OBJECTIVES: Systemic lupus erythematosus (SLE)-related organ damage (OD) is difficult to measure as it accumulates over time. Treatment effect on long-term OD for an SLE intervention is typically not available in time for a health technology appraisal. This analysis (GSK Study 213264) aimed to calibrate an existing cost-effectiveness model with the observed treatment effect of belimumab on long-term OD.

METHODS: A recent propensity-score matched analysis (PSMA) established a 5-year change in OD with a biologic intervention versus standard SLE therapy (SST) by matching patients from a long-term extension trial of intravenous belimumab 10 mg/kg and the Toronto Lupus Cohort of patients with SLE receiving SST.¹ The Systemic Lupus International Collaborating Clinics/American College of Rheumatology Damage Index (SDI) measured OD. An existing cost-effectiveness model was run with settings identical to the PSMA; SDI progression was modelled using longitudinal statistical models from the Johns Hopkins Lupus cohort.² Compared with the PSMA results, the modelled SDI scores at 5 years overestimated and underestimated OD progression with belimumab and SST, respectively. Calibration factors were derived to adjust for the difference between model predictions and observed data.

RESULTS: The calibration factors were applied in the existing model such that the resulting SDI scores from the model matched those observed in the PSMA cohort. This lowered the incremental cost-effectiveness ratio by incorporating the additional OD benefit.

CONCLUSIONS: Outputs from the PSMA demonstrating the treatment effect of intravenous belimumab could be used to calibrate an existing cost-effectiveness model to better reflect long-term evidence. This demonstrates how cost-effectiveness models can be updated to account for newly emerging data years after an intervention has entered the market.

Study funding: GSK. Editorial support (GSK-funded): Olga Conn, PhD, Fishawack Indicia Ltd, UK.

References:

1. Urowitz M, et al. Annals of Rheumatic Dis 2019;78:372−79.
2. Watson P, et al. Rheumatology 2015;54(4):623–32.