Advancing unanchored simulated treatment comparisons: A novel implementation and simulation study.

Area of Science:

• Health Technology Assessment
• Biostatistics
• Comparative Effectiveness Research

Background:

• Health technology assessment (HTA) bodies increasingly use population-adjusted indirect comparisons for treatment evaluation.
• Unanchored indirect comparisons are crucial for single-arm trials lacking a common comparator.
• Unanchored simulated treatment comparison (STC) is underutilized due to implementation complexities.

Purpose of the Study:

• To develop a novel, understandable implementation of unanchored STC for HTA.
• To derive a marginal treatment effect without aggregation bias for single-arm trial comparisons.
• To address the underutilization of STC in HTA decision-making.

Main Methods:

• Incorporation of standardization/marginalization and the NORmal To Anything (NORTA) algorithm for covariate sampling.
• Utilized non-parametric bootstrap for uncertainty quantification.
• Proposed separate standard error calculation for individual patient data (IPD) and comparator studies.

Main Results:

• The proposed unanchored STC approach was evaluated via a simulation study focusing on binary outcomes.
• Findings demonstrated that the developed method is asymptotically unbiased.
• The approach ensures appropriate quantification of uncertainty in treatment effect estimation.

Conclusions:

• Unanchored STC offers a robust methodology for population-adjusted indirect comparisons with single-arm studies.
• This novel implementation facilitates better understanding and application of STC in HTA.
• Unanchored STC should be considered for HTA decision-making involving single-arm trial data.