Skip to contentScholarGate
LibraryBookshelfDeskReview StudioAssistant
Sign in
On this page
IntuitionHow it worksWhen to use itStrengths & limitationsCommon pitfallsApplicationsFrequently asked🔒 Read the full methodSourcesRelated methods
Cite this pageSpotted an issue on this page? Report or suggest a fix →
Home›Experimental design›Dose-Escalation Design (Continual Reassessment Method)
Process / pipelineClinical trial design

Dose-Escalation Design (Continual Reassessment Method)

Also known as: Continual Reassessment Method, CRM Design, Phase I Dose-Finding Design, Doz Artırma Tasarımı

Dose-Escalation Design, formalized as the Continual Reassessment Method (CRM), is a Bayesian adaptive algorithm for identifying the Maximum Tolerated Dose (MTD) in Phase I clinical trials. Introduced by John O'Quigley, Margaret Pepe, and Lloyd Fisher in 1990, CRM treats dose-toxicity response as a parametric curve, updates a prior probability model after each patient's outcome, and assigns subsequent patients to the dose currently estimated closest to a pre-specified target toxicity probability.

ScholarGate
  1. Process / pipeline
  2. v1
  3. 1 Sources
  4. PUBLISHED
Cite this page →
Tools & resources
Download slides
Learn & explore

Read the full method

Members only

Sign in with a free account to read this section.

Sign in

Method map

The neighbourhood of related methods — select a node to explore.

Dose-Escalation Design
Bayesian InferenceSequential Design

When to use it

CRM is appropriate for Phase I oncology or pharmacology trials where the primary goal is to estimate the MTD — the highest dose associated with an acceptable toxicity rate. It requires binary toxicity outcomes, a pre-specified target toxicity probability, a parametric dose-toxicity skeleton, and sufficient sample size (typically 20–40 patients) for the posterior to converge. It is less suitable when toxicity endpoints are delayed, ordinal without clear binary collapse, or when regulatory or institutional constraints mandate rule-based designs such as 3+3. Alternatives include keyboard design, BOIN, and mTPI.

Strengths & limitations

Strengths
  • Efficiently allocates more patients near the true MTD, reducing unnecessary exposure to overtly toxic or subtherapeutic doses.
  • Formally incorporates prior clinical knowledge through the Bayesian skeleton, making it coherent and updatable.
  • Consistently outperforms the traditional 3+3 rule-based design in MTD accuracy across simulation studies.
  • Flexible enough to accommodate different target toxicity rates, model skeletons, and cohort sizes.
Limitations
  • Requires up-front specification of a dose-toxicity skeleton; misspecification of the skeleton can bias MTD estimates.
  • Implementation demands statistical software and real-time computation, which increases operational complexity compared to rule-based designs.
  • The original one-parameter model may lack flexibility when the true dose-toxicity curve deviates substantially from the assumed parametric form.
  • Regulatory agencies and ethics boards may be less familiar with CRM, occasionally creating barriers to approval or institutional resistance.

Frequently asked

How does CRM differ from the traditional 3+3 design?

The 3+3 design uses fixed, pre-specified escalation rules and stops as soon as a toxicity threshold is crossed, making no use of the full data history. CRM continuously updates a probabilistic model of the dose-toxicity curve using all observed outcomes and assigns each new patient to the currently best-estimated MTD, resulting in more accurate MTD identification with the same or smaller sample sizes.

What is the 'skeleton' in CRM and why does it matter?

The skeleton is the set of initial toxicity probability estimates assigned to each dose level before the trial begins, forming the parametric shape of the dose-toxicity curve. A well-chosen skeleton that brackets the true MTD allows CRM to converge quickly; a poorly specified skeleton that places the true MTD outside the assumed range can lead to systematic bias in the final MTD recommendation.

Is CRM safe to use in practice given its Bayesian complexity?

Yes, when properly implemented. Regulatory guidance (FDA, EMA) now acknowledges model-based adaptive designs including CRM. Pre-trial simulation studies are mandatory to evaluate the design's operating characteristics — particularly the probability of recommending the correct MTD and the rate of overdose — under realistic scenarios before ethics approval and trial initiation.

Sources

  1. O'Quigley, J., Pepe, M., & Fisher, L. (1990). Continual reassessment method: a practical design for phase 1 clinical trials in cancer. Biometrics, 46(1), 33–48. DOI: 10.2307/2531628 ↗

How to cite this page

ScholarGate. (2026, June 2). Dose-Escalation Design (Continual Reassessment Method). ScholarGate. https://scholargate.app/en/experimental-design/dose-escalation-design

Related methods

Bayesian InferenceSequential Design

Which method?

Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.

  • Bayesian InferenceStatistics↔ compare
  • Sequential DesignExperimental design↔ compare
Compare side by side →

Similar methods

Bayesian Phase I clinical trialAdaptive Phase I Clinical TrialRisk-adjusted Phase I clinical trialMeta-analytic Phase I clinical trialBayesian Phase II Clinical TrialPhase I Clinical TrialAdaptive Phase II Clinical TrialAdaptive Dose-Response Analysis

Related reference concepts

Bayesian Forecasting in Personalized DosingCombination Regimens and Drug InteractionsPrecision Dosing and Therapeutic Drug MonitoringClinical Trial Design and InterpretationDose-Response Relationships and Therapeutic WindowTherapeutic Drug Monitoring and Individualized Dosing

Spotted an issue on this page? Report or suggest a fix →

ScholarGate — Dose-Escalation Design (Dose-Escalation Design (Continual Reassessment Method)). Retrieved 2026-07-21 from https://scholargate.app/en/experimental-design/dose-escalation-design · Dataset: https://doi.org/10.5281/zenodo.20539026
Quick facts
Originator
John O'Quigley, Margaret Pepe & Lloyd Fisher
Year
1990
Type
Adaptive Bayesian dose-finding design
Subfamily
Clinical trial design
Target Parameter
Maximum Tolerated Dose (MTD)
Trial Phase
Phase I
Related methods
Bayesian InferenceSequential Design
ScholarGate

A content-first reference library for research methods — what each one is, how it works, and where it comes from.

Open data (CC-BY)

Explore

  • Library
  • Search the library…
  • Browse by field
  • Fields
  • Journey
  • Compare
  • Which method?

Reference

  • Subjects
  • Atlas
  • Glossary
  • Methodology
  • Philosophy

Your tools

  • Bookshelf
  • Desk
  • Chat

Company

  • About
  • Pricing
  • Contact
  • Suggest a method

Entries are compiled from published sources for reference. Verifying the accuracy and suitability of any information for your own use remains your responsibility.

© 2026 ScholarGate · A research-method reference library
  • Privacy
  • Cookies
  • Terms
  • Delete account