ordcrm: Likelihood-Based Continual Reassessment Method (CRM) Dose Finding Designs

Provides the setup and calculations needed to run a likelihood-based continual reassessment method (CRM) dose finding trial and performs simulations to assess design performance under various scenarios. 3 dose finding designs are included in this package: ordinal proportional odds model (POM) CRM, ordinal continuation ratio (CR) model CRM, and the binary 2-parameter logistic model CRM. These functions allow customization of design characteristics to vary sample size, cohort sizes, target dose-limiting toxicity (DLT) rates, discrete or continuous dose levels, combining ordinal grades 0 and 1 into one category, and incorporate safety and/or stopping rules. For POM and CR model designs, ordinal toxicity grades are specified by common terminology criteria for adverse events (CTCAE) version 4.0. Function 'pseudodata' creates the necessary starting models for these 3 designs, and function 'nextdose' estimates the next dose to test in a cohort of patients for a target DLT rate. We also provide the function 'crmsimulations' to assess the performance of these 3 dose finding designs under various scenarios.

Getting started

Package details

AuthorEmily V. Dressler, PhD Zhengyan Huang
MaintainerEmily V. Dressler <EmilyVDressler@gmail.com>
LicenseGPL (>= 2)
Package repositoryView on CRAN
Installation Install the latest version of this package by entering the following in R:

Try the ordcrm package in your browser

Any scripts or data that you put into this service are public.

ordcrm documentation built on May 2, 2019, 1:10 p.m.