multidisciplinary team and patients with collagen vascular disease, hypersensitivity pneumonia or other interstitial lung disease (ILD) were excluded. We chose to include only patients with a firm diagnosis of IPF and CPFE in order to avoid bias due to inclusion of patients with other type of interstitial lung diseases with different prognosis. All tests performed in the IPF and CPFE groups were obtained before patients started antifibrotic therapy.
Patients with chronic obstructive pulmonary disorder (COPD) were diagnosed according to the Global Initiative for Chronic Obstructive Lung Disease (GOLD) 2017. 15 COPD patients with emphysema phenotype were selected based on clinical and functional features (patients with < 2 exacerbations/year and carbon monoxide diffusion capacity <50%). Patients with other respiratory diseases such as asthma, bronchiectasis, and severe neuromuscular diseases that could alter lung function were excluded. Patients who were in infectious exacerbation at the time of selection were included only 3 months after clinical improvement.
All patients were interviewed by the examiner, after which peripheral blood samples were collected to measure periostin, mucin-16, PDGF-BB, and human TGF-β 1 and for functional evaluation (spirometry, measurement of the carbon monoxide diffusion capacity (DLco) and transfer coefficient (Kco), plethysmography of whole body, and single-breath nitrogen washout test). All tests were performed by a single operator and with the same equipment, HD CPL (nSpire Health Inc., Longmont, CO, USA) and followed American Thoracic Society standardization. 16 Knudson's equations for spirometry 17 and Neder's equations 18,19 were used to determine static pulmonary volumes, carbon monoxide diffusion capacity, and the transfer factor. The single-breath nitrogen washout test was performed (SBWN 2 ) using the HD PFR 3000 apparatus (nSpireHealth, Inc.). The single-breath maneuver washout was performed according to the recommendations of the American Thoracic Society/European Respiratory Society 20 and analysed using Buist equations. 21 The concentration level of each biomarker in the sera of patients was determined with enzyme-linked post hoc were used to differentiate groups. Differences were considered significant when the p value was less than 0.05.

Results
A flow diagram design is shown in Figure 1. The ages of patients in the three groups (IPF, CPFE, and COPD) and the degree of dyspnea as assessed by the modified Medical Research Council scale did not significantly differ between groups. However, there was a significant difference in smoking load among the three groups; COPD patients showed the highest smoking load (Table 1). Turkey´s posttest analysis showed that the smoking load in the CPFE group was significantly higher than in the IPF group.
Functional analysis of the three groups revealed a significant difference in many of the parameters studied. Statistically significant values of forced expiratory volume in the first second (FEV 1 ), DLco, Kco, residual volume (RV)/total lung capacity (TLC), and TLC (%) were observed in the CPFE group compared to in the COPD group. However, no significant differences were observed between the IPF group and CPFE group ( Table 1). The parameters evaluated by SBWN 2 , phase III slope (SIII), and ΔN 2 750-1250mL presented values above the normal range in all groups, but the differences were not significant. However, the closing volume (CV)/vital capacity (VC) (%) and closing capacity (CC)/TLC (%) ratios were significantly higher in individuals with CPFE than in those with IPF (p = 0.005 and p = 0.03, respectively, according to Turkey´s post-test). These data are presented in Table 2.
In the analysis of serum biomarkers, according to Table 3

Discussion
Previous studies suggested that patients with CPFE form a subgroup of IPF or have a different disease, 14 as they present with different clinical, radiological and functional alterations than patients with IPF. 22 Isolating this group of patients can facilitate diagnosis and enable early treatment.
Evaluating CPFE in a study series is difficult because many patients with various types of interstitial diseases are included, such as patients with connective tissue diseases, particularly rheumatoid arthritis 23 and non-specific fibrotic interstitial pneumonia. 24 The inclusion of several standards makes it difficult to assess prognosis. We included only patients with a well-established UIP pattern and identified important differences from a functional perspective and in blood biomarker profiles. In the blood samples, patients with CPFE had higher levels of periostin while patients with IPF had higher levels of mucin-16. Functionally, CV/VC (%) and CC/TLC (%) were higher in patients with CPFE than in those with IPF, suggesting increased air trapping associated with emphysema.
The smoking history was common in the three groups, while the smoking load was significantly higher in the COPD group compared to the other groups and higher in the CPFE group than in the IPF group.
Patients with CPFE are typically smokers or former smokers with a history of smoking over 40 packets per year. 25 This suggests a minimum smoking load leading to progression of the combined disease.
However, CPFE has been reported in patients with no smoking history, suggesting that there is a genetic predisposition for the development of the syndrome. 24 As in previous studies, differentiating between IPF and CPFE by spirometry was difficult. 2,26 Patients with CPFE showed flow measurements close to those with IPF and volume measurements close to those in patients with COPD. No spirometric variable could differentiate CPFE from the IPF, which agrees the results of previous studies. 22 This may delay diagnosis, as spirometry is the most easily accessible pulmonary function test. 27 Although some studies reported lower values of DLco in patients with CPFE compared to in IPF, 24 including in our study and a study by Jacob et al 28 , similar values were observed in both groups (IPF and CPFE). Interestingly, Kco was lower in the CPFE group. The coexistence of emphysema and fibrosis leads to normal or subnormal volumes and pulmonary fluxes, while DLco was substantially reduced. 24,28 SBWN 2 is an important tool for evaluating the homogeneity of alveolar ventilation and for the early detection of small airway changes in patients with COPD. 5 The main parameters obtained by SBWN 2 are SIII and CV, which are directly correlated with FEV 1 and FVC in patients with COPD. 5 SIII is characterized by changes in the N 2 concentration expired between 25% and 75% of the VC and ΔN 2 750-1250mL . Both parameters reflect the distribution of ventilation, ie, whether there is homogeneity in ventilation. CV is the portion of the VC that begins after the start of airway closure (phase IV) and affects the RV. In ordinary individuals, this value is <20% of VC; higher values are observed in both obstructive and restrictive patients. The CC is the CV associated with RV. 4 No studies comparing SBWN 2 have been conducted in patients with IPF and CPFE. Silva et al.
conducted SBWN 2 in patients with systemic sclerosis and pulmonary involvement and compared them with a healthy population. They found that SIII was the most sensitive pulmonary function alteration in these patients, even when other pulmonary function tests were normal. 29 The SBWN 2 test showed heterogeneity in pulmonary ventilation in all three conditions (IPF, CPFE and COPD), without large differences between groups, which did not aid in the functional differentiation between IPF and CPFE. Additionally, our data suggested that patients with CPFE experienced significant air trapping, even when the lung volumes were within normal limits. Patients included in our study had evident fibrosis in CT scanning. Thus, SBWN 2 may be a sensitive test for evaluating the progression of fibrosis in patients with initial fibrosing disease.
Periostin levels were significantly higher in the serum of the CPFE group than in the other two groups.
Periostin is involved in processes that lead to pulmonary fibrosis formation and pulmonary remodeling, as well as in eosinophil recruitment and mucus production. 6 Periostin has been evaluated in the context of pulmonary fibrosis and asthma, but a recently published study revealed no significant difference between the levels of periostin in asthma compared to in COPD. 30 Several studies suggested that periostin is a biomarker in patients with IPF, 6,7 but no studies have evaluated periostin behaviour in CPFE. Caswell-Smith et al. reported reference values for periostin in adults without asthma and no COPD at 50 ng/mL. 30 In our study, only the CPFE group showed values higher than the reference. As this biomarker may be increased in both fibrosis and COPD, it is possible that the association of the two changes contributed to the outstanding increase in patients with CPFE.
Mucin-16 (MUC16 or CA-125) is a tumor biomarker widely evaluated in medical practice and is mainly related to gynecological tumors, such as those in the ovaries. 12,13 However, some authors observed increased levels (>35 U/mL) in patients with interstitial lung diseases, particularly IPF, even suggesting a worse prognosis with increased mortality and increased risk of lung cancer. 12,13 We found that patients with IPF had higher levels of mucin-16 than patients with CPFE. This may be related to the fibrosis degree. One hypothesis to explain this results is that the presence of emphysema reduces lung density, as mucin-16 is present on the pulmonary epithelial surface.
PDGF-BB and TGF-β 1 have been proposed as biomarkers for differentiating between IPF and CPFE.
These molecules participate in collagen formation and complement system regulation 9-11 and may help in this differentiation; however, the data revealed no significant difference in PDGF-BB and TGFβ 1 among the three groups.
Because of the cross-sectional design of the study, the prognostic role of these functional and biochemical markers could not be assessed. Another limitation of this study was that the study was conducted in a single center and included only patients with an evident UIP pattern, and it was not possible to evaluate whether the observed changes also occurred in patients with incipient lesions. In addition, the lack of a control group with healthy patients may be a limiting factor. However, this study provides insight useful for future studies that include patients with IPF and CPFE. Conclusions SBWN 2 is an easy test to perform and may assist in the differential diagnosis of IPF and CPFE.
Additionally, periostin appears to be an important biomarker for differentiating CPFE from IPF, just as mucin-16 appears to be related to IPF. These data suggest that CFPE has distinct functional and serologic characteristics from isolated IPF.