ABSTRACT
OBJECTIVES: Despite the well-established value of currently used classification criteria for the early diagnosis of rheumatoid arthritis (RA) there is a constant demand for novel biomarkers notably in autoantibody-negative patients. Interleukin 7 (IL-7) has been reported as a candidate diagnostic biomarker based on ACR-1987 criteria. However, clinical practice has moved to using the EULAR 2010 classification criteria. Therefore, to advance the use of IL-7 alongside the RA biomarker pipeline, we repeated the original study in a new cohort. METHODS: 255 patients were recruited. IL-7 was quantified by ELISA. Univariate and regression analyses were used to model RA diagnosis. RESULTS: 123 patients were diagnosed with RA (EULAR 2010) while 132 were classified as non-RA. In univariate analysis, RA was associated with autoantibodies and SE-positivity, higher joint counts, DAS28 (all p<0.001) and CRP (p=0.024). IL-7 was lower in RA (p=0.05). Logistic regression analysis in 227 patients with complete data set confirmed IL-7 was the second best predictive marker (p=0.035) following SJC (p=0.007) with good model fit (AUROC=0.889). A second model investigated 147 ACPA-negative patients: lower IL7 was the second best predictive marker (p=0.075) behind SJC (p=0.013). CONCLUSIONS: This study validates our previous results from a UK cohort using EULAR 2010 criteria although the predictive power associated with IL-7 is lower than in the study using ACR 1987 criteria (both French/UK cohorts). IL-7 remains a potential biomarker for ACPA-negative RA although further validation with larger numbers of ACPA-negative patients is still needed notably to translate these results into clinical applicability.
