AB0944 Baseline serum 14-3-3 ETA levels predict ACR50 response to adalimumab therapy in psoriatic arthritis

Background TNFα blockers have had a significant impact on the management of psoriatic arthritis (PsA). Despite their effectiveness, patient response is heterogeneous, underscoring the need for markers to assist in therapy selection for likely responders. The 14-3-3 family of proteins represents intracellular chaperones that regulate various biological processes. The η (eta) isoform, is detectable in the synovial fluid and serum of patients with arthritis, is up-regulated by TNFα and correlates with measures of clinical improvement in rheumatoid arthritis (RA). Objectives To study the serum levels of 14-3-3η in PsA patients in relationship to clinical response to adalimumab therapy. Methods Serum 14-3-3η levels of 24 patients (15 males and 9 females) with active PsA fulfilling the CASPAR classification criteria were measured before receiving adalimumab therapy for either 8- or 12-weeks. Clinical assessment included: SJC68, PASI, HAQ, CRP, ESR, DAS28 and ACR response. Two-tailed Mann-Whitney U-tests were run to compare differences between ACR50 responders and non-responders. Pearson correlations were generated to determine relationships between 14-3-3η and clinical variables. Contingency and regression analyses were performed to identify which variables were predictors of an ACR50 response. Receiver operating characteristic (ROC) curves were used to examine differential expression for response and to determine the accuracy of the regression model. Results Baseline 14-3-3η serum concentrations were significantly higher in ACR50 responders (0.46ng/ml) versus non-responders (0ng/ml, p=0.02) delivering a ROC AUC of 0.75, p=0.04. Consistent with published data [1] median CRP levels were also higher in ACR50 responders; 18.3mg/l versus 5.0mg/l in non-responders, p=0.05. There were no statistically significant differences in any other clinical variables between the two response groups. Across the entire cohort, 14-3-3η titres correlated moderately with CRP levels at baseline (r=0.429, p=0.03). No correlations with other clinical measures were observed. Regression analysis revealed that 14-3-3η titres were predictive of an ACR50 response delivering a significant likelihood ratio (LR) of 5.53 (p=0.018), an odds ratio (OR) of 4.62 with a ROC AUC=0.75. Similarly, a contingency analysis showed that 14-3-3η positivity reflected likelihood of ACR50 response to adalimumab therapy [LR=5.83 p=0.016; OR=8.56 and a relative risk (RR) of 3.27]. Combining 14-3-3η positivity with CRP titres strengthens the model wherein the ROC AUC for CRP as a single variable is 0.74, LR=5.86 (p=0.015) while combining the two markers delivers a ROC AUC of 0.86, LR=10.3 (p=0.006) increasing the prediction value of CRP nearly 2-fold when 14-3-3η is added. The PPV and NPV for CRP are 0.35 and 0.65 whereas with 14-3-3η they are 0.54 and 0.46, respectively. Conclusions Serum 14-3-3η titres and positivity both predict ACR50 response to adalimumab therapy in PsA patients. 14-3-3η sero-positivity when taken together with CRP titres significantly improves prediction of response. References F Van den Bosch et al. Effectiveness of adalimumab in treating patients with active psoriatic arthritis and predictors of good clinical responses for arthritis, skin and nail lesions. Ann Rheum Dis 2010;69:394-399 Disclosure of Interest A. Marotta Shareholder of: Co-founder, A. Kuijk: None Declared, W. Maksymowych: None Declared, P. Tak: None Declared