Demystifying different types of Bayesian and frequentist designs that may be used for first-in-human or other types of phase I trials in oncology
ABSTRACT:
Historically, phase 1 trials in oncology have mostly used the 3+3 design for dose finding, even though many alternatives have been proposed over the years. That design is useful mainly for its simplicity, but it suffers from disadvantages that can be circumvented by designs which provide for more flexibility and an enhanced performance in terms of various desirable outcomes.
- In this webinar, our goal is to demystify different types of Bayesian and frequentist designs that may be used for first-in-human or other types of phase 1 trials in oncology.
- We will start by reviewing the recent literature describing general trends in the design of phase 1 trials over the past few years.
- We will then summarize the most salient features of Bayesian and frequentist designs in this setting.
- Finally, we will formulate recommendations for use of phase-1 designs in different situations and touch upon the inclusion of expansion cohorts in phase 1 studies.
- The interested registrant should notice that this webinar will be followed by a similar one dedicated to phase 2 trials, which will cover different statistical caveats, including early consideration of randomization, issues related to sample-size calculation and the use of biomarkers.
KEY TAKEAWAYS:
- Be updated on current trends in phase-1 designs in oncology
- Gain insights into the advantages and disadvantages of various Bayesian and frequentist approaches to the design of phase 1 trials
SPEAKERS:
