Predictors Of Response To TNF-α Antagonist Therapy In Rheumatoid Arthritis

Rheumatoid arthritis(RA) is a chronic, symmetrical, erosive autoimmune disease, which is charcteristed with multiple joints envolved and eventually ended by bone destruction, deformation and loss of function. The pathology of RA is maily synovitis, with or without vascular pannus and bone erosion. Clinically, traditional disease-modifying antirheumatic drugs(DMARDs) and biologic DMARDs are used for treatment. In our country, there are maily two kinds of biologic DMARDs in use which are tumor necrosis factor(TNF)-α antagonists and interleukin(IL)-6 antagonists. TNF-α antagonist is more commonly used, however. TNF-α antagonists are categorized into the tumor necrosis factor receptor antibody fusion protein, Inflixamab and Adalimumab. Compared with traditional DMARDs, TNF-α antagonist works much faster, having fewer side effects, but it is more expensive.So far, the exact pathogenesis of RA is still unclear. Multiple immune cells and cytokines might participate in the pathogenesis. Although TNF-α was considered to play a central role in the cytokine network, there are still some parts of the patients who treated with TNF-antagonist did not response well. According to the management of RA recommended by ACR/EULAR, to assess the response of the patients who were treated with TNF-antagonist, it is necessary to take the medcine at full dose for at least 3 months. After 3 month-treatment, the physician then evaluate the responsiveness according to one of DAS28, EULAR or ACR criteria. Thus, it would not only be a great financial loss to these non-responded patients, but also a delay for treatment, and it might bring the patients side effects of the drug. Therefore, how to predict the response of TNF-α antagonist precisely before initial administration would be of great clinical significance. Unfortunately, there are no international accepted predictors, and domestic report in this area is even more rare.Objective: To investigate predictors of response to TNF-α antagonist at different time-point after initial treatment. To find new biomarkers for evaluating the disease activity. And to find new biomarkers for assessment of efficacy of treatment.Methods: We collected the clinical characteristics and laboratory parameters of all the 90 RA patients enrolled in this study and evaluated the radiographic stage and Sharp score for each patient at baseline. Serum of patients at different time were collected as well and stored until used. Meanwhile, we collected serum from 35 healthy volunteers to be control group.(1) We divided the patiets into groups according to ACR20 criteria at two different time after treatment, that is 1 month and 3 months: Group A(responsive at 1 month), Group B(nonresponsive at 1 month), Group C(responsive at 3 month) and Group D(nonresponsive at 3 month).(2) We also divded the patients into five groups according to their radiographic stage of hands X-ray: Stage 0 group, Stage Ⅰ group, Stage Ⅱgroup, Stage Ⅲ group and Stage Ⅳgroup.We measured the serum levels of TNF-α, interleukin(IL)-6, IL-8, IL-34, MMP-3 and anti-CCP antibody in RA patients at three time-points and in control group with ELISA. Relaitons between each two of the parametres were analyzed with Spearman test. We analyzed the comparison between groups with Student t test or Mann-Whitney U test. We analyzed the comparison of parameters before and after treatment with paired t test or Wilcoxon test. ROC curve was used to determine the threshold of the predictors, their sensitivity and specificity. p<0.05 was considered to be of statistical significance.Results:(1)Comparison between Group A and B: Subjects in Group A showed lower level of ESR and higher level of ALB at baseline than that of Group B(p<0.05). Group A had lower values of morning stiffness, SJC, ESR, DAS28, SDAI, CDAI, HAQ, GLB, IL-34 and higher levels of MPV, ALB at one-month after TNF-α antagonist therapy(p<0.05). The cut off value of ALB for prediction was ≥34.9g/L(sensitivity: 54.4%; specificity: 44.2%), and that of ESR was ≤ 55.5mm/1h(sensitivity: 77.8%; specificity: 61.1%).(2)Comparison between Group C and D: Group C showed lower levels of ESR, HAQ and IL-34 at baseline(p<0.05). Group C also had lower levels of morning stiffness, SJC, TJC, ESR, CRP, DAS28, SDAI, HAQ, GLB, IL-6, IL-8, IL-34, MMP-3 and anti-CCP antibody at three-month therapy(p<0.05). The cut off value for prediction were: ESR≤60mm/1h(sensitivity: 88.2%; specificity: 60.7%); HAQ≤1.3125(sensitivity: 79.4%; specificity: 82.1%); IL-34≤194.12pg/ml(sensitivity: 76.5%; specificity: 71.4%).The serum levels of TNF-α, IL-6, IL-8, IL-34, MMP-3 and anti-CCP antibody were all significantly higher than those in control group(p<0.01). Serum IL-34 was higher in Group Ⅲ than in Group 0,GroupⅠand GroupⅡ(p<0.05). The serum levels of PLT, TNF-α, IL-6, IL-8, IL-34 and MMP-3 were positively correlated with CRP and disease activityindex, while that of ALB, MPV and PALB werenegatively correlated with CRP and disease activity, respectively. Sharp score didn’t have a relation with ESR, CRP or disease activity(p>0.05), and it wasn’t related to morning stiffness, SJC or TJC either(p>0.05).Conclusions: Both ESR and ALB were predictors for response to TNF-α antagonist at 1 month after treatment, and ESR, HAQ and IL-34 were predictors for that after 3-month-treatment. When ESR≤55mm/1h or ALB≥34.9g/L predict a good clinical response at 1month-therapy, while ESR≤60mm/1h, HAQ≤1.3125 or IL-34≤194.12pg/ml might predict a good response at 3-month. Cytokine TNF-α, IL-6, IL-8, IL-34 and MMP-3 were all envolved in RA. PLT, MPV, ALB, PALB, TNF-α, IL-6, IL-8, IL-34 and MMP-3 all could be used as markers of disease activity in RA. Except from ESR, CRP, DAS28, SDAI, CDAI and HAQ, morning stiffness, SJC, TJC, PLT, MPV, ALB, PALB, TNF-α, IL-6, IL-8, IL-34, MMP-3 and anti-CCP antibody were all parameters for monitoring the efficacy for RA treatment.