Sequential changes of serum KL-6 predict the progression of interstitial lung disease

Abstract

Background: Interstitial lung disease (ILD) is a slowly progressing fatal fibrotic lung disease with a widely variable clinical course and a poor prognosis. Clinicians and patients would benefit from a highly efficient and accurate predictor for ILD. The purpose of this study was to evaluate whether blood biomarkers can predict ILD progression.

Methods: In this study, 85 patients diagnosed as having ILD at the Guangzhou Institute of Respiratory Health participated, including 20 patients with idiopathic pulmonary fibrosis (IPF). During the mean follow-up time of 12 months, every patient was examined during four or five visits in our center. Serum samples were collected at baseline, and after 1, 2, 6, and 12 months and tested for the Klebs von den Lungen-6 (KL-6) concentration. Dynamic fluctuations in this biomarker concentration were examined using a logistic regression model to see if they reflected the progression of ILD.

Results: The baseline levels of serum KL-6 in the ILD patients were significantly increased compared to healthy controls. Serum KL-6 levels were significantly elevated in patients with progression of disease (1,985.2±1,497.8 vs. 1,387.6±1,313.1 µg/mL; P<0.001). Logistic regression revealed sequential changes of KL-6 was a significant predictor of ILD progression in the next follow-up (OR, 2.569; 95% CI, 2.260-2.880; P=0.001), and that sequential changes of KL-6 were significant predictors for the progression of IPF (OR, 3.611; 95% CI, 1.048-12.442; P<0.01). Baseline concentrations were not predictive for ILD or IPF. Univariate Cox analysis showed that KL-6 was significantly associated with survival [relative risk (RR), 1.901; 95% CI, 1.294-2.793; P<0.001], along with other variables.

Conclusions: Serum levels of KL-6 were elevated in ILD patients with severe respiratory function compared to those without. The rate of poor prognosis and mortality was associated with increased biomarker concentrations. Sequential measurements of biomarkers could be valuable in disease monitoring and evaluations in clinical management.

Keywords: Interstitial lung disease (ILD); KL-6; biomarker; prediction.

Figure 1

Receiver operating characteristic curve analysis. The curves show the power of initial serum KL-6 for predicting disease progression. At the cut-off level of ≥759.5 U/mL, serum KL-6 levels yielded a sensitivity of 86.4% and a specificity of 41.7% to predict disease progression (AUC =0.655, P<0.0001). KL-6, Klebs von den Lungen-6.

Figure 2

Predicting value of sequential changes in biomarkers for disease progression in ILD patients assessed by logistic regression model. KL-6, Klebs von den Lungen-6; FVC, forced vital capacity; FEV₁, forced expiratory volume in 1 second; DLCO, diffusion capacity for carbon monoxide; TLC, total lung capacity; GGO, ground glass opacities; HC, honeycombing; OR, odds ratio; CI, confidence interval; ILD, interstitial lung disease.

Figure 3

Predicting value of sequential changes in biomarkers for disease progression in IPF patients assessed by logistic regression model. KL-6, Klebs von den Lungen-6; FEV₁, forced expiratory volume in 1 second; DLCO, diffusion capacity for carbon monoxide; TLC, total lung capacity; OR, odds ratio; CI, confidence interval; IPF, idiopathic pulmonary fibrosis.

Figure S1

Comparison of serum KL-6 concentration the FVC >50% group and the FVC <50% group. FVC >50% group n=210, FVC<50% group n=122, an independent sample t-test was performed. KL-6 concentrations were significantly higher in the FVC% <50% group compared to FVC% >50% group, P=0.03. FVC, forced vital capacity. KL-6, Klebs von den Lungen-6.

Figure S2

Comparison of serum KL-6 concentration between the progressive group and the non-progressive group. FVC >50% group n=88, FVC <50% group n=244, an independent sample t-test was performed. KL-6 concentration was highly elevated in progressive compared with non-progressive period, P<0.001. “Progressive” meant over 10% decrease in %FVC or over 15% decrease in %DLCO. “Non-progressive” meant within 10% change of %FVC and within 15% change of %DLCO, or increase in %FVC or increase in %DLCO. KL-6, Klebs von den Lungen-6; FVC, forced vital capacity; DLCO, diffusion capacity for carbon monoxide.

Figure S3

Predicting value for mortality in patients with ILD assessed by Cox regression model. Using Cox regression analysis, KL-6 was the best independent predictor of mortality after adjustment for other covariates (HR, 1.90; 95% CI, 1.29–2.79; P=0.001). Only FVC% is the predictor of mortality in multivariate analysis (n=322). FVC, forced vital capacity; FEV₁, forced expiratory volume in 1 second; DLCO, diffusion capacity for carbon monoxide; TLC, total lung capacity; GGO, ground glass opacity; HC, honeycombing; KL-6, Klebs von den Lungen-6; CI, confidence interval; ILD, interstitial lung disease.