Background/Purpose: Patients with established Rheumatoid arthritis (RA) present evidence for an exhausted immune system. It is not known whether patients seen at the Early inflammatory PolyArthritis (EPA; 1-12 month symptom duration) stage already present with immune exhaustion. Quantification of T-cell receptor excision circles (TREC) in peripheral blood is a readily available measure of T cell thymic output.

Objective. To explore if low TREC numbers in EPA patients are associated with poorer disease outcomes.

Methods: Variables allowing calculation of disease activity indices (DAS-28 and SDAI), hand and feet X-rays scored using the modified Sharp method (SHS) and DNA from peripheral blood were serially collected at inclusion and at pre-specified times during follow up from consecutive EPA patients from the Early Undifferentiated PolyArthritis (EUPA) cohort and from healthy controls with similar sex and age distribution. TREC numbers per 10⁵ nucleated cells in peripheral blood were estimated by quantitative polymerase chain reaction (qPCR) using the method described by Cheynier et al (Methods Mol Biol 2007;380:197-213). The effect of sex and increasing age on log-transformed TREC numbers were analyzed using Pearson correlations and repeated measures ANOVA. Receiver Operating Characteristics (ROC) curves using baseline TREC numbers were drawn to determine the best cut-off to predict significant joint damage (SHS ≥5) at 30 months. Levels of TREC at baseline and outcomes over time were correlated using linear regression, repeated measures ANOVA or generalized estimating equations for binary outcomes (GEE).

Results: We report on 73 EPA patients (93% RA; median duration 4.7 months; 48 women; mean 54.8 years) and on 98 healthy controls with similar age and sex. The median (IQR) number (per 10⁵ nucleated cells) of TREC at inclusion was 88.6 (25.1-269.1) for EPA patients and 129.3 (45.4-289.4) for controls (p=0.112). In both patients and controls, TREC numbers decreased with increasing age, and women as a group had higher numbers. TREC numbers at baseline inversely correlated with disease activity but this correlation was lost during follow up. At baseline, ROC curves revealed that a value of < 128 TREC per 10⁵nucleated cells best identified patients with SHS ≥5 at 30 months (AUC = 0.613; sensitivity = 0.600; specificity = 0.731; RR (95% CI) = 1.91 (1.01-3.31), p=0.024). TREC < 128 also associated with SHS ≥ 5  at baseline (AUC = 0.716; sensitivity = 0.786; specificity = 0.678; RR (95% CI) = 3.77 (1.64-8.66), p<0.001). There was a trend between TREC levels < 128 at baseline and radiographic progression (ΔSHS ≥ 5) from baseline to follow up: RR (95%CI) = 2.24 (0.92-5.50); p=0.077.

Conclusion: Our age and sex results in healthy controls parallel published results. In EPA patients, baseline TREC numbers were correlated with initial but not subsequent disease activity. Low baseline TREC levels were associated with more joint damage at baseline and trended to predict more progression during follow up.  Measurement of TREC numbers represents a novel promising biomarker in EPA patients, potentially identifying patients who already have developed the exhausted RA phenotype, those most likely to develop severe joint damage.

« Back to 2015 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/quantification-of-t-cell-receptor-excision-circles-trec-from-peripheral-blood-in-patients-with-inflammatory-polyarthritis-of-recent-onset-epa-association-with-radiographic-joint-damage/