JAK inhibitors worked less well in rheumatoid arthritis as BMI rose

Design

Individual patient data meta-analysis of 16 phase 3 randomised trials

Population

11,883 adults with rheumatoid arthritis in tofacitinib, baricitinib and upadacitinib trials

Primary outcome

ACR20 response and DAS28-CRP at the primary time point

Effect

ACR20 RR vs healthy weight: 0.94 overweight to 0.78 (95% CI 0.71 to 0.85) class 3 obesity

An individual patient data meta-analysis in The Lancet Rheumatology (3 August 2026) pooled 16 phase 3 trials of tofacitinib, baricitinib and upadacitinib in 11,883 people with rheumatoid arthritis, obtained through the Vivli platform. Nearly a third had obesity.

Compared with a healthy weight, the relative likelihood of an ACR20 response was 0.94 with overweight, 0.92 with class 1 obesity, 0.88 with class 2 and 0.78 with class 3 (BMI 40 or more). DAS28-CRP was correspondingly higher, by 0.54 points in class 3 obesity. There was no such gradient on placebo, so the difference reflects a smaller drug effect rather than worse disease alone. Heterogeneity was low.

Obesity is common in rheumatoid arthritis and rising in India. These data do not say JAK inhibitors should be avoided in heavier patients. They say weight is part of the treatment plan, and that a patient with severe obesity who responds less well may not have failed the drug for any other reason.

For prescribers, it supports discussing weight management alongside any JAK inhibitor and setting expectations accordingly.

• Record BMI before starting a JAK inhibitor and review it at follow-up.
• Discuss weight management as part of rheumatoid arthritis treatment, not as separate advice.
• Set realistic expectations of response in patients with severe obesity.
• Consider BMI when judging an incomplete response before switching drugs.
• Do not withhold a JAK inhibitor on BMI alone; a smaller expected response is still a response.

Why it matters

It turns obesity from background comorbidity into a modifier of how well the drug works.

Don't overread it

This is an analysis of trial data by baseline BMI; it does not show that losing weight restores response.

The statistics, in plain English

A relative risk of 0.78 means patients with class 3 obesity were about 22% less likely to reach ACR20 than those of healthy weight. Because placebo responses did not show the same gradient, the effect is on the drug's benefit rather than on disease measurement alone.

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.