优化晚期经典霍奇金淋巴瘤的风险分层:临床预测模型的批判性分析
本文综述晚期经典霍奇金淋巴瘤(cHL)临床预测模型的演变、优势与局限。国际预后评分(IPS)于1998年提出,2012年更新、2015年简化为IPS-3,因治疗进步预测性能下降。2023年HoLISTIC联盟提出晚期霍奇金淋巴瘤国际预后指数(A-HIPI),纳入连续变量,但对老年患者的适用性尚不确定,需进一步验证。现有模型均未纳入中期PET/CT等动态治疗反应标志物,未来需整合动态生物标志物与治疗反应指标。
要点
- IPS于1998年提出,2012年更新为updated IPS,2015年简化为IPS-3,因cHL治疗进步预测性能下降。
- 2023年HoLISTIC联盟提出A-HIPI,纳入连续变量以提供更精确的风险评估,但对老年患者适用性尚不确定。
- 现有模型均未纳入中期PET/CT等动态治疗反应标志物,未来需整合动态生物标志物与治疗反应指标。
核心结论
- 现有晚期cHL临床预测模型因治疗进步预测性能下降,A-HIPI纳入连续变量可提供更精确风险评估,但对老年患者适用性尚不确定,需进一步验证。
- 现有模型均未纳入中期PET/CT等动态治疗反应标志物,未来需整合动态生物标志物与治疗反应指标以优化风险分层。
研究设计与样本
- 综述,讨论晚期cHL临床预测模型的演变、优势与局限及其临床意义。
主要发现
- IPS于1998年提出,2012年更新为updated IPS,2015年简化为IPS-3。
- 2023年HoLISTIC联盟提出A-HIPI,纳入连续变量,但对老年患者适用性尚不确定。
- 现有模型均未纳入中期PET/CT等动态治疗反应标志物,而后者已显示较强预后价值。
局限性
- A-HIPI对老年患者的适用性尚不确定,需进一步验证研究。
- 现有模型均未纳入中期PET/CT等动态治疗反应标志物。
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
Classical Hodgkin lymphoma (cHL) is a haematological malignancy with high curability; however, prognosis varies significantly based on clinical and biological factors. To enhance risk stratification, several clinical prediction models have been developed over time, particularly for advanced-stage cHL. The International Prognostic Score (IPS), introduced in 1998, was the first widely adopted model, later refined in 2012 (updated IPS) and further simplified in 2015 (IPS-3). Despite their prognostic utility, these models have demonstrated declining predictive performance due to advancements in cHL treatment. In response, the HoLISTIC consortium recently introduced the Advanced-Stage Hodgkin Lymphoma International Prognostic Index (A-HIPI) in 2023. Unlike previous models, A-HIPI incorporates continuous variables, aiming to provide a more precise risk assessment. However, its applicability to older patients remains uncertain, necessitating further validation studies. Additionally, none of the existing models incorporate dynamic treatment response markers such as interim positron emission tomography/computed tomography (PET/CT), which have shown strong prognostic value. This review comprehensively discusses the evolution, strengths and limitations of these prediction models, their clinical implications and the necessity for future refinements integrating dynamic biomarkers and treatment response indicators. The integration of machine learning and multi-omics approaches could further enhance risk stratification, improve treatment personalization and optimize patient outcomes in cHL.
引用
- Koca O, Eskazan AE. Refining the risk stratification in advanced-stage classical Hodgkin lymphoma: A critical analysis of clinical prediction models. Br J Haematol. 2025. PMID: 40843643. doi:10.1111/bjh.70115
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。