Bortezomib treatment induces a higher mortality rate in lupus model mice with a higher disease activity

Background: Bortezomib (Bz) is a proteasome inhibitor that directly targets antibody-producing plasma cells. We recently reported the first randomized control trial that evaluated the effects of Bz in patients with systemic lupus erythematosus (SLE). In that study, we demonstrated that Bz treatment is associated with many adverse reactions in patients with refractory disease. In the present study, we examine the therapeutic and toxic effects of Bz on MRL/MpJ-lpr/lpr (MRL/lpr) mice with severe disease activity.

Methods: Female MRL/lpr mice at 10 and 14 weeks of age were treated with phosphate buffered saline (PBS) (n = 19), Bz (750 μg/kg twice weekly) (n = 27), or cyclophosphamide (Cyc) (1 mg/body, once in 2 weeks) (n = 20). Cellular subsets, serum immunoglobulin, anti-double-stranded DNA (anti-dsDNA) antibody titer, and a pathological index of glomerulonephritis were then analyzed at 22 weeks of age. Survival curves of the 10-week-old and 14-week-old Bz-treated groups were compared. Blood counts, creatinine, liver enzymes, and serum cytokine levels were measured 1 week after Bz treatment. Gene expression profiling of spleens from Bz and Cyc treatment mice were compared with those from control mice.

Results: The anti-dsDNA antibody levels were significantly higher in 14-week-old than in 10-week-old mice, indicating a higher disease activity at 14 weeks. A significant decrease in the number of splenic cells and glomerulonephritis index was observed in Bz-treated and Cyc-treated mice. Bz, but not Cyc, significantly decreased serum immunoglobulin and anti-dsDNA antibody titer levels. Survival curve analysis revealed a significantly higher mortality rate in 14-week-old than in 10-week-old Bz-treated and control groups. Following two injections of Bz, serum IL-6 and TNF-α levels were significantly more elevated in 14-week-old than in 10-week-old mice. Potentially immunogenic molecules, such as heat shock proteins, were characteristically upregulated in spleens of Bz-treated but not Cyc-treated mice.

Conclusions: In spite of its therapeutic effect, Bz treatment had more toxic effects associated with increased proinflammatory cytokine levels in mice with a higher disease activity. Understanding the mechanism of the toxicity and developing preventive strategies against it is important for the safe clinical application of Bz in human SLE.