Systemic inflammatory profile and response to anti-tumor necrosis factor therapy in chronic obstructive pulmonary disease

Abstract

Background: Chronic obstructive pulmonary disease (COPD) is characterized by progressive worsening of airflow limitation associated with abnormally inflamed airways in older smokers. Despite correlative evidence for a role for tumor necrosis factor-alpha in the pathogenesis of COPD, the anti-tumor necrosis factor-alpha, infliximab did not show clinical efficacy in a double-blind, placebo-controlled, phase II clinical trial. This study sought to evaluate the systemic inflammatory profile associated with COPD and to assess the impact of tumor necrosis factor neutralization on systemic inflammation.

Methods: Serum samples (n = 234) from the phase II trial were collected at baseline and after 24 weeks of placebo or infliximab. Additionally, baseline serum samples were obtained from an independent COPD cohort (n = 160) and 2 healthy control cohorts (n = 50; n = 109). Serum concentrations of a broad panel of inflammation-associated analytes were measured using a 92-analyte multiplex assay.

Results: Twenty-five proteins were significantly elevated and 2 were decreased in COPD, including highly elevated CD40 ligand, brain-derived neurotrophic factor, epidermal growth factor, acute-phase proteins, and neutrophil-associated proteins. This profile was largely independent of smoking status, age, and clinical phenotype. The majority of these associations of serum analytes with COPD are novel findings. Increased serum creatine kinase-muscle/brain and myoglobin correlated modestly with decreased forced expiratory volume at 1 second, suggesting cardiac involvement. Infliximab did not affect this systemic inflammatory profile.

Conclusions: A robust systemic inflammatory profile was associated with COPD. This profile was generally independent of disease severity. Because anti-tumor necrosis factor-alpha did not influence systemic inflammation, how to control the underlying pathology beyond symptom suppression remains unclear.

Trial registration: ClinicalTrials.gov NCT00056264.

Figure 1

Serum analytes associated with COPD vs controls. Hierarchical clustering (average linkage with Euclidean distances) was performed restricted to the analytes reported in Table 2 to be associated with COPD vs controls. Serum concentrations were normalized as log₂ of fold over the geometric mean (G_m) of the control cohorts and presented as a heatmap with subjects across x-axis and analytes on y-axis. Disease status, smoking status, and GOLD stage are represented at bottom. IGF-1, insulin-like growth factor-1; Ig, immunoglobulin; EGF, epidermal growth factor; TNF-RII, tumor necrosis factor-receptor II; IL, interleukin; TIMP-1, tissue inhibitor of metalloproteinases-1; VEGF, vascular endothelial growth factor; EN-RAGE, extracellular newly identified-receptor for advanced glycation end-binding protein; RANTES, regulated upon activation, normally T-cell expressed, and secreted; ENA-78, epithelial-derived neutrophil activating protein-78; PAI-1, plasminogen activating factor-1; MCP-1, monocyte chemoattractant protein-1; MIP-1beta, macrophage inflammatory protein-1beta.

Figure 2

Baseline serum analyte concentrations in current smokers. The serum concentrations of the indicated analytes in healthy control and COPD cohorts, restricted to only current smokers, are shown for subgroups < 50 (n = 36 and 22 for controls and COPD, respectively) or between 50- to 70-years-old (n = 16 and 107 for controls and COPD, respectively). Data presented as box (median, interquartile range) and whiskers (10^th-90^th percentiles). Dotted line indicates LDD. *COPD vs control, False discovery rate < 10^-8 within 50-70 year age group. BDNF, brain-derived neurotrophic factor; CD40L, CD40 ligand; EGF, epidermal growth factor; IgE, immunoglobulin E.

Figure 3

Correlations of serum analyte concentrations and FEV₁. Serum concentrations of the indicated analytes (y-axes) vs percent predicted FEV₁ (x-axes) are plotted for the T54 COPD cohort at baseline week 0 visit (left) and week 24 visit (middle) and for the BioServe COPD cohort (right). Correlation coefficients and significance of correlations are reported for each plot.

Figure 4

Changes in analyte concentrations after infliximab treatment. The serum concentrations of the indicated analytes at the 24-week timepoint for placebo and 3 and 5 mg/kg infliximab treatment groups in the T54 COPD cohort are shown as signed-fold over respective baseline concentrations. Data presented as box (median, interquartile range) and whiskers (10^th-90^th percentiles). *p < 0.05 for placebo vs 3 and 5 mg/kg infliximab groups combined; † p < 0.05 for 5 mg/kg infliximab group vs placebo. IL, interleukin; IGF-1, insulin-like growth factor-1; CRP, C-reactive protein.