The escalation package
escalation is an R package for dose-escalation clinical trials, providing a consistent, extensible, modular approach.
It largely uses existing R implementations of popular dose-finding designs, including the continual reassessment method (CRM) design from the
dfcrm package, the Bayesian optimal interval (BOIN) design from the
BOIN package, and the perennial 3+3.
escalation bestows these model-fitting functions with a common interface.
This allows the provision of add-on behaviours that affect the way that doses are selected, or trial stopping decisions are made.
For example, it is simple to compose a CRM design that will not skip doses and will stop once a recommended dose has been given to $n$ patients; or a BOIN design that will demand that a certain dose is tested sufficiently before stopping is countenanced; and so on.
Functions are provided to run simulations and enumerate all possible future dose paths.
See related blog posts or the package documentation site for many illustrative examples.
- CRM Simulation Checklist
- How to design a dose-finding study using the continual reassessment method
- Combination Lenalidomide and Azacitidine - A Novel Salvage Therapy in Patients Who Relapse After Allogeneic Stem-Cell Transplantation for Acute Myeloid Leukemia
- Implementing the EffTox dose-finding design in the Matchpoint trial