Multi-Cohort Trial Simulator

VISION-like randomized trial simulation with power calculations and virtual patient cohorts.

Trial Design Parameters

Trial Selection

Trial Preset

Preset sets default cohort size, randomization ratio, and target HR.

Study Design

Total Cohort Size

Randomization Ratio

Follow-up Duration (months)

Expected Treatment Effect

Target Hazard Ratio (OS)

VISION: HR=0.62 [0.52-0.74]

Assumptions (Editable)

Apply PSA prognostic effect

PSA effect exponent

Hazard scales by (PSA / 68.6) ^ exponent.

Apply Gleason prognostic effect

Gleason 8 hazard multiplier

Gleason 9-10 hazard multiplier

Apply PSMA SUVmax predictive effect (treatment arm only)

PSMA high-uptake effect (HR multiplier)

Applied if SUVmax ≥ 15 (cutoff-based; avoids per-unit linear extrapolation).

Dropout rate (%)

Baseline Weibull shape (k)

Baseline Weibull scale (months)

Simulation Parameters

Monte Carlo iterations

Ready to Simulate

Configure trial parameters to run virtual cohort simulation.

Assumptions and Restrictions

All assumptions are listed here and are adjustable in the form. If you do not accept an assumption, uncheck or edit it.

Baseline survival distribution is Weibull with user-specified shape and scale.
Dropout is applied as random censoring at a user-specified rate.
PSA and Gleason effects are prognostic only (apply to both arms).
PSMA SUVmax effect is predictive (treatment arm only).
Power is estimated by repeated simulation and log-rank p-value threshold (0.05).

Restriction: This simulator is for study design sensitivity analysis only. It does not replace published trial statistics or clinical decision-making.

Methodology and References

Methodology: Individual event times are generated from a Weibull baseline with user-selected parameters. A treatment hazard ratio is applied to the treatment arm. Prognostic and predictive effects are optional and user-controlled.

Survival curves are plotted as Kaplan-Meier step functions (time vs cumulative survival).
Hazard ratio is estimated via log-rank calculations across simulated arms.
Power is estimated by repeated simulation and the proportion of p-values < 0.05.

Validation approach (internal diagnostics): Compare simulated median HR and its CI to the target HR you entered. Use the direction and separation of the Kaplan-Meier curves as a sanity check. This is not an external clinical validation.

Restrictions: This simulator is for study design sensitivity analysis only and is not a clinical predictor. Outputs are only as valid as the assumptions you approve.

Primary trial references:

1. Sartor O, et al. Lutetium-177-PSMA-617 for Metastatic Castration-Resistant Prostate Cancer. N Engl J Med. 2021;385(12):1091-1103. doi:10.1056/NEJMoa2107322

2. Hofman MS, et al. [177Lu]Lu-PSMA-617 versus cabazitaxel in patients with metastatic castration-resistant prostate cancer (TheraP). Lancet. 2021;397(10276):797-804. doi:10.1016/S0140-6736(21)00237-3

Statistical methods:

3. Kalbfleisch JD, Prentice RL. The Statistical Analysis of Failure Time Data. 2nd ed. Wiley; 2002. doi:10.1002/9781118032985

4. Therneau TM, Grambsch PM. Modeling Survival Data: Extending the Cox Model. Springer; 2000. doi:10.1007/978-1-4757-3294-8