Serum IP-10 as a biomarker of human rhinovirus infection at exacerbation of COPD

Abstract

Background: Human rhinovirus (HRV) is the most frequent virus associated with COPD exacerbations. Viral infections increase exacerbation severity and likelihood of hospitalization. As ease of sampling blood makes serum a more practical marker than sputum, we investigated whether changes in serum interferon-gamma-inducible protein 10 (IP-10) from baseline to exacerbation were higher in airway HRV-positive exacerbations and whether IP-10 levels related to HRV load.

Methods: One hundred thirty-six patients with COPD and 70 controls were included over 2 years and 72 exacerbations sampled. HRV positivity and load were determined by reverse transcriptase-polymerase chain reaction in nasopharyngeal swabs and/or sputum at baseline and exacerbation. IP-10 was measured by enzyme-linked immunosorbent assay in serum and compared with HRV load.

Results: At baseline, serum IP-10 was higher in patients with COPD than controls; medians were 149.4 pg/mL (103-215) and 111.7 pg/mL (82-178), P = .02. The presence of HRV at baseline did not increase IP-10: patients with COPD, 166.9 pg/mL (110-240) and 149.4 pg/mL (103-215), P = .30; controls, 136.4 pg/mL (77-204) and 111.7 pg/mL (82-178), P = .53. IP-10 increased significantly from baseline to exacerbation in HRV-positive exacerbations: 154.9 pg/mL (114.0-195.1) to 207.5 pg/mL (142.1-333.5), P = .009. There was no change in IP-10 between baseline and exacerbation in HRV-negative exacerbations: 168.3 pg/mL (94.3-249.8) and 175.6 pg/mL (107.2-290.4), P = .49. At exacerbation, IP-10 correlated with sputum viral load: rho = 0.48; P = .02. In receiver operating characteristics analysis, the combination of IP-10 and coryzal symptoms gave an area under the curve of 0.82 (95% CI, 0.74-0.90).

Conclusions: IP-10 increases from baseline to exacerbation in HRV-positive exacerbations and correlates with sputum HRV load. Serum IP-10 may be useful as a novel marker for these events.

Figures

Figure 1.

Flow diagram showing patients with COPD and control subjects studied. HRV = human rhinovirus; NPS = nasopharyngeal swab.

Figure 2.

Median serum IP-10 in 48 controls and 87 patients with COPD when HRV negative in the stable state. Circles represent extreme outliers. IP-10 = interferon-γ-inducible protein 10. See Figure 1 legend for expansion of other abbreviation.

Figure 3.

Increase in IP-10 from baseline to exacerbation in 31 paired COPD exacerbations that were positive for HRV (A). Median serum IP-10 at baseline and exacerbation in 21 paired baseline-exacerbations that were negative for HRV (B). Circles represent extreme outliers. See Figures 1 and 2 legends for expansion of abbreviations.

Figure 4.

At exacerbation IP-10 levels correlated with sputum viral load (A). IP-10 levels did not correlate with CRP (B) or IL-6 (C). CRP = C-reactive protein. See Figures 1 and 2 legends for expansion of other abbreviations.

Figure 5.

Thirty-four patients reporting coryzal symptoms had significantly higher levels of IP-10 than 36 patients not reporting coryzal symptoms. Circles represent extreme outliers (A). In 21 HRV-negative exacerbations in which 11 patients reported cold symptoms there was no difference in IP-10 (B). See Figures 1 and 2 for expansion of abbreviations.

Figure 6.

HRV load in 13 baseline-exacerbation pairs. The horizontal line represents the cutoff chosen for positivity at exacerbation. See Figure 1 legend for expansion of abbreviation.

Figure 7.

ROC curves. Using IP-10 alone, AUC = 0.78 (95% CI, 0.65-0.91) (A). Using the presence of coryzal symptoms alone, AUC = 0.66 (0.51-0.82) (B). Using the combination of IP-10 and coryzal symptoms, AUC = 0.80 (0.66-0.94) (C). AUC = area under the curve; ROC = receiver operating characteristics. See Figure 2 legend for expansion of other abbreviations.