Therapeutic Drug Monitoring Plays an Important Role in Patients with Noninfectious Uveitis Receiving Adalimumab

Weidong Cheng¹Jiayu Xiao²Shengjie Li³Huaru Wang²Shasha Wang²Yujing Qian¹Chan Zhao¹Fei Gao¹XUZHEN QIN^2*Meifen Zhang^1*

• ¹Department of Ophthalmology, Peking Union Medical College Hospital (CAMS), Beijing, China
• ²Peking Union Medical College Hospital Laboratory Department, Beijing, China
• ³Peking Union Medical College Hospital State Key Laboratory of Complex Severe and Rare Diseases, Dongcheng, China

Objective: To evaluate the association between serum adalimumab (ADA) concentrations, antibodies against adalimumab (AAAs), tumor necrosis factor α (TNFα) levels, and clinical response in patients with noninfectious uveitis (NIU), as well as changes in TNFα after administration of ADA, to explore the role of therapeutic drug monitoring in these patients. Methods: This retrospective study included NIU patients treated with ADA at Peking Union Medical College Hospital between June 2024 and April 2025, who underwent testing for serum ADA and AAA levels. Most patients also had serum TNFα measured concurrently. Clinical data collected included concomitant medications, ADA dosing frequency, and ocular inflammation status, etc. High-performance size-exclusion chromatography was used to characterize forms of TNFα in serum samples of patients. THP-1 cells were stimulated with free TNFα or TNFα-ADA complexes to compare their pro-inflammatory activity. Results: Among 164 test results from 147 patients included (aged 5~56 years), median ADA level was significantly lower in AAA-positive patients than in AAA-negative patients (1.9 vs. 6.4 μg/mL; P<0.001), and lower in those with active inflammation than in those with quiescent inflammation (2.2 vs. 6.0 μg/mL; P<0.001). An ADA level below 4.1 μg/mL was associated with poor clinical response. Concomitant antimetabolite use was linked to a lower proportion of detectable AAAs compared with ADA monotherapy (34.3% vs. 54.5%; P=0.036). Median ADA level was significantly higher when testing occurred ≤14 days after the last dose of ADA (P<0.001), though many patients maintained therapeutic levels even with extended dosing intervals. TNFα levels increased in most patients after ADA therapy, predominantly in the form of TNFα-ADA complexes, which exhibited significantly weaker pro-inflammatory effects than free TNFα. Conclusion: The presence of AAAs was associated with reduced ADA levels and an increased risk of treatment failure. Despite the limitations of a retrospective design, these findings suggest that therapeutic drug monitoring may help identify causes of treatment failure and optimize regimens in stable patients.