Singh Janharpreet, Anwer Sumayya, Palmer Stephen, Saramago Pedro, Thomas Anne, Dias Sofia, Soares Marta O, Bujkiewicz Sylwia
Biostatistics Research Group, Department of Population Health Sciences, University of Leicester, Leicester, UK.
Centre for Reviews and Dissemination, University of York, York, UK.
Med Decis Making. 2025 Jan;45(1):17-33. doi: 10.1177/0272989X241295665. Epub 2024 Nov 18.
Multi-indication cancer drugs receive licensing extensions to include additional indications, as trial evidence on treatment effectiveness accumulates. We investigate how sharing information across indications can strengthen the inferences supporting health technology assessment (HTA).
We applied meta-analytic methods to randomized trial data on bevacizumab, to share information across oncology indications on the treatment effect on overall survival (OS) or progression-free survival (PFS) and on the surrogate relationship between effects on PFS and OS. Common or random indication-level parameters were used to facilitate information sharing, and the further flexibility of mixture models was also explored.
Treatment effects on OS lacked precision when pooling data available at present day within each indication separately, particularly for indications with few trials. There was no suggestion of heterogeneity across indications. Sharing information across indications provided more precise estimates of treatment effects and surrogacy parameters, with the strength of sharing depending on the model. When a surrogate relationship was used to predict treatment effects on OS, uncertainty was reduced only when sharing effects on PFS in addition to surrogacy parameters. Corresponding analyses using the earlier, sparser (within and across indications) evidence available for particular HTAs showed that sharing on both surrogacy and PFS effects did not notably reduce uncertainty in OS predictions. Little heterogeneity across indications meant limited added value of the mixture models.
Meta-analysis methods can be usefully applied to share information on treatment effectiveness across indications in an HTA context, to increase the precision of target indication estimates. Sharing on surrogate relationships requires caution, as meaningful precision gains in predictions will likely require a substantial evidence base and clear support for surrogacy from other indications.
We investigated how sharing information across indications can strengthen inferences on the effectiveness of multi-indication treatments in the context of health technology assessment (HTA).Multi-indication meta-analysis methods can provide more precise estimates of an effect on a final outcome or of the parameters describing the relationship between effects on a surrogate endpoint and a final outcome.Precision of the predicted effect on the final outcome based on an effect on the surrogate endpoint will depend on the precision of the effect on the surrogate endpoint and the strength of evidence of a surrogate relationship across indications.Multi-indication meta-analysis methods can be usefully applied to predict an effect on the final outcome, particularly where there is limited evidence in the indication of interest.
随着治疗有效性的试验证据不断积累，多适应症癌症药物会获得许可扩展以纳入更多适应症。我们研究了跨适应症共享信息如何强化支持卫生技术评估（HTA）的推断。
我们将荟萃分析方法应用于贝伐单抗的随机试验数据，以跨肿瘤学适应症共享关于总生存期（OS）或无进展生存期（PFS）的治疗效果以及PFS与OS效果之间替代关系的信息。使用共同或随机的适应症水平参数来促进信息共享，同时也探讨了混合模型的进一步灵活性。
单独汇总目前每个适应症内可用的数据时，对OS的治疗效果缺乏精确性，尤其是对于试验较少的适应症。没有迹象表明各适应症之间存在异质性。跨适应症共享信息能更精确地估计治疗效果和替代参数，共享的强度取决于模型。当使用替代关系来预测对OS的治疗效果时，只有在除了替代参数之外还共享对PFS的效果时，不确定性才会降低。使用特定HTA可获得的更早、更稀疏（适应症内和适应症间）的证据进行的相应分析表明，共享替代和PFS效果并未显著降低OS预测中的不确定性。各适应症之间几乎没有异质性意味着混合模型的附加值有限。
荟萃分析方法可有效地应用于在HTA背景下跨适应症共享治疗有效性信息，以提高目标适应症估计的精确性。在替代关系上进行共享需要谨慎，因为预测中显著的精确性提高可能需要大量的证据基础以及来自其他适应症对替代关系的明确支持。
我们研究了跨适应症共享信息如何在卫生技术评估（HTA）背景下强化对多适应症治疗有效性的推断。多适应症荟萃分析方法可以更精确地估计对最终结局的影响或描述替代终点与最终结局之间关系的参数。基于对替代终点的影响预测对最终结局的效果的精确性将取决于对替代终点的效果的精确性以及各适应症间替代关系的证据强度。多适应症荟萃分析方法可有效地应用于预测对最终结局的影响，特别是在感兴趣的适应症证据有限的情况下。
