Farinasso Cecília Menezes, Ferreira Vinícius Lins, Medeiros Flávia Cordeiro, da Rocha Aline Pereira, Parreira Patrícia do Carmo Silva, Oliveira Layssa Andrade, Marra Lays Pires, Lucchetta Rosa Camila, de Oliveira Haliton Alves
Health Technology Assessment Unit, Hospital Alemão Oswaldo Cruz, São Paulo, São Paulo, Brazil.
Health Technology Assessment Unit, Hospital Alemão Oswaldo Cruz, São Paulo, São Paulo, Brazil.
Value Health Reg Issues. 2025 May;47:101088. doi: 10.1016/j.vhri.2025.101088. Epub 2025 Feb 24.
Matching-adjusted indirect comparisons (MAICs) can be used in case of cross-trial heterogeneity or availability of only single-arm trials. Although the National Institute for Health and Care Excellence (NICE) provides MAIC-development orientation, many still do not adhere to it. Our goal was to map MAIC oncology studies and whether NICE recommendations were observed.
We included MAIC studies comparing treatments in oncology from 2010. We searched PubMed, Embase, and the Cochrane Library up to October 1, 2024. We analyzed MAIC characteristics such as previous systematic reviews, whether the analysis was anchored or unanchored, selection of variables, and individual patient data (IPD) reporting. We adopted NICE recommendations for the assessment of MAIC studies.
We included 117 MAIC studies, which often explored multiple myeloma (n = 19%) and non-small cell lung cancer (17%) more frequently. Most MAICs were unanchored (72%), with an average of 1.9 comparisons per study. MAIC studies generally reported using pseudo-IPD (69%) but did not report the source of IPD (78%). In general, MAICs did not conduct systematic reviews to select trials for inclusion (66%). The average sample size reduction, in comparison with the original trials, was 44.9%. Only 3 MAICs fulfilled all NICE recommendations. The least reported aspects were the adjustment for all effect modifiers and prognostic variables (for unanchored MAICs), evidence of effect modifier status, and distribution of weights.
Most MAIC models did not follow NICE recommendations. Our review highlights the importance of rigorous methodological standards and thorough reporting of MAIC studies to enhance their credibility.
在存在交叉试验异质性或仅有单臂试验可用的情况下，可使用匹配调整间接比较（MAIC）。尽管英国国家卫生与临床优化研究所（NICE）提供了MAIC开发指南，但许多人仍未遵循。我们的目标是梳理MAIC肿瘤学研究情况以及是否遵循了NICE的建议。
我们纳入了2010年以来比较肿瘤学治疗方法的MAIC研究。检索了截至2024年10月1日的PubMed、Embase和Cochrane图书馆。我们分析了MAIC的特征，如先前的系统评价、分析是锚定的还是非锚定的、变量选择以及个体患者数据（IPD）报告情况。我们采用NICE的建议来评估MAIC研究。
我们纳入了117项MAIC研究，这些研究经常更频繁地探讨多发性骨髓瘤（n = 19%）和非小细胞肺癌（17%）。大多数MAIC是非锚定的（72%），每项研究平均有1.9次比较。MAIC研究通常报告使用了伪IPD（69%），但未报告IPD的来源（78%）。总体而言，MAIC未进行系统评价以选择纳入试验（66%）。与原始试验相比，平均样本量减少了44.9%。只有3项MAIC符合所有NICE建议。报告最少的方面是对所有效应修饰因素和预后变量的调整（对于非锚定MAIC）、效应修饰因素状态的证据以及权重分布。
大多数MAIC模型未遵循NICE建议。我们的综述强调了严格的方法学标准和MAIC研究全面报告的重要性，以提高其可信度。
