Mortality and adverse events of special interest with intravenous belimumab for adults with active, autoantibody-positive systemic lupus erythematosus (BASE): a multicentre, double-blind, randomised, placebo-controlled, phase 4 trial

Findings Between Nov 27, 2012, and July 28, 2017, we randomly assigned 4018 patients. The as-treated population included 2002 patients in the belimumab group versus 2001 in the placebo group. Ten (0?50%) patients in the belimumab group died versus eight (0?40%) in the placebo group (difference 0?10%, 95% CI ?0?31 to 0?51). Incidences were similar in the belimumab and placebo groups for serious infections (75 [3?75%] of 2002 vs 82 [4?10%] of 2001; difference ?0?35%, 95% CI ?1?55 to 0?85), opportunistic infections and other infections of interest (36 [1?80%] vs 50 [2?50%]; ?0?70%, ?1?60 to 0?20), non-melanoma skin cancers (4 [0?20%] vs 3 [0?15%]; 0?05%, ?0?21 to 0?31) and other malignancies (5 [0?25%] vs 5 [0?25%]; 0?00%, ?0?31 to 0?31). A higher proportion of patients in the belimumab group than in the placebo group had infusion and hypersensitivity reactions (8 [0?40%] vs 2 [0?10%]; 0?30%, ?0?01 to 0?61), serious depression (7 [0?35%] vs 1 [0?05%]; 0?30%, 0?02 to 0?58), treatmentemergent suicidality (28 [1?42%] of 1972 patients vs 23 [1?16%] of 1986; 0?26%, ?0?44 to 0?96), and sponsoradjudicated serious suicide or self-injury (15 [0?75%] of 1972 patients vs 5 [0?25%] of 1986; post hoc difference 0?50%, Background Belimumab is approved for the treatment of active systemic lupus erythematosus (SLE). Although clinical trials showed a favourable benefit-risk profile, numerical differences in the incidence of mortality and adverse events of special interest (AESIs) have been reported. We assessed the frequency of these events in patients with SLE receiving belimumab or placebo plus standard therapy. Methods BASE was a double-blind, randomised, placebo-controlled, phase 4 trial done in 33 countries. Adults with active SLE were randomly assigned (1:1) to receive intravenous belimumab (10 mg/kg) or placebo, plus standard therapy, for 48 weeks. The primary endpoints were incidences of all-cause mortality and AESIs during the on-treatment period (first-to-last study drug dose + 28 days). Safety analyses were done in the as-treated population (patients grouped by actual treatment received >50% of the time). This study was registered with ClinicalTrials.gov (NCT01705977). Findings Between Nov 27, 2012, and July 28, 2017, we randomly assigned 4018 patients. The as-treated population included 2002 patients in the belimumab group versus 2001 in the placebo group. Ten (0 & middot;50%) patients in the belimumab group died versus eight (0 & middot;40%) in the placebo group (difference 0 & middot;10%, 95% CI & minus;0 & middot;31 to 0 & middot;51). Incidences were similar in the belimumab and placebo groups for serious infections (75 [3 & middot;75%] of 2002 vs 82 [4 & middot;10%] of 2001; difference & minus;0 & middot;35%, 95% CI & minus;1 & middot;55 to 0 & middot;85), opportunistic infections and other infections of interest (36 [1 & middot;80%] vs 50 [2 & middot;50%]; & minus;0 & middot;70%, & minus;1 & middot;60 to 0 & middot;20), non-melanoma skin cancers (4 [0 & middot;20%] vs 3 [0 & middot;15%]; 0 & middot;05%, & minus;0 & middot;21 to 0 & middot;31) and other malignancies (5 [0 & middot;25%] vs 5 [0 & middot;25%]; 0 & middot;00%, & minus;0 & middot;31 to 0 & middot;31). A higher proportion of patients in the belimumab group than in the placebo group had infusion and hypersensitivity reactions (8 [0 & middot;40%] vs 2 [0 & middot;10%]; 0 & middot;30%, & minus;0 & middot;01 to 0 & middot;61), serious depression (7 [0 & middot;35%] vs 1 [0 & middot;05%]; 0 & middot;30%, 0 & middot;02 to 0 & middot;58), treatment-emergent suicidality (28 [1 & middot;42%] of 1972 patients vs 23 [1 & middot;16%] of 1986; 0 & middot;26%, & minus;0 & middot;44 to 0 & middot;96), and sponsor-adjudicated serious suicide or self-injury (15 [0 & middot;75%] of 1972 patients vs 5 [0 & middot;25%] of 1986; post hoc difference 0 & middot;50%, 0 & middot;06 to 0 & middot;94). Interpretation In line with previously published data, incidences of all-cause mortality and AESIs were similar in patients given belimumab and placebo, except for serious infusion or hypersensitivity reactions, serious depression, treatment-emergent suicidality, and sponsor-adjudicated serious suicide or self-injury events.