Serum tenascin-C discriminates patients with active SLE from inactive patients and healthy controls and predicts the need to escalate immunosuppressive therapy: a cohort study

Abstract

Introduction: The aim of this study was to examine whether circulating levels of the proinflammatory glycoprotein tenascin-C (TNC) are useful as an activity-specific or predictive biomarker in systemic lupus erythematosus (SLE).

Methods: Serum TNC levels were determined by enzyme-linked immunosorbent assay at inception visit in a prospective cohort of 59 SLE patients, and in 65 healthy controls (HC). SLE patients were followed for a mean of 11 months, disease activity was assessed using the Systemic Lupus Erythematosus Disease Activity Index 2000 (SLEDAI-2 K) and British Isles Lupus Assessment Group disease activity index (BILAG-2004), clinical and laboratory data were recorded every 3-6 months, and changes in glucocorticoids (GC) and immunosuppressants (IS) were recorded serially. We examined cross-sectionally the relationships between serum concentrations of TNC and SLE status, SLEDAI-2 K scores, strata of disease activity, and levels of conventional biomarkers [anti-double-stranded DNA (dsDNA), anti-nucleosome antibodies, C3 and C4]. We also explored the utility of TNC levels for predicting disease flares, defined as (i) new/increased GC, (ii) new/increased GC or IS, and (iii) increase in SLEDAI by ≥3 or (iv) BILAG A or B flare.

Results: There was no significant difference in the mean levels of TNC between the SLE patients and HC. However, in SLE patients with active disease (SLEDAI ≥6), the TNC levels were significantly higher than in the HC (p = 0.004) or in patients with no/low disease activity (p = 0.004). In SLE patients, TNC levels were significantly associated with positivity of anti-dsDNA (p = 0.03) and anti-nucleosome antibodies (p = 0.008). Flares defined by a need to escalate immunosuppressive therapy were captured more frequently and earlier than flares defined by standard activity indices. Higher baseline levels of serum TNC presented a significantly greater risk of flare (i) [hazard ratio (HR) 1.39, 95% confidence interval (CI) 1.11-1.73] or (ii) (HR 1.25, 95% CI 1.02-1.52) but not of flares (iii) or (iv). The baseline serum TNC level was the single most important independent predictor of flare (i) compared with conventional biomarkers.

Conclusions: TNC is not disease-specific, but it seems to indicate the activity of SLE and may predict the need to escalate immunosuppressive therapy. TNC levels may thus serve as a useful activity-specific and predictive biomarker in SLE.

Fig. 1

Mean levels of tenascin-C in healthy controls and patients with systemic lupus erythematosus with low [Systemic Lupus Erythematosus Disease Activity Index 2000 (SLEDAI-2 K) <6] and high (SLEDAI-2 K ≥6) disease activity

Fig. 2

Receiver operating characteristic curve analysis of tenascin (TNC) serum levels as a predictor of active systemic lupus erythematosus (defined as Systemic Lupus Erythematosus Disease Activity Index 2000 ≥ 6). At the optimal cutoff point of 659 ng/ml, the area under the curve for TNC serum levels that discriminated between active and inactive disease was 0.69 (95 % CI 0.53–0.86, p = 0.02) with a sensitivity of 53 % and specificity of 92 %

Fig. 3

Differential in time to flare according to the definitions of flare used. Systemic Lupus Erythematosus Disease Activity Index 2000 (SLEDAI) and British Isles Lupus Assessment Group disease activity index (BILAG) scores were collected only at prespecified time points (months 0, 3 and 6 and then every 6 months; however, the graphed points are exact dates when the per-protocol visits actually occurred), and reflected only the systemic lupus erythematosus activity within a maximum 30-day time window before each visit. Changes in glucocorticoids and/or immunosuppressive therapy were tracked using the source documentation, and thus these data have finer granularity in time (analysis with the Kaplan-Meier approach for survival estimation). GC glucocorticoids, IS immunosuppressants

Fig. 4

Differentials in time to (a) flare (i) and (b) flare (ii) according to the baseline level of serum tenascin-C. Flare (i) was defined as the need to start or escalate glucocorticoids, and flare (ii) was defined as the need to start or escalate any immunosuppressant (analysis with the Kaplan-Meier approach for survival estimation). CI confidence interval