Cost-effectiveness of janus kinase inhibitors for rheumatoid arthritis: A systematic review and meta-analysis of cost-utility studies.

Janus kinase inhibitors (JAK-i), a class of targeted synthetic disease-modifying antirheumatic drugs (tDMARDs), are suggested as second or third-line therapies in rheumatoid arthritis (RA). Synthesized cost-effective evidence would aid in informed decision-making given the similar clinical effectiveness of JAKi, but incongruent cost-effectiveness reports. Literature search was conducted in PubMed, Embase, Scopus, and Tufts Medical Centers' cost-effective analysis registry. We pooled the incremental net benefit (INB) with 95% confidence interval (CI) using random-effects model and the heterogeneity was assessed using Cochrane-Q test and I2 statistic. Modified economic evaluation bias checklist was used to assess the quality of selected studies. Publication bias was assessed using a funnel plot and Egger's test. The Grading of Recommendation, Assessment, Development, and Evaluation (GRADE) assessment was performed to assess the certainty of outcomes presented. We included seventeen relevant studies for systematic review, of which fifteen were eligible for meta-analysis. The meta-analysis results showed that JAK-i is cost-effective compared to csDMARDS/bDMARDs with a pooled INB (INBp) of $19,886 (95% CI, 1,635 to 38,137) but with considerable heterogeneity (I2 = 99.14). As a second-line treatment for csDMARD failed RA, JAK-i is cost-effective than csDMARD/bDMARD with a pooled INB of $23,144 (74.1-46,214) and high heterogeneity (I2 = 99.67). But on a separate analysis JAK-i as second-line treatment is not cost-effective than TNF-a-i (INBp = $25,813, -5,714 to 57,340). However, leave-one-out analysis found that omitting a single outlier makes JAK-i cost-effective. Further, JAK-i is not cost-effective as a third-line treatment for csDMARD-TNF-a-I failed RA, compared to csDMARDs/bDMARDs with INBp $26,157 (-7,284 to 59,598). Meta-analysis suggests that JAK-i is cost-effective when used after csDMARD failure but not cost-effective when used after csDMARD-TNF-a-i failure with low certainty of evidence. https://www.crd.york.ac.uk/prospero/display_record.php?ID=CRD42021222541, identifier CRD42021222541.