Performance analysis of Luminex and ELISA to profile serum IP-10 as a biomarker in systemic lupus erythematosus.

ABSTRACT

OBJECTIVE: Interferon-γ inducible protein-10 (IP-10) is a promising biomarker in systemic lupus erythematosus (SLE). The optimal quantification platform has not yet been identified. We compared the performance of bead-based multiplex assay (Luminex) and high-sensitivity enzyme-linked immunosorbent assay (hs-ELISA) for profiling serum IP-10 as a biomarker of lupus activity. METHODS: A cross-sectional study was conducted on outpatients with SLE. Serum IP-10 was measured simultaneously on Luminex and hs-ELISA, and correlation between platforms was assessed. Additionally, IP-10 levels were tested against disease activity and organ involvement. RESULTS: One-hundred and forty-one patients (88% women; 38 years old) were studied. Median IP-10 levels were 100.9 (125.2) pg/mL by Luminex and 156.5 (191.7) pg/mL by hs-ELISA. Correlation analysis showed Spearman's ρ = 0.621 (p < 0.0001) between Luminex and hs-ELISA. Quantification of IP-10 by Luminex showed a significant correlation (ρ = 0.198; p = 0.021) with disease activity, while this was not observed (ρ = 0.036; p = 0.683) when measured using hs-ELISA. Serum IP-10 levels were lower in quiescent patients than in those with active disease (70.8 [68.4] versus 114.3 [123.9] pg/mL; p = 0.024), with an AUC-ROC = 0.62 (p = 0.029), sensitivity = 47.9%, specificity = 77.5%, and positive likelihood ratio = 2.1. Patients with active arthritis had higher IP-10 levels than non-arthritis patients (158.1 [505.4] versus 94.1 [114.0] pg/mL; p = 0.008), with an AUC-ROC = 0.73 (p = 0.0009), sensitivity = 72.7%, specificity = 66.4%, and positive likelihood ratio = 2.1. No other type of organ involvement was identified by serum IP-10. CONCLUSIONS: Luminex performs better than hs-ELISA as a quantification platform for IP-10 as it correlates with disease activity and identifies active arthritis in SLE.