RECIST 1.1

RECIST 1.1 quantifies measurement variability in radiologic assessments of tumor burden from computed tomography (CT) scans to evaluate impacts on objective response rates (ORR) and progression rates in clinical trials.

Key Features:

• Hierarchical Modeling: Employs a hierarchical model to estimate the distribution of measurement variability using data from clinical trial CT scans.
• Simulation-Based Analysis: Uses simulation to calculate probabilities representing the impact of measurement errors on categorical diagnoses across scenarios.
• Algorithm Development: Generates a 95% central range for ORR and progression rates from simulated data to evaluate the reliability of those metrics.
• Validation with External Data: Validated on an external dataset with reported coverage of 93% for the estimated measurement-variability distribution and 100% for the evaluation algorithm.

Scientific Applications:

• Clinical Trial Data Analysis: Quantifies variation in ORR and progression rates attributable to measurement variability to aid interpretation of trial endpoints.
• Enhanced Reliability of Radiologic Assessments: Assesses the effect of measurement errors on radiologic tumor burden assessments to inform evaluation of disease monitoring and treatment response.

Methodology:

Combines hierarchical modeling and simulation techniques by estimating measurement-variability distributions from clinical trial datasets and simulating their impact on diagnostic outcomes.