Determinants of neuropsychiatric flares in patients with systemic lupus erythematosus: results from five phase III trials of belimumab

Objective To identify determinants of neuropsychiatric (NP) flares in patients with SLE treated for active SLE yet no ongoing severe NPSLE with non-biologic standard therapy plus belimumab or placebo. Methods We analysed data from five phase III trials (BLISS-52, BLISS-76, BLISS-NEA, BLISS-SC, EMBRACE; n = 3638) after exclusion of patients with baseline NP BILAG A. Factors associated with NPSLE flare, defined as a new NP BILAG A or B, were investigated using Cox regression. In a subgroup analysis, we studied patients with baseline NP BILAG E for determinants of de novo NPSLE flare. Organ damage was assessed using the SLICC/ACR Damage Index (SDI). Results We documented 105 (2.9%) NPSLE flares. In multivariable analysis, male sex (HR = 2.37; 95% CI: 1.31, 4.28; P = 0.004), baseline NP BILAG B-D (HR = 5.91; 95% CI: 3.86, 9.06; P < 0.001), and increasing SDI scores (HR = 1.35; 95% CI: 1.21, 1.50; P < 0.001) were strongly associated with NPSLE flare. Belimumab use yielded no association at any dose or administration form. In analysis of SDI domains, NP damage was the strongest determinant of NPSLE flare (HR = 3.25; 95% CI: 2.72, 3.88; P < 0.001), holding true for cognitive impairment (HR = 14.29; 95% CI: 9.22, 22.14; P < 0.001), transverse myelitis (HR = 21.89; 95% CI: 5.40, 88.72; P < 0.001), and neuropathy (HR = 8.87; 95% CI: 5.59, 14.09; P < 0.001). Male sex was the strongest determinant of de novo NPSLE flare (HR = 3.26; 95% CI: 1.51, 7.04; P = 0.003). Conclusion Male sex, NPSLE history, and NP damage were strong determinants of impending NPSLE flare. No clear protection or predisposition was conferred from add-on belimumab.