A Policy Framework for Multi-Indication Evidence Synthesis in Oncology Health Technology Assessment (HTA)

OBJECTIVES: Health Technology Assessment (HTA) typically disregards evidence of a treatment’s effect on other indications, even when relevant, leading to high uncertainty. This study introduces a novel framework to guide HTA in appropriately considering multi-indication evidence to improve decisions and reduce uncertainty.
METHODS: The framework addresses key elements of multi-indication evidence synthesis, including selection of appropriate methods and incorporation of expert judgements. It builds on recent research, including an application to bevacizumab in oncology and a simulation study examining the performance of alternative methods. The framework quantifies cost-effectiveness impacts for two of bevacizumab’s appraisals.
RESULTS: Multi-indication evidence synthesis can improve decision making by increasing precision while generating unbiased and calibrated estimates. Important precision gains were shown in bevacizumab’s case study, even in early indications. For example, even the weaker hierarchical methods yielded an Overall Survival (OS) hazard ratio in breast cancer of 0.88 (95% credible interval: 0.78, 0.98), compared to 0.89 (0.73, 1.04) using only target indication evidence alone, leading to a reduction in decision uncertainty.
Under heterogeneity, where strong sharing assumptions may not hold, the sparseness typical of HTA data structures (few studies per indication and few indications overall) limits the applicability of complex methods like mixture or surrogacy models. While hierarchical models can still provide unbiased estimates in such cases, precision gains may be low making the burden of evidence identification and analyses potentially unjustified.
The framework further outlines when and how to share, what to share on (whether OS or surrogacy) and how to gather and incorporate clinical judgement.
CONCLUSIONS: Sharing information from plausibly related indications can strengthen HTA decisions. This framework defines a structured approach to integrating multi-indication evidence into HTA decision making.