Low numbers of blood and salivary natural killer cells are associated with a better response to belimumab in primary Sjogren's syndrome: results of the BELISS study.

INTRODUCTION: In this study, we sought to address changes in blood lymphocyte subpopulations and labial salivary gland (LSG) inflammation after belimumab treatment in patients with primary Sjogren's syndrome (pSS) and to identify predictors of response to treatment. METHODS: Sequential blood lymphocyte subsets and LSG biopsies were analysed between week 0 (W0) and W28 in 15 patients with pSS treated with belimumab. Systemic response to treatment was defined as a decrease in the European League Against Rheumatism Sjogren's Syndrome Disease Activity Index score of >/=3 points at W28. RESULTS: After belimumab, we observed a decrease in blood B lymphocytes primarily involving CD27-negative/immunoglobulin D-positive naive B cells (p=0.008). Lymphocytic sialadenitis (focus score >1) that was present in 12 patients (80.0 %) before belimumab treatment became negative in 5 of them after treatment (p=0.03). The median (interquartile range) LSG B-cell/T-cell ratio decreased from 0.58 (0.5-0.67) to 0.50 (0.5-0.5) (p=0.06). B-cell activating factor (BAFF) staining was detected in 11 (78.6 %) of 14 patients before belimumab treatment compared with 7 (50.0 %) of 14 after belimumab therapy (p=0.10). The median percentage of BAFF-positive cells in foci significantly decreased from 27.5 % (10-40) to 5 % (0-20) (p=0.03). A systemic response was achieved in six patients (40 %). The only predictor of response was the presence of a low number of natural killer (NK) cells, both in blood (8.5 % [7-10] vs 11 % [9-21]; p=0.04) and in LSG (20.6/mm(3) [20.0-21.4] vs 30.0/mm(3) [25.0-100.0], p=0.003). Serum BAFF levels did not influence response to treatment. CONCLUSIONS: Low blood and salivary NK cell numbers are associated with a better response to belimumab. This suggests that two distinct subsets of pSS may exist: one with a predominant type I interferon (IFN)-BAFF-B-cell axis, representing good responders to belimumab; and one with a predominant type II IFN-NK cell axis, representing non-responders. TRIAL REGISTRATION: ClinicalTrials.gov identifier: NCT01160666 . Registered 9 July 2010.