Case studies

How simulation reshapes trial design

Explore realistic scenarios where Monte Carlo simulation helped teams de‑risk protocol assumptions, refine sample size, and improve trial decision‑making before first patient first visit.

Propose a studyTry a simulation
Phase 2Phase 3MetabolicOncologySuperiorityPlatform trials
Phase 2 · Type 2 diabetesSuperiority

Detecting under‑power risk in a glycaemic control trial

A sponsor planned a two‑arm trial targeting a modest HbA1c difference with optimistic variance assumptions. Simulation runs revealed a high probability of sub‑80% power under more realistic variability scenarios.

Original plan
220 participants · fixed follow‑up
After simulation
280 participants · extended follow‑up

Outcome: Reduced chance of an inconclusive result by adjusting sample size and visit schedule before protocol finalisation.

Phase 3 · OncologyTime‑to‑event

Stress‑testing event‑driven timelines

A global oncology trial relied on historical hazard rates and optimistic recruitment curves. Scenario simulations highlighted how modest delays in site activation and lower event rates could push database lock 9–12 months later than planned.

Planned duration
30 months to final analysis
Simulated range
30–42 months across scenarios

Outcome: Protocol and governance team agreed contingency triggers and communication plans anchored in simulation outputs.

Platform trialMulti‑arm

Evaluating operating characteristics in a platform design

A multi‑arm platform considered staggered entry of experimental arms and shared control. Simulation helped quantify false positive and false negative rates under varying recruitment and drop‑out assumptions.

Arms evaluated
3 experimental + shared control
Runs per scenario
10,000 Monte Carlo simulations

Outcome: Governance committee selected more conservative decision boundaries aligned with long‑run error control.

Early developmentGo / No‑go

Clarifying go/no‑go criteria before first patient

A biotech team needed to align internal stakeholders on credible success criteria for a small proof‑of‑concept study. Simulation mapped observed effect sizes to posterior beliefs about true efficacy.

Planned sample
60 participants · 2 arms
Decision rule
Proceed only if probability of achieving target effect exceeds pre‑agreed threshold

Outcome: Transparent, simulation‑based framework for go/no‑go decisions agreed before recruitment.

Have a trial concept you'd like to explore?

Case studies here are illustrative. For real‑world collaborations, sensitive protocol details are never shared without appropriate agreements.

Discuss a potential case study