Outcome Measure Discordance in Systemic Lupus Erythematosus Clinical Trials

Researchers evaluated the degree of concordance between anifrolumab trials to better understand the drivers of disparate results.

Discordance in outcome measures between the British Isles Lupus Assessment Group (BILAG)-based Composite Lupus Assessment (BICLA) and the systemic lupus erythematosus (SLE) Responder Index (SRI) in 1 of 3 anifrolumab clinical trials highlights the need to monitor baseline disease activity and glucocorticoid tapering in future SLE trials, according to an article published in Annals of the Rheumatic Diseases.

Three trials recently evaluated anifrolumab for the treatment of SLE, including the phase 2b MUSE trial (ClinicalTrials.gov Identifier: NCT01438489), the phase 3 TULIP-1 (ClinicalTrials.gov Identifier: NCT02446912 ) and TULIP-2 (ClinicalTrials.gov Identifier: NCT02446899) trials. All 3 trials used the BICLA and SRI(4) to assess clinical responses.

In MUSE and TULIP-2, anifrolumab demonstrated efficacy as measured by the BICLA and SRI(4) endpoints. In TULIP-1, the endpoint for BICLA, but not SRI(4), was met.  This study was undertaken to more closely evaluate the discordance between the BICLA and SRI(4) outcomes in TULIP-1.

Researchers analyzed data from patients who received either 300 mg anifrolumab or placebo every 4 weeks for 48 weeks as part of the randomized, double-blind TULIP-1, TULIP-2, and MUSE clinical trials. In TULIP-1 and TULIP-2, patients receiving at least 10 mg/day of oral glucocorticoids were required to attempt to taper to  7.5 mg/day or fewer between weeks 8 and 40. Tapering was encouraged, but not mandatory in MUSE. 

BICLA and SRI(4) responses were measured at week 52. Patients were grouped by concordance of outcomes. Concordant subgroups had dual positive or negative BICLA and SRI(4) responses. Discordant subgroups had differing responses between BICLA and SRI(4). The subgroups were evaluated for baseline demographics and clinical characteristics, glucocorticoid taper, and changes in organ domain scores and joint counts between baseline and 52 weeks.

Across the 3 trials, 85.4% (TULIP-1), 83.7% (TULIP-2), and 78.0% (MUSE) of patients had concordant BICLA and SRI(4) outcomes, and the proportion of dual BICLA/SRI(4) responders favored the anifrolumab group in each trial. A smaller proportion of patients who were BICLA nonresponders/SRI(4) responders were identified in each trial. The proportion of patients in this subgroup between the treatment and placebo arms was similar in the TULIP-2 and MUSE studies. In the TULIP-1 study, the proportion of patients in the BICLA nonresponder/SRI(4) responder subgroup was higher in the placebo (15.2%) vs the anifrolumab (6.7%) arms.

Within the TULIP-1 BICLA nonresponder/SRI(4) responder subgroup, the resolution of arthritis was the main reason for the SRI(4) response among patients who received placebo. These patients had lower baseline disease activity scores, joint counts, and glucocorticoid tapering rates than those who received anifrolumab. The combination of less glucocorticoid tapering and fewer active joints may have magnified the SRI(4) response. The BICLA response, which required at least a partial improvement in all domains that were scored as moderate or severe at entry, was less sensitive to these imbalances.

Limitations of the study included the post-hoc analysis, small numbers of patients in the discordant subgroups, complex inclusion criteria for the clinical trials, and circumstantial connections to explain discordance in the TULIP-1 trial.

The authors concluded, “Given the emphasis placed on primary endpoint attainment in phase 3 trials by regulators, factors that [jeopardize] study outcomes need to be [recognized] and mitigated during trial design and execution. Confirmation of our observations in other trial cohorts may also suggest ways in which we can improve on current composite endpoints in a data-driven fashion. For now, we suggest that careful attention to baseline factors and maintaining uniformity in glucocorticoid tapering are essential in future SLE clinical trials to reduce the likelihood of discordant results and [maximize] the ability to detect efficacy signals.”

Disclosure: Some study authors declared affiliations with biotech, pharmaceutical, and/or device companies. Please see the original reference for a full list of authors’ disclosures.


Bruce IN, Furie RA, Morand EF, et al. Concordance and discordance in SLE clinical trial outcome measures: analysis of three anifrolumab phase 2/3 trials. Ann Rhuem Dis. Published online June 10, 2022. doi:10.1136/annrheumdis-2021-221847