魏正宗WEI, CHENG-CHUNG白洧瑄BAI, WEI-XUANWEI-XUANBAI醫學研究所碩士班2026-10-292030-07-0720252025-06-26https://handle.ncl.edu.tw/11296/bs58n2https://scholars.csmu.edu.tw/handle/123456789/36731研究目的：本研究旨在比較接受 JAK 抑制劑與腫瘤壞死因子抑制劑治療之脊椎關節炎患者，其葡萄膜炎首次發生風險之差異。研究方法：本研究採用模擬目標試驗（target trial emulation）設計，資料來源為 TriNetX 美國協作網絡之電子病歷資料庫。納入對象為 2016 年 1 月 1 日至 2023 年 12 月 31 日期間，首次接受JAK 抑制劑與腫瘤壞死因子抑制劑之 18 歲以上僵直性脊椎炎、乾癬及乾癬性關節炎患者。依初始治療類型分為JAK 抑制劑與腫瘤壞死因子抑制劑兩組，並透過 1:1 傾向分數配對（Propensity score matching）以平衡兩組之人口統計變項、共病症、共藥物與實驗室數據。主要結果為治療後首次出現葡萄膜炎診斷。追蹤期最長可達九年，並採用 Cox 比例風險回歸模型估算風險比（HR）及 95% 信賴區間（CI），另以 Kaplan–Meier 生存分析評估九年累積發生率。研究結果：研究共納入 697,850 名患者，其中 5,886 名接受JAK 抑制劑，48,029 名接受腫瘤壞死因子抑制。經傾向分數配對後， JAK 抑制劑組相較於腫瘤壞死因子抑制組之葡萄膜炎發生風險顯著較低（HR = 0.596，95% CI = 0.403–0.880）。次群分析顯示，此保護效果於年齡 ≥51 歲（HR = 0.433，95% CI = 0.237–0.791）、白人族群（HR = 0.592，95% CI = 0.376–0.931）、發炎指數升高者（CRP ≥ 3 mg/L：HR = 0.500，95% CI = 0.260–0.960；ESR ≥ 20 mm/h：HR = 0.410，95% CI = 0.190–0.870）皆一致顯著。此外，此風險降低效果無論是否合併使用傳統疾病修飾性抗風濕藥物（csDMARDs）均仍具統計顯著性（合併csDMARDs：HR = 0.502，95% CI = 0.276–0.916；未合併 csDMARDs：HR = 0.556，95% CI = 0.330–0.937）。結論與建議：本大型真實世界世代研究結果顯示，JAK 抑制劑相較於腫瘤壞死因子抑制劑，可顯著降低僵直性脊椎炎、乾癬及乾癬性關節炎患者新發生葡萄膜炎之風險。此結果支持 JAK 抑制劑於減緩此族群眼部發炎表現上的潛在應用價值，未來仍需更多前瞻性研究與隨機對照試驗，以進一步驗證其療效並指引臨床治療策略之制定。Objectives: To compare the risk of incident uveitis among patients with spondyloarthritis initiating treatment with Janus kinase inhibitors (JAKi) versus tumor necrosis factor inhibitors (TNFi). Methods: We conducted an emulated target trial using real-world electronic health records from the TriNetX US Collaborative Network. Adults with ankylosing spondylitis (AS), psoriasis (PsO), or psoriatic arthritis (PsA) who newly initiated a JAKi or a TNFi between January 1, 2016, and December 31, 2023, were identified. Patients were stratified into JAKi and TNFi cohorts based on initial treatment exposure. Propensity score matching (1:1) was performed to balance baseline demographics, comorbidities, prior medication use, and laboratory values. Cox proportional hazards models were used to estimate hazard ratios (HRs) and 95% confidence intervals (CIs) for the development of incident uveitis, with TNFi as the reference. Kaplan–Meier analysis was conducted to compare the 9-year cumulative incidence of uveitis between cohorts. The primary outcome was incident uveitis following initiation of therapy, with follow-up extending up to nine years. Results: Among 697,850 patients identified, 5,886 received JAKi and 48,029 received TNFi. After propensity score matching, JAKi use was associated with a significantly lower risk of uveitis compared with TNFi (HR = 0.596, 95% CI = 0.403–0.880). Subgroup analyses showed a consistent protective association in older patients (≥ 51 years: HR = 0.433, 95% CI = 0.237–0.791), White individuals (HR = 0.592, 95% CI = 0.376–0.931), and those with elevated inflammatory markers (CRP ≥3 mg/L: HR = 0.500, 95% CI = 0.260–0.960; ESR ≥20 mm/h: HR = 0.410, 95% CI = 0.190–0.870). The reduced risk persisted regardless of concomitant use of conventional synthetic DMARDs (with csDMARDs: HR = 0.502, 95% CI = 0.276–0.916; without csDMARDs: HR = 0.556, 95% CI = 0.330–0.937). Conclusions: In this large-scale, real-world cohort study, JAKi therapy was associated with a significantly reduced risk of incident uveitis compared to TNFi therapy in patients with AS, PsO, or PsA. These findings suggest a potential role for JAKi in mitigating ocular inflammation in this population. Further prospective studies and randomized controlled trials are warranted to validate these results and inform future clinical guidelines.脊椎關節炎自體免疫疾病葡萄膜炎腫瘤壞死因子抑制劑JAK 抑制劑目標試驗模擬世代研究spondyloarthritisautoimmune diseasesuveitistumor necrosis factor inhibitorsJanus kinase inhibitorstarget trial emulationcohort studythesis