TY - JOUR

T1 - Principles of dose finding studies in cancer

T2 - a comparison of trial designs

AU - Jaki, Thomas

AU - Clive, Sally

AU - Weir, Christopher

PY - 2013/5

Y1 - 2013/5

N2 - PurposeOne key aim of Phase I cancer studies is to identify the dose of a treatment to be further evaluated in Phase II. We describe, in non-statistical language, three classes of dose-escalation trial design and compare their properties.MethodsWe review three classes of dose-escalation design suitable for Phase I cancer trials: algorithmic approaches (including the popular 3 + 3 design), Bayesian model-based designs and Bayesian curve-free methods. We describe an example from each class and summarize the advantages and disadvantages of the design classes.ResultsThe main benefit of algorithmic approaches is the simplicity with which they may be communicated: it may be for this reason alone that they are still employed in the vast majority of Phase I trials. Model-based and curve-free Bayesian approaches are preferable to algorithmic methods due to their superior ability to identify the dose with the desired toxicity rate and their allocation of a greater proportion of patients to doses at, or close to, that dose.ConclusionsFor statistical and practical reasons, algorithmic methods cannot be recommended. The choice between a Bayesian model-based or curve-free approach depends on the previous information available about the compound under investigation. If this provides assurance about a particular model form, the model-based approach would be appropriate; if not, the curve-free method would be preferable.

AB - PurposeOne key aim of Phase I cancer studies is to identify the dose of a treatment to be further evaluated in Phase II. We describe, in non-statistical language, three classes of dose-escalation trial design and compare their properties.MethodsWe review three classes of dose-escalation design suitable for Phase I cancer trials: algorithmic approaches (including the popular 3 + 3 design), Bayesian model-based designs and Bayesian curve-free methods. We describe an example from each class and summarize the advantages and disadvantages of the design classes.ResultsThe main benefit of algorithmic approaches is the simplicity with which they may be communicated: it may be for this reason alone that they are still employed in the vast majority of Phase I trials. Model-based and curve-free Bayesian approaches are preferable to algorithmic methods due to their superior ability to identify the dose with the desired toxicity rate and their allocation of a greater proportion of patients to doses at, or close to, that dose.ConclusionsFor statistical and practical reasons, algorithmic methods cannot be recommended. The choice between a Bayesian model-based or curve-free approach depends on the previous information available about the compound under investigation. If this provides assurance about a particular model form, the model-based approach would be appropriate; if not, the curve-free method would be preferable.

KW - 3 + 3 design

KW - Bayesian method

KW - Clinical trial, Phase I

KW - Continual reassessment method

KW - CRM

KW - Curve free

U2 - 10.1007/s00280-012-2059-8

DO - 10.1007/s00280-012-2059-8

M3 - Journal article

VL - 71

SP - 1107

EP - 1114

JO - Cancer Chemotherapy and Pharmacology

JF - Cancer Chemotherapy and Pharmacology

SN - 0344-5704

IS - 5

ER -