In non-renal SLE patients (n=6,168), this target trial emulation compared infection-related hospitalization risk over 24 months among those initiating azathioprine, belimumab, methotrexate, or mycophenolic acid analogues (MPAA) using U.S. insurance claims data (March 2011–September 2023), with both intention-to-treat and per-protocol analyses weighted for baseline and time-varying glucocorticoid exposure.
MPAA carried the highest 24-month cumulative incidence of infection-related hospitalization (13%), and was significantly riskier than belimumab (HR 1.55; 95%CI 1.07–2.25) and methotrexate (HR 1.32; 95%CI 1.04–1.68) in ITT analyses; however, after adjusting for time-varying glucocorticoid dose, all between-treatment differences were no longer statistically significant.
Time-varying glucocorticoid dose may be both a confounder and a mediator (i.e., on the causal pathway through which treatments affect infection risk), making full adjustment potentially over-controlling; the study relied on administrative claims, which may misclassify disease activity and drug exposure; patients with prior lupus nephritis or heavy immunosuppression were excluded, limiting generalizability to more severe SLE.
Glucocorticoid co-exposure—rather than the choice of immunosuppressant alone—appears to be the dominant driver of serious infection risk in non-renal SLE; clinicians should prioritize minimizing steroid burden alongside immunosuppressant selection, noting that belimumab was associated with the lowest infection-related hospitalization rate (8% at 24 months).
