AI-generated research brief — verify at source
All Four Oral JAK Inhibitors Raise LDL Cholesterol Significantly Versus Placebo
A systematic review and meta-analysis of 13 placebo-controlled trials found that all four oral JAK inhibitors examined—baricitinib, upadacitinib, tofacitinib, and decernotinib—produced consistent, statistically significant elevations in LDL cholesterol compared to placebo, with increases ranging from approximately 9 mg/dL for baricitinib to nearly 16 mg/dL for tofacitinib. Although HDL cholesterol also rose across all agents, the concurrent LDL elevation represents a class-wide lipid signal with potential cardiovascular implications that warrants systematic clinical attention.
What Was Studied
This analysis investigated whether oral Janus kinase inhibitors (JAKi) are associated with clinically meaningful changes in lipid profiles—specifically HDL, LDL, total cholesterol, and triglycerides—compared to placebo, across patients with immune-mediated diseases. The question is clinically important because JAKi are increasingly used across multiple inflammatory conditions, and characterizing their lipid effects is critical for evaluating long-term cardiovascular safety in treated populations.
How It Was Studied
Researchers conducted a systematic review and meta-analysis drawing on phase 2 and phase 3 placebo-controlled randomized clinical trials identified through searches of Embase, PubMed, and Web of Science up to March 2025. Eligible trials enrolled patients with immune-mediated diseases including rheumatoid arthritis (nine studies), atopic dermatitis (two), Crohn’s disease (one), and psoriasis (one). A total of 13 studies were included, encompassing 3,978 patients receiving oral JAKi and 1,680 controls receiving placebo, for a combined population of 5,658 participants. Lipid changes were expressed as mean differences in mg/dL between active treatment and placebo arms.
What Was Observed
- Tofacitinib produced the largest LDL elevation among the agents studied: an average increase of approximately 15.7 mg/dL more than placebo (95% CI 12.9–18.6), alongside an HDL increase of about 7.0 mg/dL (95% CI 5.7–8.3). The narrow confidence intervals indicate a precise and consistent estimate across contributing trials.
- Upadacitinib and decernotinib also showed substantial LDL increases: upadacitinib raised LDL by approximately 12.4 mg/dL above placebo (95% CI 8.9–15.9) with an HDL rise of 5.4 mg/dL (95% CI 3.2–7.7), while decernotinib was associated with an LDL increase of about 14.9 mg/dL (95% CI 3.6–26.3). The wide confidence interval for decernotinib reflects limited data from few trials and considerable uncertainty around its true lipid effect.
- Baricitinib showed the smallest LDL increase in this analysis, at approximately 9.05 mg/dL above placebo (95% CI 7.78–10.32), with an accompanying HDL rise of about 6.07 mg/dL (95% CI 5.01–7.14). Both estimates were precise, suggesting a reliable signal from the rheumatoid arthritis trial data underpinning this drug’s estimates.
Why This Matters
This meta-analysis provides the clearest quantitative synthesis to date that dyslipidemia—particularly LDL elevation—is a consistent pharmacological effect shared across structurally distinct oral JAKi agents and multiple disease indications, rather than an idiosyncratic finding. The class-wide nature of the LDL signal strengthens the case that lipid monitoring should be a routine component of JAKi management, irrespective of the underlying condition being treated. The authors specifically note that these lipid changes could translate into cardiovascular risk, underscoring the need to integrate cardiovascular risk factor assessment into the standard care of patients on these therapies.
How to Read This Result
These findings are drawn from short-duration phase 2 and 3 RCTs pooled across heterogeneous disease conditions, and while the meta-analytic signal is consistent, the medium study quality and the absence of long-term cardiovascular outcome data mean that the clinical magnitude of cardiovascular risk from these lipid changes remains uncertain.
Limitations
The included trials varied in design, patient populations, disease indications, and treatment durations, introducing heterogeneity that may affect the precision and generalizability of pooled estimates. Only a small number of trials contributed data for certain agents—most notably decernotinib, whose very wide confidence interval reflects this sparsity. Because the included studies were relatively short-term RCTs, the analysis captures early lipid changes but cannot assess whether these elevations translate into actual cardiovascular events over longer follow-up. Variation in background therapies and disease activity across indications may also have confounded lipid measurements.