Introduction and Overview | Hypothesis tests | Design and analysis priors | Using the package | ICT-107 Phase II Trial Overview | 1. Bayes Factor Analysis | 2. Operating characteristics for actual sample sizes | 3. Power and sample size planning | 4. Informative design priors | 5. Unequal randomization probabilities | 6. Design Recommendations based on the calibration | Summary | References

Introduction | How to design a Bayesian trial: Overview of the calibration algorithm | Basic calibration | Operating-characteristic curves | Adding a constraint on the probability of compelling evidence | Full Bayes-frequentist calibration | Frequentist calibration | Hybrid calibration | Relationship to the two-stage design | Summary | References

Introduction | Bayes factors in single-arm phase II designs | Types of priors | Calibration modes | Vignette overview | Single-arm designs | Two-arm designs | References

Introduction | Design setup | Operating characteristics | [\mathcal{A}_{\mathrm{eq}}(n) | [\mathrm{freq_type1}^{\max}(n) | Calibration modes | Bayesian calibration | Frequentist calibration | Hybrid calibration | Full Bayes–frequentist calibration | Tuning parameters for frequentist calibration | The posterior probability threshold (\gamma_{\mathrm{eq}}) | The ROPE half-width (\delta) | The Point alternative (dp) | Summary

Introduction | Hypotheses and thresholds | Priors and calibration inputs | Analysis prior | Bayesian design prior | Frequentist point alternative | Interpretation of Bayesian and frequentist power | Bayesian power | Frequentist power | The function design_singlearm_bf() | Overview of the calibration algorithm | Interpretation of the corrected operating characteristics | Detailed example: Nonsmall cell lung cancer phase II trial | Optimal design search | Printing and summarizing the result | Accessing and plotting the results | Example with a constraint on compelling evidence for the null hypothesis | Summary | References

Introduction | Hypotheses and Bayes factors | Priors: design vs analysis | Design priors | Analysis priors | Overview of the calibration algorithm | Riociguat phase II trial: fixed-sample design and optimal two-stage design | The two-stage design function design_twoarm_twostage_bf() | Riociguat phase II trial: Setup | Riociguat phase II trial: Fixed-sample comparator via design_twoarm_onestage_bf() | Riociguat phase II trial: Optimal Bayesian two-stage design via design_twoarm_twostage_bf() | Interpretation of the small futility probability | A note on the expected sample size | Reducing the expected sample size substantially | One-stage design | Two-stage design without power cushion | Interpretation of the conv flag | How priors and thresholds affect the calibration grid (and runtime) | Design priors and required sample size | Evidence threshold $k$ and required sample size | Probability of compelling evidence for $H_0$ and feasibility | Practical recommendation for vignettes and examples | Summary | References

Introduction | Frequentist calibration: overview | Manual evaluation of a two-stage design | Optimal frequentist design | Interpreting the frequentist design | Practical recommendations for frequentist calibration | References

Introduction: Hybrid and full calibration | Bayesian calibration | Hybrid Bayes-frequentist calibration | Current implementation | Full Bayesian and frequentist calibration | Overview of the different calibration modes | Full calibration example | Why the design is selected | Comparison of calibration modes | Practical recommendations for full calibration | References

Hybrid calibration: Overview | Basics of hybrid calibration | Finding an optimal hybrid design | Interpretation of hybrid calibration | Bayesian and frequentist expected sample sizes | Frequentist expected sample size | Bayesian expected sample size | Sensitivity to the $H_1$ design prior | Adding a compelling-evidence constraint under $H_0$ | Power cushions in hybrid calibration | Interpretation of the power_cushion parameter | Practical guidance | Practical recommendations for hybrid calibration | References

Introduction | Setup | The region of practical equivalence (ROPE) | Design and analysis priors | ROPE decision illustrations | 1) Equivalence accepted | 2) Type-I error: equivalence concluded under the null hypothesis | 3) Indecisive result | 4) Clear non-equivalence | A first ROPE-based design example | Summarizing the design | Plotting the design | Example 1: Oncology phase II equivalence trial | Clinical hypotheses and ROPE | Analysis and design priors | Calibration target | Visualization | Example 2: Sensitivity analysis via grid exploration | Example 3: revisiting the first example with PCE(H0) and frequentist power | Summary | References

Introduction | Bayes factors in single-arm phase II designs | Types of priors | Calibration modes | Vignette overview | Single-arm designs | Two-arm designs | References

Introduction | Design setup | Operating characteristics | [\mathcal{A}_{\mathrm{eq}}(n) | [\mathrm{freq_type1}^{\max}(n) | Calibration modes | Bayesian calibration | Frequentist calibration | Hybrid calibration | Full Bayes–frequentist calibration | Tuning parameters for frequentist calibration | The posterior probability threshold (\gamma_{\mathrm{eq}}) | The ROPE half-width (\delta) | The Point alternative (dp) | Summary

Introduction and Overview | Hypothesis tests | Design and analysis priors | Using the package | ICT-107 Phase II Trial Overview | 1. Bayes Factor Analysis | 2. Operating characteristics for actual sample sizes | 3. Power and sample size planning | 4. Informative design priors | 5. Unequal randomization probabilities | 6. Design Recommendations based on the calibration | Summary | References

Introduction | How to design a Bayesian trial: Overview of the calibration algorithm | Basic calibration | Operating-characteristic curves | Adding a constraint on the probability of compelling evidence | Full Bayes-frequentist calibration | Frequentist calibration | Hybrid calibration | Relationship to the two-stage design | Summary | References

Introduction | Hypotheses and thresholds | Priors and calibration inputs | Analysis prior | Bayesian design prior | Frequentist point alternative | Interpretation of Bayesian and frequentist power | Bayesian power | Frequentist power | The function design_singlearm_bf() | Overview of the calibration algorithm | Interpretation of the corrected operating characteristics | Detailed example: Nonsmall cell lung cancer phase II trial | Optimal design search | Printing and summarizing the result | Accessing and plotting the results | Example with a constraint on compelling evidence for the null hypothesis | Summary | References

Introduction | Hypotheses and Bayes factors | Priors: design vs analysis | Design priors | Analysis priors | Overview of the calibration algorithm | Riociguat phase II trial: fixed-sample design and optimal two-stage design | The two-stage design function design_twoarm_twostage_bf() | Riociguat phase II trial: Setup | Riociguat phase II trial: Fixed-sample comparator via design_twoarm_onestage_bf() | Riociguat phase II trial: Optimal Bayesian two-stage design via design_twoarm_twostage_bf() | Interpretation of the small futility probability | A note on the expected sample size | Reducing the expected sample size substantially | One-stage design | Two-stage design without power cushion | Interpretation of the conv flag | How priors and thresholds affect the calibration grid (and runtime) | Design priors and required sample size | Evidence threshold $k$ and required sample size | Probability of compelling evidence for $H_0$ and feasibility | Practical recommendation for vignettes and examples | Summary | References

Introduction | Frequentist calibration: overview | Manual evaluation of a two-stage design | Optimal frequentist design | Interpreting the frequentist design | Practical recommendations for frequentist calibration | References

Introduction: Hybrid and full calibration | Bayesian calibration | Hybrid Bayes-frequentist calibration | Current implementation | Full Bayesian and frequentist calibration | Overview of the different calibration modes | Full calibration example | Why the design is selected | Comparison of calibration modes | Practical recommendations for full calibration | References

Hybrid calibration: Overview | Basics of hybrid calibration | Finding an optimal hybrid design | Interpretation of hybrid calibration | Bayesian and frequentist expected sample sizes | Frequentist expected sample size | Bayesian expected sample size | Sensitivity to the $H_1$ design prior | Adding a compelling-evidence constraint under $H_0$ | Power cushions in hybrid calibration | Interpretation of the power_cushion parameter | Practical guidance | Practical recommendations for hybrid calibration | References

Introduction | Setup | The region of practical equivalence (ROPE) | Design and analysis priors | ROPE decision illustrations | 1) Equivalence accepted | 2) Type-I error: equivalence concluded under the null hypothesis | 3) Indecisive result | 4) Clear non-equivalence | A first ROPE-based design example | Summarizing the design | Plotting the design | Example 1: Oncology phase II equivalence trial | Clinical hypotheses and ROPE | Analysis and design priors | Calibration target | Visualization | Example 2: Sensitivity analysis via grid exploration | Example 3: revisiting the first example with PCE(H0) and frequentist power | Summary | References