Clustering-aided prediction of outcomes in patients with idiopathic pulmonary fibrosis

Abstract

Background: Blood biomarkers predictive of the progression of idiopathic pulmonary fibrosis (IPF) would be of value for research and clinical practice. We used data from the IPF-PRO Registry to investigate whether the addition of "omics" data to risk prediction models based on demographic and clinical characteristics improved prediction of the progression of IPF.

Methods: The IPF-PRO Registry enrolled patients with IPF at 46 sites across the US. Patients were followed prospectively. Median follow-up was 27.2 months. Prediction models for disease progression included omics data (proteins and microRNAs [miRNAs]), demographic factors and clinical factors, all assessed at enrollment. Data on proteins and miRNAs were included in the models either as raw values or based on clusters in various combinations. Least absolute shrinkage and selection operator (Lasso) Cox regression was applied for time-to-event composite outcomes and logistic regression with L1 penalty was applied for binary outcomes assessed at 1 year. Model performance was assessed using Harrell's C-index (for time-to-event outcomes) or area under the curve (for binary outcomes).

Results: Data were analyzed from 231 patients. The models based on demographic and clinical factors, with or without omics data, were the top-performing models for prediction of all the time-to-event outcomes. Relative changes in average C-index after incorporating omics data into models based on demographic and clinical factors ranged from 1.7 to 3.2%. Of the blood biomarkers, surfactant protein-D, serine protease inhibitor A7 and matrix metalloproteinase-9 (MMP-9) were among the top predictors of the outcomes. For the binary outcomes, models based on demographics alone and models based on demographics plus omics data had similar performances. Of the blood biomarkers, CC motif chemokine 11, vascular cell adhesion protein-1, adiponectin, carcinoembryonic antigen and MMP-9 were the most important predictors of the binary outcomes.

Conclusions: We identified circulating protein and miRNA biomarkers associated with the progression of IPF. However, the integration of omics data into prediction models that included demographic and clinical factors did not materially improve the performance of the models.

Trial registration: ClinicalTrials.gov; No: NCT01915511; registered August 5, 2013; URL: www.

Clinicaltrials: gov .

Keywords: Biomarkers; Disease progression; Interstitial lung disease.

Fig. 1

Workflow for evaluating the performance of different models

Fig. 2

C-indices of models for (A) composite of time to death or lung transplant and (B) composite of time to death, lung transplant, or decline in forced vital capacity > 10%. The left panel shows models based on demographic and clinical characteristics with or without omics data. The middle panel shows models based on demographics with or without omics data. The right panel shows the model based on omics data alone. Clin, clinical; demo, demographics; lbl_miRNA, cluster label of miRNA; lbl_prot, cluster label of protein; raw_miRNA, raw values of miRNAs; raw_prot, raw values of proteins.

Fig. 3

Variable importance of demographic and omics variables for (A) composite outcome of death or lung transplant and (B) composite outcome of death, lung transplant, or decline in forced vital capacity (FVC) > 10%. Green bars denote proteins, blue bars denote demographic variables and the red bar denotes the cluster label of miRNA. Demo, demographics; lbl_miRNA, clustering label of miRNA; raw_prot, raw values of proteins

Fig. 4

Area under the curve (AUC) of models for absolute decline in forced vital capacity > 10% at 1 year. The left panel shows models based on demographic and clinical characteristics with or without omics data. The middle panel shows models based on demographics with or without omics data. The right panel shows the models based on omics data alone. Clin, clinical; demo, demographics; lbl_miRNA, cluster label of miRNA; lbl_prot, cluster label of protein; raw_miRNA, raw values of miRNAs; raw_prot, raw values of proteins

Fig. 5

Variable importance of demographic and omics variables for absolute decline in forced vital capacity > 10% at 1 year. Green bars denote proteins, blue bars denote demographic variables and the red bar denotes the cluster label of miRNA. Demo, demographics; lbl_miRNA, clustering label of miRNA; raw_prot, raw values of proteins

Fig. 6

Area under the curve (AUC) of models for disease progression (decline in forced vital capacity [FVC] % predicted > 10%, decline in diffusing capacity of the lungs for carbon monoxide [DLco] % predicted > 15%, death, or lung transplant) at 1 year. The left panel shows models based on demographics and clinical characteristics with or without omics data. The middle panel shows models based on demographics with or without omics data. The right panel shows the models based on omics data alone. Clin, clinical; demo, demographics; lbl_miRNA, cluster label of miRNA; lbl_prot, cluster label of protein; raw_miRNA, raw values of miRNAs; raw_prot, raw values of proteins

Fig. 7

Variable importance by selected frequency for disease progression (decline in forced vital capacity [FVC] % predicted > 10%, decline in diffusing capacity of the lungs for carbon monoxide [DLco] % predicted > 15%, death or lung transplant) at 1 year. Green bars denote proteins, blue bars denote demographic variables and the red bar denotes the cluster label of miRNA. Demo, demographics; lbl_miRNA, clustering label of miRNA; raw_prot, raw values of proteins