Treat and Extend as a Threshold Search: Monte Carlo Simulation of Extension Rules Under Study-Derived Durability Priors

Purpose:To model treat and extend (TREX) therapy as a threshold-search problem and compare common extension and shortening rules using durability distributions derived from published studies. Design: Computational modelling study using Monte Carlo (MC) simulation and Pareto multi-objective optimisation analysis of TREX rules. Subjects: 10,000 simulated eyes per durability distribution. Methods: Each simulated eye was assumed to have a fixed maximum dry interval (T_max), defined as the longest injection interval maintaining disease stability. Four TREX search rules were evaluated: +2/−2, +4/−4 with midpoint refinement, +4/−2, and a midpoint bracketing strategy. Deterministic pathways were enumerated for each T_max from 4 to 16 weeks. Durability priors were derived from published maintenance interval data from TENAYA/LUCERNE, the FARIT study, and a real-world aflibercept 2 mg cohort, with intermediate weekly survival values interpolated using a piecewise constant hazard interpolation. Monte Carlo simulation of 10,000 eyes was used to estimate visits to maintenance interval, cumulative overshoot, and maximum single overshoot. Pareto analysis identified dominated and non-dominated strategies. Main outcome measures: Number of visits to find maintenance interval, cumulative and maximum overshoot beyond a disease-controlling interval. Results: Under TENAYA/LUCERNE priors, +4/−2 required the fewest visits to maintenance, while +2/−2 had the lowest overshoot. The midpoint strategy achieved similar efficiency to +4/−2 with lower overshoot. Across all three durability distributions, +4/−4 was consistently dominated. The Pareto-optimal set varied by durability prior: midpoint performed favourably under longer-durability distributions but carried greater overshoot risk under shorter-durability priors. Conclusion: TREX rule choice involves explicit trade-offs between search efficiency and overshoot. Pareto analysis provides a structured framework for comparing extension strategies and tailoring rule choice to expected durability and clinical risk tolerance.