Abstract
Objective. To assess the relationship between serum concentrations of tumor necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL) and osteoprotegerin (OPG) and the therapeutic response to disease-modifying antirheumatic drugs (DMARD) in patients with early rheumatoid arthritis (RA).
Methods. Circulating levels of TRAIL and its soluble receptor OPG were measured by ELISA in paired serum samples obtained from 66 patients with early RA at their first visit (baseline) and after 1 year of therapy. Levels of TRAIL and OPG were analyzed in relation to the clinical response, defined by the 28-joint count Disease Activity Score (DAS28).
Results. Both serum TRAIL and OPG increased after DMARD therapy. Baseline levels of TRAIL, but not OPG, were significantly higher (p < 0.05) in the patients that achieved a clinical response by DAS28 after 1 year of therapy, versus patients without clinical response to DMARD. Baseline serum levels of TRAIL were higher (p < 0.01) in rheumatoid factor-negative patients.
Conclusion. Our data suggest that the basal level of circulating TRAIL is an important determinant in the therapeutic response to DMARD in patients with early RA.
- EARLY RHEUMATOID ARTHRITIS
- DISEASE MODIFYING ANTIRHEUMATIC DRUGS
- TUMOR NECROSIS FACTOR RELATED APOPTOSIS INDUCING LIGAND
Early diagnosis of rheumatoid arthritis (RA) is crucial to prevent unfavorable disease outcome1. RA treatment includes early use of disease-modifying antirheumatic drugs (DMARD) such as methotrexate (MTX), leflunomide, sulfasalazine (SSZ), and hydroxychloroquine, and, for the most aggressive disease subsets, biologic agents initiated as soon as the diagnosis is confirmed. To achieve remission, optimal management of RA is needed within 3–6 months after onset of disease2.
Tumor necrosis factor (TNF)-related apoptosis-inducing ligand (TRAIL), also known as Apo-2 ligand, is a member of the structurally related TNF family of cytokines, and exists as either a type II membrane protein or as a soluble protein3. Four transmembrane TRAIL receptors belonging to the apoptosis-inducing TNF-receptor (R) family have been described. While TRAIL-R1 (DR4) and TRAIL-R2 (DR5) transduce apoptotic signals upon binding of TRAIL, TRAIL-R3 (DcR1) and TRAIL-R4 (DcR2) are homologous to DR4 and DR5 in their cysteine-rich extracellular domain, but lack intracellular death domain and apoptosis-inducing capability3. In addition, it has recently been shown that the soluble receptor osteoprotegerin (OPG) interacts with and neutralizes the biological activity of TRAIL with affinity comparable to that of RANKL, another member of the TNF family of cytokines4. Various in vitro and in vivo studies in animal models have shown that TRAIL might have a therapeutic role in ameliorating experimental osteoarthritis, as well as in promoting apoptosis of synovial fibroblasts obtained from patients with RA5–10. To date, limited data are available on the levels of circulating TRAIL in RA patients. Moreover, although it has been shown that TRAIL is able to induce apoptosis of RA-derived synoviocytes7–10, it is noteworthy that TRAIL can also promote nonapoptotic signals in a cell type-dependent manner11–18. In particular, it has been demonstrated that recombinant TRAIL can also induce in vitro proliferation of RA-derived synoviocytes surviving the initial induction of apoptosis11,14; another study proposed that high levels of OPG might inhibit the ability of TRAIL to induce apoptosis of RA-derived synoviocytes10.
The aim of our study was to measure serum concentrations of TRAIL and OPG in patients with early RA and to assess the relationship between levels of TRAIL and OPG and the therapeutic response to DMARD, defined on the basis of the 28-joint count Disease Activity Score (DAS28).
MATERIALS AND METHODS
Patients
All patients presenting with early arthritis between January 2006 and January 2008 were included in the study cohort. “Early arthritis” was defined by the presence of 2 of the following 4 clinical criteria: no antecedent trauma, at least one tender joint, at least one swollen joint, and morning stiffness lasting ≥ 60 minutes. In addition, inclusion criteria included serological signs of inflammation [elevated erythrocyte sedimentation rate (ESR) or C-reactive protein (CRP)] or rheumatoid factor (RF) positivity and anticitrullinated protein antibodies (anti-CCP) and a symptom duration not longer than 24 weeks. Clinical assessment of patients included evaluation by DAS28 and the Health Assessment Questionnaires (HAQ). Venous serum was obtained from each patient. All patients were treated with DMARD therapy, consisting of a single DMARD (with or without steroids) or combination DMARD (Table 1). Paired serum samples were available for each patient: one sample obtained at baseline (before starting DMARD therapy) and a second sample after about 1 year of therapy. In these patients, at 1 year followup, a decrease in DAS28 > 1.2 with respect to baseline was defined as clinical response to treatment, while increase or decrease in DAS28 < 1.2 was defined as no response, as reported19. The laboratory and clinical information for each patient was obtained from clinical records. As a control group, sera were collected from 51 age- and sex-matched healthy donors (age 59.3 ± 13.8 yrs; 70.6% were female).
Characteristics of the study population.
Written consent for all procedures was obtained from all subjects in accord with the Declaration of Helsinki and with approval obtained from the institutional review boards of the university hospitals of Ferrara and Pavia.
Measurement of TRAIL and OPG
Measurement of TRAIL and OPG was carried out in the serum samples of patients with early RA and healthy controls. Analyses were performed using specific commercial ELISA kits purchased from R&D Systems (Minneapolis, MN, USA) and Alexis Biochemicals (Lausen, Switzerland) for TRAIL and OPG, respectively. Assays were carried out in duplicate in accord with the manufacturer’s instructions and analyzed with an ELISA reader at 450 nm. Sensitivity of the TRAIL assay was 2.9 pg/ml and the intra- and interassay coefficients of variation (CV) were 3.9% and 6%, respectively, while the sensitivity of the OPG assay was 2.8 pg/ml and intra- and interassay CV were 9% and < 10%.
Statistical analysis
Data were calculated and shown as mean ± SD or as median and interquartile range (IQR), according to the distribution. Comparisons between patient groups and controls were performed with Student’s t test and chi-square test. Differences in TRAIL, OPG, and disease variables across phases of the study were analyzed by ANOVA for repeated measures. Statistical significance was defined as p < 0.05.
RESULTS
Levels of TRAIL and OPG increased after therapy
Demographic and clinical characteristics of 66 early RA patients analyzed for serum TRAIL and OPG levels are given in Table 1. At baseline (before starting DMARD therapy), the RA patient group was characterized by 66% positivity of CRP, mean DAS28 score 5.07 ± 1.68, mean HAQ score 1.31 ± 0.8, total swollen joint count 10.2 ± 6.5, and tender joint count 10.2 ± 8.4.
Levels of serum TRAIL in RA patients at baseline (mean 65.3 ± 24.4 pg/ml) were comparable to those in healthy controls (mean 68.6 ± 20.6 pg/ml); after 1 year of DMARD therapy, followup analysis of the same RA patients showed a significant increase (p < 0.01) in serum levels of TRAIL compared to baseline levels in patients and controls (Figure 1A). Also, levels of serum OPG in RA patients at baseline (mean 79.4 ± 44 pg/ml) were comparable with those in controls (mean 74.8 ± 26.6 pg/ml), and 1 year after DMARD therapy serum levels of OPG were significantly increased compared to those in controls (p < 0.05), but not compared to baseline levels of OPG in RA patients (p = 0.14; Figure 1B).
Comparison of levels of circulating TRAIL (A) and OPG (B) in controls and patients with early RA, determined by ELISA. Patient samples were collected at baseline, i.e., before starting therapy, and again after 1 year of DMARD therapy. Horizontal bars are median, upper and lower edges of box are 75th and 25th percentiles, extensions are 10th and 90th percentiles.
Baseline serum TRAIL but not OPG levels predict clinical response to therapy
After 1 year, significant decreases in the DAS28 score and HAQ score and in the percentage of CPR-positive patients were observed in the early RA population (Figure 2). Interestingly, when the RA population was subdivided into responders and nonresponders to DMARD therapy on the basis of differences in DAS28 score (delta DAS28, Figure 3A), we observed that the patients showing a good response to DMARD therapy had significantly higher levels (p < 0.05) of baseline serum TRAIL compared to patients whose delta DAS28 showed minimal (< 1.2) or absent variation after therapy (Figure 3B). In contrast, no significant differences were observed in baseline serum OPG levels between these 2 groups of patients (Figure 3B).
Clinical response of patients with early RA after 1 year of DMARD therapy. DAS28 and HAQ scores and positivity for CRP were monitored at baseline (before starting therapy) and after 1 year of DMARD therapy. For DAS28 and HAQ scores, horizontal bars are median, upper and lower edges of box are 75th and 25th percentiles, extensions are 10th and 90th percentiles. Percentages of patients positive for CRP at the 2 time-points are shown.
Circulating TRAIL and OPG levels in patients with early RA in relationship to their clinical response to DMARD therapy. A. Patients were divided into DMARD responders and non-responders on the basis of delta DAS28. B. Serum levels of TRAIL and OPG in patients at baseline compared to clinical response to DMARD. Horizontal bars are median, upper and lower edges of box are 75th and 25th percentiles, extensions are 10th and 90th percentiles.
Since these findings suggest that high baseline serum levels of TRAIL, but not OPG, might have predictive value in identifying those patients that would show a favorable response to DMARD, we examined further the baseline levels of serum TRAIL and OPG in relation to negative prognostic factors, such as presence of erosive disease (Figure 4A) and RF positivity (Figure 4B). Interestingly, while serum levels of OPG were significantly higher (p < 0.01) in patients with erosive disease, serum levels of TRAIL were not significantly different in patients with and those without erosive bone disease (Figure 4A). On the other hand, serum levels of TRAIL were significantly lower (p < 0.01) in patients positive for RF (Figure 4B), in keeping with a role of TRAIL in counteracting autoimmune phenomena3.
Circulating TRAIL and OPG levels in patients with early RA at baseline in relationship to negative prognostic markers: A. presence or absence of erosive disease; B. rheumatoid factor status. Horizontal bars are median, upper and lower edges of box are 75th and 25th percentiles, extensions are 10th and 90th percentiles.
DISCUSSION
A major finding of our study is that patients with higher levels of serum TRAIL, irrespective of their disease activity score at baseline, obtained the optimal response from DMARD therapy. On the other hand, pretreatment levels of serum OPG did not predict clinical outcome in our study population. In this respect, although data concerning the levels of OPG in sera of patients with early RA are heterogeneous20–25, a possible pathogenic link between elevated levels of OPG and inflammation has been suggested by in vitro studies26–28. However, as in disorders of elevated bone turnover29, the upregulation of OPG could also represent a compensatory mechanism to limit bone erosion. We observed consistently higher basal serum OPG levels in patients with erosive disease.
Moreover, we found that serum levels of OPG as well as TRAIL increased in response to DMARD therapy. These findings are in agreement with reports from in vitro studies that therapy with MTX and SSZ each augmented the secretion of OPG30. In addition, serum OPG and TRAIL were both higher after 1 year of DMARD therapy in patients with a good clinical response. In this respect, it is noteworthy that although OPG is able to potentially interact with TRAIL, and both OPG and TRAIL display antiosteoclastic activity31,32, we have recently shown that when added in vitro at a 1:1 (TRAIL:OPG) ratio, the antiosteoclastic activity of OPG is not impaired by the presence of TRAIL33. Further, it should be considered that while we measured concentrations of OPG, which are particularly elevated in the synovial fluid of RA patients (mean 667.3 ± 424.5 pg/ml), in keeping with a potential antierosive role of synovial OPG, we were unable to detect soluble TRAIL in the same synovial fluids (with assay sensitivity of 2.9 pg/ml), strongly suggesting that the potential biological activity of soluble TRAIL is not expressed at the synovial level.
The major finding of our study was that higher concentrations of serum TRAIL at baseline were observed in patients with early RA with a favorable response to DMARD therapy, and that TRAIL might induce its immunomodulatory activity at the systemic level.
Footnotes
-
Supported by grants from the Fondo Regione Emilia Romagna (PG 08 23943/2008/P18A4) and CariFe Foundation.
- Accepted for publication February 16, 2010.
