抗生素临床试验:过去回顾、当前挑战与潜在解决方案
该综述对2013至2022年高影响医学及专科期刊发表的抗生素随机对照试验(RCT)进行定向文献回顾,共纳入193项试验。多数试验由非产业来源支持、多中心、双臂平行设计且仅纳入成人,中位样本量500例,但中低收入国家参与严重不足。作者建议优先采用适应性、平台和病原导向等创新设计,并整合真实世界数据、协调监管要求、促进全球公平参与。
要点
- 抗生素RCT以多中心、双臂平行传统设计为主,平台等创新设计应用有限。
- 非产业支持试验占72.0%,中位样本量500例,产业支持试验样本量更大但持续时间更短。
- 高抗菌药物耐药负担国家参与严重不足,仅26.9%试验涉及中低收入国家站点。
核心结论
- 传统平行RCT仍是抗生素临床试验的主流设计,需优先发展适应性、平台和病原导向等创新方法以提升效率与相关性。
- 未来试验应整合真实世界数据、协调监管要求并支持全球公平参与。
研究设计与样本
- 对2013至2022年发表在高影响医学及专科期刊的抗生素RCT进行定向文献回顾,共识别193项试验。
- 提取数据包括研究人口学、试验特征和设计要素。
主要发现
- 72.0%的试验由非产业来源支持,90.7%为多中心,88.6%为双臂,98.4%为平行设计,68.4%仅纳入成人。
- 仅52项试验(26.9%)涉及中低收入国家站点。
- 中位样本量为500例(IQR 253-777),产业支持试验样本量显著更大。
- 非产业支持试验持续时间长于产业支持试验(中位36个月 vs 18个月)。
- 所有试验均采用频率学派方法。
关键数值与推荐
- 非产业支持试验占比72.0%,中低收入国家站点参与试验占比26.9%。
- 中位样本量500例(IQR 253-777)。
- 非产业支持与产业支持试验中位持续时间分别为36个月和18个月。
临床怎么用
- 开展抗生素RCT时可参考当前主流设计特征,并考虑采用适应性、平台或病原导向设计以提高效率。
- 应关注高抗菌药物耐药负担地区的参与不足问题,促进全球公平参与。
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
BACKGROUND: Increasing antimicrobial resistance (AMR) is a global public health concern. Rapid and effective evaluation of interventions for antibiotic-resistant organisms is needed. Randomized controlled trials (RCTs) provide the strongest evidence of causal effects of an intervention. However, conducting antibiotic RCTs has many challenges. We explored antibiotic RCT designs and methodology, to gain an understanding of current approaches and potential solutions for future antibiotic trials. METHODS: We conducted a targeted literature review for antibiotic RCTs from 2013 to 2022 published in high impact medical and subspecialty journals. Data extracted included study demographics, characteristics, and trial design features. RESULTS: The search identified 193 trials. The majority were supported by non-industry sources (72.0%), multicenter (90.7%), 2-arm trials (88.6%) with a parallel design (98.4%) and enrolled adults only (68.4%). Only 52 trials (26.9%) involved sites in low- or middle-income countries (LMICs). The median sample size was 500 (interquartile range [IQR] 253-777) with a significantly larger sample size in industry supported trials. However, non-industry supported trials were of longer duration than industry supported trials (median 36 vs 18 months). All trials utilized a frequentist method. CONCLUSIONS: The most common trials were multicenter, 2-arm trials with parallel traditional RCT designs. Although platform trials are currently ongoing, traditional designs were overwhelmingly used. Study involvement in countries with the highest AMR burden were grossly under-represented. Innovative trial methodologies-such as adaptive, platform, and pathogen-directed designs-should be prioritized to improve efficiency, flexibility, and relevance. Future trials must integrate real-world data, harmonize regulatory requirements, and support equitable global participation.
引用
- Hardy MJ, Harris PNA, Mo Y, et al. Antibiotic Clinical Trials: A Review of the Past, Present Challenges, and Potential Solutions. Clin Infect Dis. 2026. PMID: 42227053. doi:10.1093/cid/ciag017
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。