临床表型在结核病精准医疗中的应用前景
本文探讨利用临床表型指导结核病精准治疗的前景。临床表型由患者特征、影像范围、菌负荷、药敏和宿主反应定义,需通过组学、模型引导精准给药和人工智能算法在临床试验中验证以优化分层治疗。治疗中可通过菌负荷定量、转录组和表观遗传生物标志物及影像症状评分监测反应,实现及时调整药物、防止耐药并缩短疗程治愈。
要点
- 临床表型整合患者特征、影像、菌负荷、药敏和宿主反应,可指导结核病分层治疗。
- 组学、模型引导精准给药和人工智能算法需在临床试验中验证以改善患者分类。
- 菌负荷定量、转录组和表观遗传标志物及影像症状评分可用于无痰监测治疗反应。
核心结论
- 基于临床表型的整合监测可及时调整药物,快速降低菌负荷,防止耐药并实现短疗程无复发治愈。
研究设计与样本
- 本文为综述,讨论利用临床表型、组学、模型引导精准给药和人工智能算法优化结核病分层治疗与监测。
主要发现
- 临床表型由患者特征、影像范围、菌负荷、药敏和宿主反应定义。
- 组学、模型引导精准给药和人工智能算法需在临床试验中适应和验证以改善患者分类。
- 菌负荷定量、转录组和表观遗传生物标志物可用于无痰监测治疗反应。
- 影像和症状评分工具可评估疾病负担,许多工具适用于高流行地区。
临床怎么用
- 治疗前基于临床表型对患者分层,以选择个体化方案。
- 治疗中整合菌负荷、转录组/表观遗传标志物及影像症状评分监测反应,及时调整药物。
局限性
- 组学、模型引导精准给药和人工智能算法尚未在临床试验中充分验证。
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
Developing shorter treatment regimens for tuberculosis requires careful characterization of the clinical phenotype, which is defined by patient characteristics, radiological extent of disease, mycobacterial burden, drug susceptibility, and host response. Advances in 'omics and model-informed precision dosing, as well as integrated algorithms using artificial intelligence, need to be adapted and validated in clinical trials to improve classification of patients for stratified treatment. When treatment is initiated based on the clinical phenotype, monitoring of treatment response can be improved by quantification of bacterial load, transcriptomic and epigenetic biosignatures for sputum-free monitoring, and assessing disease burden by radiological and symptom scoring tools. Many of these tools are suitable for high-endemic settings. Such integrated monitoring allows prompt drug adjustments for rapid reduction in bacterial load, which prevents development of drug resistance and achieves relapse-free cure even with shorter treatment.
引用
- Niward K, Braian C, Svärdhagen G, et al. Future Prospects for Using Clinical Phenotypes in Tuberculosis Precision Medicine-An Approach for Clinical Management. Clin Infect Dis. 2026. PMID: 41537599. doi:10.1093/cid/ciaf663
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。