# Dose-escalation designs (3+3, BOIN, dose-expansion)

Source: https://onco.cc/terms/dose-escalation-design/  
OnCo record `dose-escalation-design` (Term). Data CC BY-NC 4.0, attribute "Data from OnCo (onco.cc)"; commercial use needs a licence.

## TL;DR

How a phase 1 trial climbs from a tiny starting dose to a useful one: the old 3+3 design treats three patients at a time and moves up if none has serious toxicity; newer statistical designs (BOIN, CRM) use all the data to pick doses more accurately with fewer patients on ineffective levels.

## Summary

The 3+3 design (still the most used) is simple but imprecise, treats a large share of patients at sub-therapeutic doses and identifies the MTD poorly; model-based designs such as the continual reassessment method and the Bayesian optimal interval design estimate the toxicity curve continuously and are recommended by the FDA and methodologists, with time-to-event versions handling late toxicities. After escalation, dose-expansion cohorts (often 20-40 patients per tumour type) gather efficacy signals and may be randomised between two doses under Project Optimus. Accelerated titration and single-patient cohorts speed the early low-dose levels, and backfill cohorts add patients at lower doses to inform dose optimisation.

## Fields

- Kind: Term
- Last checked: 2026-09-09
- Also known as: 3+3; 3 + 3; 3+3 design; rule-based design; BOIN; Bayesian optimal interval; CRM; continual reassessment method; mTPI; TITE-CRM; model-based dose finding; model-based design; dose-finding; dose finding; dose-finding study; dose escalation cohort; dose escalation; dose-escalation; dose escalation phase; dose-expansion; dose expansion; expansion phase; accelerated titration; backfill cohort; dose level; dose levels; cohort of 3; backfill cohorts; model-based designs

## Sources

- Wikipedia: https://en.wikipedia.org/wiki/Phase_I_clinical_trial#Dose_escalation
- Wikipedia: https://en.wikipedia.org/wiki/Phase_I_clinical_trial#Dose_escalation

## Connected records

- terms: [Bayesian trial design](https://onco.cc/terms/bayesian-trial-design/), [Dose-limiting toxicity (DLT)](https://onco.cc/terms/dose-limiting-toxicity/), [First-in-human (FIH) trial](https://onco.cc/terms/first-in-human/), [Maximum tolerated dose (MTD)](https://onco.cc/terms/mtd/), [Phase 1, 2 and 3 trials](https://onco.cc/terms/trial-phases/), [Project Optimus](https://onco.cc/terms/project-optimus/), [Recommended phase 2 dose (RP2D)](https://onco.cc/terms/rp2d/), [Seamless, adaptive and Bayesian trial designs](https://onco.cc/terms/seamless-adaptive/)
- fronts: [Drug Discovery Platforms](https://onco.cc/fronts/drug-discovery/)
- trials: [Low-dose nivolumab plus metronomic chemotherapy (Tata Memorial)](https://onco.cc/trials/low-dose-nivolumab-tmh/), [REJOICE-Ovarian01](https://onco.cc/trials/rejoice-ovarian01/)
- key papers: [Reverse swing-M, phase 1 study of repurposing mebendazole in recurrent high-grade glioma](https://onco.cc/key-papers/paper-patil-mebendazole-glioma-phase-1-cancer-med-2020/)

---
JSON: https://onco.cc/api/v1/entities/dose-escalation-design.json