Objective: To compare parotid diffusion-weighted images (DWIs) taken before and after abatacept therapy in patients with Sjögren's syndrome (SS) associated with rheumatoid arthritis (RA) and to examine the utility in evaluation and prediction of response to therapy.
Methods: DWIs of the parotid glands taken at baseline and 52 weeks after initiation of abatacept were analyzed in nine SS patients with RA using relative standard deviation (RSD) of the entire glands and signal intensity ratio (SIR) within the residual parenchyma. The correlation between changes in RSD and SIR and changes in salivary secretion based on Saxon's test was examined. Furthermore, baseline characteristics were compared in patients with increased and decreased salivary secretion after treatment. The predictive power of the parameter at baseline was examined using receiver operating characteristic (ROC) analysis.
Results: Abatacept improved salivary secretion from 2076 ± 1535 at baseline to 2857 ± 1431 mg/2 min at 52 weeks (n = 9, p = .05). Increase of salivary secretion was significantly higher in patients with decreased RSD (n = 6) than increased RSD (n = 3) (1241 ± 713, -137 ± 142 mg/2 min, p = .02). The increase and decrease in RSD completely accorded with those of salivary secretion. Furthermore, SIR was the only parameter that was significantly different between patients with posttreatment increase and decrease in salivary secretion (p = .04). ROC analysis showed the sensitivity and specificity of SIR at baseline of ≥13.0 × 10-2 for the prediction of the response to abatacept were 75.0% and 83.3%, respectively.
Conclusions: Parotid DWI seems to be useful for evaluating and predicting the response in salivary secretion to abatacept in SS patients with RA.
Keywords: Abatacept; magnetic resonance imaging; rheumatoid arthritis and Sjögren’s syndrome.