Challenge:
Our client was preparing a first-in-human phase 1/2a study for a new anti-cancer agent that showed potent anti-tumor activity in cancer models.
The client wanted help designing a dose escalation procedure that would allow flexible cohort sizes, facilitate the selection of doses for evaluation in expanded cohorts, and ultimately to determine the recommended phase 2 dose.
Solution:
Cytel developed an adaptive Bayesian Logistic Regression Model (BLRM) to estimate the dose toxicity relationship based on accumulating data, and used this model to design a procedure for escalation guided by the overdose control principle.
The properties of the escalation procedure were evaluated through simulations performed in EAST ESCALATE, and shown to compare favorably to a standard 3+3 design.
Read our Introduction to the BLRM method here
Value Added:
The dose escalation procedure allows flexible cohort sizes, thereby providing the safety review committee the freedom to make dose recommendations that respond to emerging data and enrollment variability.
The Bayesian model can be used to estimate toxicity rates at any dose using all data accumulated in the trial, facilitating an assessment of the best doses to carry forward for further evaluation, and ultimately to determine the recommended dose for phase 2a.
To find out more about dose escalation in East Escalate watch the replay of our recent webinar below.
Are you interested in applying adaptive trials in oncology? Click the button below to download our article ' Are Adaptive Trials the Answer to Oncology Development Success?'
Further Resources
Cytel Animation: Modern Dose Escalation Phase 1 trials
Blog: A Bayesian approach to Phase 1 combination trials in Oncology
