The Role of Therapeutic Apheresis in the Management of Rheumatoid Arthritis A Systematic Review and Meta‐Analysis

ABSTRACT

Rheumatoid arthritis (RA) cannot be cured in many patients even with improved drug therapies. Therapeutic apheresis has been explored as an alternative intervention for refractory RA, yet relevant clinical evidence is scattered and mostly derived from early studies. We systematically retrieved eligible studies from PubMed, Web of Science and the Cochrane Library up to December 12, 2024, enrolling randomized controlled trials (RCTs) and case series containing at least five patients that assessed all types of apheresis for RA treatment. A random‐effects model was adopted to carry out meta‐analysis; the Cochrane Risk of Bias 2 (ROB2) tool and Joanna Briggs Institute (JBI) checklist were used to evaluate methodological quality, while the Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework was applied to judge the certainty of evidence. Sixteen studies (10 RCTs, 6 case series) were included. Meta‐analysis of 8 RCTs revealed no statistically significant benefit of apheresis as a standalone therapy. For Health Assessment Questionnaire (HAQ), the pooled standardized mean difference (SMD) was −1.02 (95% confidence interval [CI]: −2.15 to 0.11). For erythrocyte sedimentation rate (ESR), the pooled SMD was −0.90 (95% CI: −2.41 to 0.62). For C‐reactive protein (CRP), the pooled SMD was −0.75 (95% CI: −1.73 to 0.24). Substantial heterogeneity was observed across all outcomes ( I ²  > 90%). However, subgroup analyses by modality revealed critical distinctions. Double‐filtration plasmapheresis (DFPP) demonstrated the most consistent benefits when used as an adjunct to disease‐modifying antirheumatic drugs (DMARDs). For ESR, the pooled mean difference (MD) was −22.32 (95% CI: −61.74 to 17.11). For CRP, the pooled MD was −3.04 (95% CI: −7.75 to 1.66). For HAQ, the pooled MD was −0.96 (95% CI: −2.14 to 0.21). In contrast, Therapeutic Plasma Exchange (TPE) and Lymphoplasmapheresis (LPE) showed limited efficacy and higher adverse event rates. The certainty of evidence was very low to low for most outcomes. While therapeutic apheresis as a broad intervention class lacks consistent efficacy in RA, this overall finding masks critical distinctions between technologies. DFPP shows promise as an adjunct to pharmacotherapy in carefully selected patients with highly refractory disease, offering a favorable benefit–risk profile. The clinical utility of apheresis in RA is inextricably linked to technical selectivity, highlighting the need for rigorous sham‐controlled trials with standardized protocols.