Infection risk in lupus follows the steroid dose, not the choice of immunosuppressant

Design

Target trial emulation using US claims, marginal structural models

Population

6,168 patients with non-renal SLE starting AZA, belimumab, MTX or MPAA

Primary outcome

Time to first infection-related hospitalisation

Effect

MPAA vs belimumab HR 1.55 (1.07 to 2.25) unadjusted for steroid; not significant after adjustment

This target trial emulation, published 17 September in Arthritis & Rheumatology, used US claims data from 2011 to 2023 on 6,168 patients with non-renal lupus starting azathioprine, belimumab, methotrexate or mycophenolate, excluding those with prior nephritis or rituximab or cyclophosphamide exposure.

Two-year infection-related hospitalisation was 10% with azathioprine, 8% with belimumab, 10% with methotrexate and 13% with mycophenolate. In intention-to-treat analysis, mycophenolate carried higher risk than belimumab (HR 1.55, 95% CI 1.07 to 2.25) and methotrexate (HR 1.32). In per-protocol analysis, azathioprine and mycophenolate both had about twice the risk of belimumab. But once time-varying monthly glucocorticoid dose was included, none of the differences between treatments remained significant.

In other words, the extra infections tracked the glucocorticoid exposure that went with each drug. The practical lesson is that the most effective infection-prevention step in lupus is getting the steroid dose down, whichever steroid-sparing agent achieves that.

• Make glucocorticoid reduction the central infection-prevention goal in lupus.
• Choose the steroid-sparing agent that best controls disease so steroids can be tapered.
• Do not avoid mycophenolate solely for infection risk when it is the right drug for the disease.
• Update vaccinations and screen for latent infections before intensifying treatment.

Why it matters

Challenges the assumption that the immunosuppressant, rather than the steroid, drives serious infections in lupus.

Don't overread it

Claims-based observational data; steroid dose partly reflects disease activity, so the two cannot be fully separated.

The statistics, in plain English

Hazard ratios such as 1.55 described extra risk before accounting for steroids. When monthly steroid dose was added, the between-drug differences were no longer statistically significant.

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.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.