Multi-indication Evidence Synthesis in Oncology Health Technology Assessment: Meta-analysis Methods and Their...

Area of Science:

• Oncology
• Biostatistics
• Health Economics

Background:

• Multi-indication drugs gain new licenses as evidence grows.
• Investigating information sharing across indications for health technology assessment (HTA).

Purpose of the Study:

• To assess how sharing data across indications strengthens inferences for HTA.
• To improve precision of treatment effect estimates for multi-indication drugs.

Main Methods:

• Applied meta-analytic methods to bevacizumab trial data across oncology indications.
• Utilized common/random indication-level parameters and mixture models.
• Examined treatment effects on overall survival (OS) and progression-free survival (PFS), and their surrogate relationship.

Main Results:

• Pooling data within indications lacked precision, especially for indications with few trials.
• Information sharing across indications increased precision of treatment effects and surrogacy parameters.
• Sharing PFS effects alongside surrogacy parameters reduced uncertainty in OS predictions.
• Limited heterogeneity across indications showed minimal added value from mixture models.

Conclusions:

• Meta-analysis methods effectively share information across indications to enhance HTA precision.
• Caution is advised when sharing surrogate relationships; substantial evidence is needed for precision gains.