MERIT: Multiple-dose Randomized Phase II Trial Design for Dose Optimization and Sample Size Determination

Number of Doses:

2 3 4

Toxicity Rates:

Null $\phi_{T,0}$

Alternative $\phi_{T,1}$

Efficacy Rates:

Null $\phi_{E,0}$

Alternative $\phi_{E,1}$

Global Type I Error Rate:

Generalized Power:

Power I Power II

Inlucde toxicity and futility monitoring

Interim Times:

Input the fraction of the total sample size at interims, separated by space.

Efficacy

Toxicity

Stopping Criteria:

Stop for futility if $p(\pi_{E,j} < \phi_{E,1} | data) > C_E$, where $C_E$

Stop for toxicity if $p(\pi_{T,j} > \phi_{T,1} | data) > C_T$, where $C_T$

Setting to Optimize the Design:

Correlation between toxicity and efficacy

positive negative

Correlation

Number of simulations

Seeds of the random number generator

MERIT Design

Note: 'NA' means that this endpoint will not be used to make go/no-go decision at the interim

Enter Simulation Scenarios:

For each scenario, enter true toxicity and efficacy rate of each dose level:

Number of simulations

Set seed

Operating Characteristics

Isotonic Transformation:

This tab performs isotonic regression on interim or final toxicity and efficacy data. Interim and final decisions are made by simply comparing the resulting isotonic transformed numbers of toxicity and efficacy with the decision boundaries provided by the MERIT design.

Enter Trial Data:

Note: a higher dose level represents a higher dosage.

Number of doses:

2 3 4

Dose level:

Sample size:

Number of toxicity:

Number of efficacy:

1

2

1

2

3

1

2

3

4

Isotonic Transformaion

Yang, P., Li, D., Lin, R., Huang, Bo., & Yuan, Y. (2024). Design and Sample Size Determination for Multiple-dose Randomized Phase II Trials for Dose Optimization. Statistics in Medicine, 43(15), 2972-2986.