Both SP-D and KL-6/MUC1 are established biomarkers of the interstitial pneumonias, including idiopathic pulmonary fibrosis (IPF), but the causes and clinical outcomes based on their independent effects are not known. Eleven asymptomatic patients, detected with honeycombing on high-resolution computed tomography (HRCT), were compared with 17 other IPF outpatients having slight respiratory symptoms and honeycombing as well. Although SP-D was increased in both groups, KL-6 was significantly higher in the symptomatic IPF group. When the patients (
Although the prognosis for patients with idiopathic pulmonary fibrosis (IPF) is very poor after a diagnosis of respiratory symptoms, that is, a median survival of 2–4 yrs [
Several studies have reported patients in early stages of interstitial lung disease (ILD) [
SP-D and KL-6/MUC1 are useful biomarkers in the diagnosis of various ILDs, such as IPF, collagen vascular disease-associated interstitial pneumonitis, radiation pneumonitis, hypersensitivity pneumonitis, pulmonary alveolar proteinosis, and drug-induced pneumonitis [
The initial aim of this study was to reveal the characteristics of the biomarkers SP-D and KL-6/MUC1 in early-stage IPF. In 2002, we found out that asymptomatic patients with honeycombing on HRCT have elevated SP-D but not KL-6/MUC1 in their serum. Therefore, as second aim, to evaluate the prognostic value of the divergence between SP-D and KL-6/MUC1, we surveyed the outcome of these patients for 8–10 years. We also performed an immunohistochemical study to confirm the dissociation of these biomarkers in the lung tissues obtained by surgical lung biopsy.
We conducted a retrospective study of 28 outpatients at Tohoku University Hospital and Sendai Kousei Hospital, Sendai, Japan, between 1999–2001. All patients met radiological findings of definite usual interstitial pneumonia (UIP) pattern on HRCT, (1) subpleural, basal predominance, (2) reticular abnormality, (3) honeycombing with or without traction bronchiectasis, and (4) absence of features listed as inconsistent with UIP pattern. In addition, none of these patients had any evident disease, such as collagen diseases or hypersensitivity pneumonia, reportedly causative of ILD. Bacteriological examinations were of no clinical significance in all patients. In the 28 patients, two groups, asymptomatic (
We surveyed the outcome of the patients by telephone call or postal mail in August 2009. We could not contact one patients of each in SP-D and SP-D
Measurements of vital capacity (VC), FVC, FEV1, ratio of FEV1 to FVC (FEV1.0%), total lung capacity (TLC), ratio of residual volume (RV) to TLC (RV/TLC), and the diffusing capacity of the lung for carbon monoxide (DLCO) were made using standard equipment (CHESTAC-55V; CHEST Co., Tokyo, Japan) according to the ATS recommendations [
Peripheral venous blood samples were collected between 1999–2001. Each serum sample was analyzed for SP-D and KL-6/MUC1. The serum level of SP-D was measured by a commercially available enzyme-linked immunosorbent assay (ELISA) kit (Yamasa Co., Japan), as previously described [
All patients first underwent conventional CT scanning of the chest using 10-mm thick sections. The HRCT scans also were performed within two weeks of the collection of the blood sample. The findings on HRCT were evaluated by discussion among the authors, including a radiologist. On the basis of previous study results [
Specimens of the three IPF lung tissues were obtained by video-assisted thoracoscopic surgery at Tohoku University Hospital. Lung specimens were fixed in 10% buffered formalin for 18 h and embedded in paraffin for immunohistochemistry. The antibody against KL-6 was provided by Sanko Junyaku Co., Ltd. (Tokyo, Japan) and SP-D by YAMASA (Tokyo, Japan); both of which are commonly used for detecting the serum levels of these biomarkers. An antihuman podoplanin monoclonal antibody (AngioBio Co., Del Mar, CA) was used for detecting lymphatic endothelial cells, and antibodies against human CD34 (Nichirei) and von Willebrand factor (vWF) (Nichirei) were used for endothelial cells in the lung. The antigen-antibody complex was visualized using Vector Red (Vector Laboratories, Burlingame, CA) and/or diaminobenzidine (DAB), and counterstained with elastica-Goldner stain, modified elastica-Mason stain. For immunofluorescent staining, the Alexa Fluor Dye (Molecular Probes, Invitrogen) appropriate for each antibody was used according to the manufacturer’s instructions.
For comparison between asymptomatic and symptomatic patients (Table
Characteristics of asymptomatic and symptomatic patients with honeycombing on HRCT.
Asymptomatic | Symptomatic | ||
---|---|---|---|
Patients (M/F) | 10/1 | 15/2 | |
Age yrs | 67 ± 3 ( | 66 ± 2 ( | .32 |
Smoking Current/Ex/Never | 5/2/1 | 5/4/5 | .44 |
Pack-yrs | 33.3 ± 6.3 ( | 33.1 ± 7.6 ( | .60 |
VC %pred | 96.9 ± 6.1 ( | 88.1 ± 3.6 ( | .25 |
FEV1.0/FVC (%) | 82.3 ± 3.4 ( | 84 ± 2.2 ( | .61 |
DLCO %pred | 79.3 ± 6.1 ( | 58.3 ± 5.8 ( | .031* |
RV/TLC % | 32.6 ± 1.8 ( | 34.5 ± 2.0 ( | .36 |
Data are presented as mean ± S.E. VC: vital capacity. EFV1.0: forced expiratory volume in one second. FEV: forced expiratory volume. DLCO: diffusing capacity of lung for carbon monoxide. RV/TLC: Residual volume/Total lung capacity. *: <.05.
Characteristics of patients in SP-D and SP-D&KL-6 group.
SP-D group | SP-D&KL-6 group | ||
---|---|---|---|
Age yrs | 71 ± 2.2 ( | 66 ± 2.1 ( | .19 |
Pack-yrs | 28.6 ± 11.5 ( | 31.0 ± 12.2 ( | .72 |
VC %pred | 102.4 ± 7.3 ( | 82.5 ± 4.5 ( | .045* |
FEV1.0/FVC (%) | 82.9 ± 5.1 ( | 87.3 ± 2.1 ( | .69 |
DLCO %pred | 71.6 ± 16.5 ( | 61.6 ± 7.8 ( | .051 |
RV/TLC (%) | 34.9 ± 4.1 ( | 32.9 ± 1.7 ( | .40 |
SP-D ng/ml | 336 ± 40.3 ( | 375 ± 30.8 ( | .48 |
KL-6/MUC1 U/ml | 393 ± 20.5 ( | 1439 ± 169 ( | <.001* |
HRCT scan fibrotic area (%) | |||
Aortic arch | 0.7 ± 0.2 ( | 14.9 ± 4.7 ( | <.001* |
Carina | 1.1 ± 0.5 ( | 15.8 ± 5.6 ( | <.001* |
Top of diaphragm | 5.4 ± 2.4 ( | 33.4 ± 7.6 ( | <.001* |
Data are presented as mean ± S.E. SP-D group: SP-D > 220 ng/ml and KL-6 < 500 U/ml; SP-D&KL-6 group: SP-D > 220 ng/ml and KL-6 > 1000 U/ml. *:
The clinical characteristics and pulmonary function data are summarized in Table
The level of serum SP-D and KL-6/MUC1 was compared among asymptomatic and symptomatic IPF patients (Figure
Comparison of serum levels of SP-D and KL-6/MUC1 between asymptomatic patients with honeycombing and symptomatic IPF patients. Both groups have elevated SP-D without statistic difference, but KL-6/MUC1 was significantly higher in symptomatic IPF patients. Dotted lines represent mean values of SP-D (110 ng/ml) and KL-6/MUC-1 (500 U/ml).
To analyze the clinical value of SP-D and KL-6/MUC1, we extracted two groups from these 28 patients by the elevated biomarkers, an SP-D group (
To analyze the relation of biomarkers and fibrotic lesions with honeycombing on HRCT, we measured the distributional area of fibrotic lesions with honeycombing in three levels of the HRCT of each patient (Upper; aortic arch, Middle; carina, Lower; top of diaphragm). The mean percentage of the area of fibrotic lesions with honeycombing was significantly lower in the SP-D group than the SP-D&KL-6 group Table
We surveyed the clinical outcome of patients in the SP-D group and the SP-D&KL-6 group for 8–10 years after initial detection. Because we could not contact one patient in each group, we compared the overall survival between the SP-D only group (
The Kaplan-Meier survival curves of patients with honeycombing in SP-D and SP-D&KL-6 groups. Patients in SP-D group (in red) had longer survival (mean survival time, MST; 3117 days) than those in SP-D&KL-6 group (in blue) (MST; 1232 days) (
There is a clear difference in the immunohistochemical distribution pattern between SP-D and KL-6/MUC1. While both of the biomarkers are produced by alveolar type II cells and bronchiolar epithelial cells, the immunoreactivity for SP-D is cytoplasmic, while KL-6 is distributed at extracellular sites facing the alveolar space or airway space (Figures
Immunohistochemical distribution of SP-D and KL-6/MUC1 in control and IPF lungs. ((a) and (b)) In control lungs, alveolar type II epithelial cells show cytoplasmic staining for SP-D (in red). In contrast, KL-6/MUC1 (in green, (b)) distributes extracellular surface facing alveolar space without cytoplasmic or basal-lateral side expression. The dotted square in (a) is the same section with (b) observed by fluorescence (scale bar = 10
Immunohistochemistry of SP-D and KL-6/MUC1 in IPF lungs. Double-fluorescent immunohistochemistry for SP-D (in red) and KL-6/MUC1 (in green) in IPF lung tissues revealed SP-D-positive materials within both pulmonary veins (PV) and lymphatics (L). In contrast, KL-6/MUC1 was not included in pulmonary veins but only in lymphatics (scale bars = 10
It was shown here that (1) the serum level of KL-6/MUC1 was maintained within the normal range in dissociation with the SP-D level, which was elevated in asymptomatic patients with honeycombing on HRCT, (2) these patients with high SP-D but low KL-6 in their serum exhibited prolonged survival, (3) there are obvious differences in the distribution patterns between SP-D and KL-6/MUC1 in the IPF lung, especially in the clearance route, which may cause the observed dissociation between these biomarkers.
The finding of honeycombing on HRCT is believed to be mostly associated with usual interstitial pneumonia (UIP), as it is reported that honeycombing in at least one lobe indicate UIP, with a 90% sensitivity and 86% specificity [
Our survey of the survival of these patients clearly noted that patients with a high level of SP-D but low KL-6 in their serum had a better prognosis. Several studies have showed the prognostic value of the biomarkers, SP-A, SP-D, and KL-6 at the initial visit [
The different mechanistic roles of the biomarkers SP-D and KL-6 are explained in part from the immunohistochemical study results, which revealed the different distribution patterns of these biomarkers in IPF lungs (Supplement Figure 1). KL-6 is expressed at the extracellular surface of alveolar type II cells and bronchiolar epithelial cells, and from these cells is released into the alveolar lining fluid, accumulated in honeycomb cysts. Since KL-6 is a high molecular weight mucin-like glycoprotein (molecular weight > 1,000 KD) [
There are several limitations in our study. First, the number of subjects in this study is low. And we did not evaluate dyspnea quantification by means of scales, such as MRC, in initial subjects enrollment. Second, we could not analyze changes of lung function from baseline, because of retrospective study. Therefore, we could not evaluate regression analysis between biomarkers concentration and lung functional changes from baseline.
In conclusion, SP-D and KL-6 are useful biomarkers for differentiating patients with honeycombing on HRCT into the stable and progressive stages, which would inform clinical implication for IPF.
This study was partly supported by a grant to the Diffuse Lung Diseases Research Group from the Ministry of Health, Labour and Welfare, Japan.