JAK抑制剂治疗成人类风湿关节炎的疗效与安全性:系统综述与累积荟萃分析
该系统综述与累积荟萃分析纳入31项随机对照试验的46篇报告,评估JAK抑制剂对比安慰剂、csDMARDs或bDMARDs治疗成人类风湿关节炎的疗效与安全性。JAK抑制剂在改善DAS28、ACR应答和HAQ-DI方面均优于对照,但不良事件、严重感染、SAEs和MACE风险因对照不同而存在差异。
要点
- JAK抑制剂改善DAS28优于安慰剂、csDMARDs和bDMARDs,差异均具统计学意义。
- ACR应答和HAQ-DI改善与DAS28结论一致,JAK抑制剂优于安慰剂或阳性对照。
- 安全性信号因对照不同而异:对比安慰剂不良事件增加,对比bDMARDs见SAEs和MACE风险增加。
核心结论
- JAK抑制剂在改善RA症状(DAS、ACR应答、HAQ-DI)方面优于安慰剂和阳性治疗。
- 安全性结局在累积荟萃分析中随时间不一致,可能与部分试验样本量小、罕见结局难以研究及随访短有关。
- 因试验把握度不足以确定安全性终点,JAK抑制剂的安全警告无法更早被预见。
研究设计与样本
- 系统综述与累积荟萃分析,检索Medline、Embase和Cochrane Library自建库至2024年6月11日。
- 纳入成人RA的随机对照试验,干预为JAK抑制剂,对照为安慰剂或阳性治疗(csDMARDs或bDMARDs)。
- 共筛查5971条引文,最终纳入31项独特试验的46篇报告,整体偏倚风险评为低。
- 采用随机效应模型计算合并效应估计,并评估异质性(I²统计量),PROSPERO注册号CRD42023460537。
主要发现
- DAS28-CRP改善:JAK抑制剂 vs 安慰剂 MD -1.06(95%CI -1.17至-0.95);vs csDMARDs MD -0.79(95%CI -1.05至-0.53);vs bDMARDs MD -0.35(95%CI -0.48至-0.23)。
- DAS28-ESR改善:JAK抑制剂 vs 安慰剂 MD -1.03(95%CI -1.21至-0.85);vs csDMARDs MD -0.54(95%CI -0.73至-0.36);vs bDMARDs MD -0.33(95%CI -0.65至-0.01)。
- 次要结局HAQ-DI和ACR应答在JAK抑制剂组也观察到类似改善。
- 安全性:JAK抑制剂 vs 安慰剂不良事件、治疗相关不良事件和严重感染风险显著增加;vs csDMARDs严重感染风险显著增加;vs bDMARDs SAEs和MACE风险显著增加。
关键数值与推荐
- 共纳入31项独特试验的46篇报告,整体偏倚风险为低。
- DAS28-CRP vs 安慰剂 MD -1.06(95%CI -1.17至-0.95),vs csDMARDs MD -0.79(95%CI -1.05至-0.53),vs bDMARDs MD -0.35(95%CI -0.48至-0.23)。
- DAS28-ESR vs 安慰剂 MD -1.03(95%CI -1.21至-0.85),vs csDMARDs MD -0.54(95%CI -0.73至-0.36),vs bDMARDs MD -0.33(95%CI -0.65至-0.01)。
- JAK抑制剂 vs 安慰剂不良事件、治疗相关不良事件和严重感染风险显著增加;vs csDMARDs严重感染风险显著增加;vs bDMARDs SAEs和MACE风险显著增加。
临床怎么用
- JAK抑制剂可视为优于csDMARDs或bDMARDs的有效、快速起效的口服治疗选择。
- 临床决策时应结合最新疗效与安全性证据,并注意特定高危患者人群的使用限制及相关安全警告。
局限性
- 部分试验样本量小,罕见结局难以研究,随访时间短,导致安全性结局在累积荟萃分析中随时间不一致。
- 试验把握度不足以确定安全性终点,因此无法更早预见JAK抑制剂的安全警告。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
OBJECTIVES: Janus kinase (JAK) inhibitors have been subject to multiple safety warnings recently. Patient treatment decisions should be evidence-based using the most up-to-date data on the efficacy and safety of JAK inhibitors. METHODS: In a systematic review and cumulative meta-analysis (CMA), Medline, Embase and the Cochrane Library were searched from inception to June 11, 2024. Eligible studies included randomised controlled trials (RCTs) of adult patients with rheumatoid arthritis (RA), treated with a JAK inhibitor, placebo or active treatment (conventional synthetic disease-modifying anti-rheumatic drugs [csDMARDs] or biologic DMARDs [bDMARDs]) and at least one of the protocol defined outcomes. Literature screening was conducted by two reviewers independently. Data extraction and risk-of-bias (ROB) assessment were done by one reviewer in a standardised manner and checked for accuracy by a second reviewer. The following outcomes were assessed: disease activity score erythrocyte sedimentation rate (DAS28-ESR) and DAS28 C-reactive protein (DAS28-CRP) [co-primary outcomes]; American College of Rheumatology (ACR) response (20/50/70), and Health Assessment Questionnaire-Disability Index (HAQ-DI) [secondary outcomes]; adverse events (AEs), treatment-related AEs (TRAEs) serious adverse events (SAEs), major adverse cardiac events (MACE), venous thromboembolism (VTE), serious infections and malignancies [safety outcomes]. Pooled effect-estimates were computed using a random-effects model and heterogeneity was assessed (I² statistic). PROSPERO registration (CRD42023460537). RESULTS: Of 5971 citations screened, 46 reports (of 31 unique trials) were included. Overall, ROB was considered 'low' across the trials. A statistically significant improvement in DAS was found for JAK inhibitors vs. placebo: (DAS28-CRP: mean difference [MD] -1.06, 95 % CI -1.17 to -0.95; DAS28-ESR:1.03, 95 % CI -1.21 to -0.85); versus csDMARDs (DAS28-CRP:0.79, 95 % CI -1.05 to -0.53; DAS28-ESR:0.54, 95 % CI -0.73 to -0.36); and versus bDMARDs (DAS28-CRP:0.35, 95 % CI -0.48 to -0.23; DAS28-ESR:0.33, 95 % CI -0.65 to -0.01). Similar efficacy improvements in secondary outcomes (HAQ-DI, ACR response) were also observed in patients treated with JAK inhibitors compared to placebo or active treatment. In patients treated with JAK inhibitors, a statistically significant increased risk of AEs, TRAEs and serious infections (versus placebo), serious infections (versus csDMARDs) and SAEs and MACE (versus bDMARDs) was found. CONCLUSION: JAK inhibitors are more effective than placebo and active treatment in improving symptoms of RA including DAS, ACR response, and HAQ-DI. This improvement in symptoms was clearly shown in the CMA over time as each new trial was published, in some instances from the second published trial. Safety outcomes were inconsistent over time in the CMA, probably owing to small sample sizes in some trials, difficulty in studying uncommon outcomes and short follow-up. Safety warnings for JAK inhibitors could not have been pre-empted sooner as the trials were not adequately powered to determine this. JAK inhibitors offer advantages to patients with RA and physicians should consider them an effective, fast-acting oral treatment compared to csDMARDs or bDMARDs, whilst acknowledging the limitations of their use in certain high-risk patient populations and the respective safety warnings.
引用
- O'Neill SM, Leahy J, Harte M, et al. The efficacy and safety of Janus kinase inhibitors in adults with rheumatoid arthritis: a systematic review and cumulative meta-analysis. Semin Arthritis Rheum. 2025. PMID: 41330174. doi:10.1016/j.semarthrit.2025.152885
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。