- Design
- Target trial emulation using claims data
- Population
- 6,168 patients with nonrenal lupus starting azathioprine, belimumab, methotrexate or a mycophenolic acid analogue
- Primary outcome
- Time to first infection-related hospitalisation
- Effect
- 24-month incidence 8% to 13%; mycophenolate vs belimumab ITT HR 1.55 (95% CI 1.07 to 2.25); differences lost after adjusting for steroid dose
This target trial emulation used US claims data for 6,168 patients with nonrenal systemic lupus starting azathioprine, belimumab, methotrexate or a mycophenolic acid analogue between 2011 and 2023. The outcome was hospitalisation for infection. People with prior lupus nephritis, transplantation, or rituximab or cyclophosphamide exposure were excluded.
At 24 months, cumulative incidence was 10% for azathioprine, 8% for belimumab, 10% for methotrexate and 13% for the mycophenolate group. In intention-to-treat analysis, the mycophenolate group had higher risk than belimumab (HR 1.55, 95% CI 1.07 to 2.25) and methotrexate (HR 1.32, 1.04 to 1.68). Per protocol, azathioprine (HR 2.04, 1.01 to 4.16) and mycophenolate (HR 2.27, 1.11 to 4.62) were higher than belimumab.
Once monthly glucocorticoid dose was modelled over time, the differences between drugs were no longer significant. That fits the view that steroid exposure drives much of the infection risk, though steroid use also reflects disease activity, so the picture is not simple. Claims data lack disease activity measures.
- Count the glucocorticoid dose when comparing infection risk between lupus drugs.
- Prioritise steroid-sparing strategies, which may matter as much as the choice of immunosuppressant.
- Review vaccination status and infection prophylaxis before and during treatment.
- Do not choose between agents on infection risk alone from this study.
Why it matters
It reframes serious infection in lupus as largely a steroid exposure problem rather than a feature of one immunosuppressant.
Don't overread it
A claims-based emulation cannot separate steroid effects from disease activity, so it does not prove the agents are equally safe.
The statistics, in plain English
A hazard ratio of 1.55 with an interval of 1.07 to 2.25 shows a raised risk that disappeared when steroid dose was added to the model. That means the drug and the dose are tangled together, not that drug choice never matters.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for rheumatology, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free