In clinical trials, response-adaptive randomization (RAR) has gained increasing attention due to its ability to assign more patients to better-performing treatments. Consequently, several RAR methods have been proposed in recent years. Among them, the efficient response adaptive randomization design (ERADE), proposed by Hu et al. (2009), stands out as an optimal approach, with the asymptotic variance of the allocation proportion achieving the Cramér–Rao lower bound, demonstrating its statistical efficiency. However, the original ERADE is limited to trials with only two treatment arms. Given the growing prevalence of multi-arm trials in modern clinical development, the original ERADE design no longer meets all practical needs. In this paper, we extend ERADE for use in multi-arm clinical trials, proposing the multi-arm ERADE algorithm. We establish the asymptotic properties of this generalized design and demonstrate its effectiveness in finite sample settings through simulations and a real-world trial redesign.
ThompsonWR. On the likelihood that one unknown probability exceeds another in view of the evidence of two samples. Biometrika1933; 25: 285–294.
2.
RosenbergerWFLachinJM. Randomization in clinical trials: theory and practice. Wiley series in probability and statistics; 2016.
3.
ZelenM. Play the winner rule and the controlled clinical trial. J Am Stat Assoc1969; 64: 131–146.
4.
WeiLJ. The generalized Polya’s urn design for sequential medical trials. Ann Stat1979; 7: 291–296.
5.
WeiL. The generalized Polya’s urn design for sequential medical trials. Ann Stat1979; 7: 291–296.
6.
AndersenJFariesDRamuraR. A randomized play-the-winner design for multi-arm clinical trials. Commun Stat-Theory Methods1994; 23: 309–323.
7.
IvanovaARosenbergerWFDurhamSD, et al.A birth and death urn for randomized clinical trials: asymptotic methods. Sankhyā: Indian J Stat Ser B2000; 62: 104–118.
8.
IvanovaAFlournoyN. A birth and death urn for ternary outcomes: stochastic processes applied to urn models. In Jones MC, Charalambides CA, Koutras MV (eds) Probability and statistical models with applications. Chapman and Hall/CRC, 2001, pp. 583–600.
9.
IvanovaA. A play-the-winner-type urn design with reduced variability. Metrika2003; 58: 1–13.
10.
HuFZhangLX. Asymptotic normality of urn models for clinical trials with delayed response. Bernoulli2004; 10: 447–463.
11.
SunRCheungSHZhangLX. A generalized drop-the-loser rule for multi-treatment clinical trials. J Stat Plan Inference2007; 137: 2011–2023.
12.
CrimaldiILeisenF. Asymptotic results for a generalized polya urn with “multi-updating” and applications to clinical trials. Commun Stat—Theory Methods2008; 37: 2777–2794.
13.
BandyopadhyayUBiswasABhattacharyaR. Drop-the-loser design in the presence of covariates. Metrika2009; 69: 1–15.
14.
ZhuHHuF. Interim analysis of clinical trials based on urn models. Can J Stat2012; 40: 550–568.
15.
EiseleJR. The doubly adaptive biased coin design for sequential clinical trials. J Stat Plan Inference1994; 38: 249–261.
16.
RosenbergerWFStallardNIvanovaA, et al.Optimal adaptive designs for binary response trials. Biometrics2001; 57: 909–913.
17.
JennisonCTurnbullBW. Group sequential tests with outcome-dependent treatment assignment. Seq Anal2001; 20: 209–234.
18.
HuFZhangLX. Asymptotic properties of doubly adaptive biased coin designs for multitreatment clinical trials. Ann Stat2004; 32: 268–301.
19.
ZhaiGLiYZhangL, et al.Group response-adaptive randomization with delayed and missing responses. Stat Med2024; 43: 5047–5059.
20.
VillarSSBowdenJWasonJ. Revisiting the multi-armed bandit model for the optimal design of clinical trials: benefits and drawbacks. Trials2013; 14: P36.
SverdlovOTymofyeyevYWongWK. Optimal response-adaptive randomized designs for multi-armed survival trials. Stat Med2011; 30: 2890–2910.
23.
WilliamsonSFJackoPJakiT. Generalisations of a Bayesian decision-theoretic randomisation procedure and the impact of delayed responses. Comput Stat Data Anal2022; 174: 107–407.
24.
GiovagnoliAVerdinelliI. Bayesian adaptive randomization with compound utility functions. Stat Sci2023; 38: 52–67.
25.
BaiZHuFShenL. An adaptive design for multi-arm clinical trials. J Multivar Anal2002; 81: 1–18.
26.
VillarSSWasonJBowdenJ. Response-adaptive randomization for multi-arm clinical trials using the forward looking gittins index rule. Biometrics2015; 71: 969–978.
27.
SmithALVillarSS. Bayesian adaptive bandit-based designs using the gittins index for multi-armed trials with normally distributed endpoints. J Appl Stat2018; 45: 1052–1076.
28.
WilliamsonSFVillarSS. A response-adaptive randomization procedure for multi-armed clinical trials with normally distributed outcomes. Biometrics2020; 76: 197–209.
29.
ZhuHHuF. Implementing optimal allocation for sequential continuous responses with multiple treatments. J Stat Plan Inference2009; 139: 2420–2430.
30.
WasonJMTrippaL. A comparison of Bayesian adaptive randomization and multi-stage designs for multi-arm clinical trials. Stat Med2014; 33: 2206–2221.
31.
HuFZhangLXHeX. Efficient randomized-adaptive designs. Ann Stat2009; 37: 2543–2560.
32.
BandyopadhyayUBiswasA. Adaptive designs for normal responses with prognostic factors. Biometrika2001; 88: 409–419.
33.
HuFRosenbergerWF. Optimality, variability, power. J Am Stat Assoc2003; 98: 671–678.
34.
ZhangLRosenbergerWF. Response-adaptive randomization for clinical trials with continuous outcomes. Biometrics2006; 62: 562–569.
35.
EfronB. Forcing a sequential experiment to be balanced. Biometrika1971; 58: 403–417.
36.
FlehingerBJLouisTA. Sequential treatment allocation in clinical trials. Biometrika1971; 58: 419–426.
37.
SmithRL. Properties of biased coin designs in sequential clinical trials. Ann Stat1984; 12: 1018–1034.
38.
JennisonCTurnbullBW. Group sequential methods with applications to clinical trials. 2nd ed.Boca Raton, FL: Chapman & Hall/CRC, 2000.
39.
Baldi AntogniniAGiovagnoliA. On the large sample optimality of sequential designs for comparing two or more treatments. Seq Anal2005; 24: 205–217.
40.
GwiseTEHuFHuJ. Optimal biased coins for two-arm clinical trials. Stat Interface2008; 1: 125–135.
41.
RosenbergerWFHuF. Maximizing power and minimizing treatment failures in clinical trials. Clin Trials2004; 1: 141–147.
42.
GoadsbyPJReuterUHallströmY, et al.A controlled trial of erenumab for episodic migraine. N Engl J Med2017; 377: 2123–2132.
43.
XuWHuFCheungSH. Adaptive designs for non-inferiority trials with multiple experimental treatments. Stat Methods Med Res2018; 27: 3255–3270.
44.
GaoJHuFCheungSH, et al.Response-adaptive treatment randomization for multiple comparisons of treatments with recurrent event responses. Stat Methods Med Res2022; 31: 1549–1565.
45.
MaWQinYLiY, et al.Statistical inference for covariate-adaptive randomization procedures. J Am Stat Assoc2020; 115: 1488–1497.
46.
LiYMaWQinY, et al.Testing for treatment effect in covariate-adaptive randomized trials with generalized linear models and omitted covariates. Stat Methods Med Res2021; 30: 2148–2164.
YangHQinYWangF, et al.. Balancing covariates in multi-arm trials via adaptive randomization. Comput Stat Data Anal2023; 179: 107642.
49.
ZhaoZSongYJiangW, et al.Consistent covariances estimation for stratum imbalances under minimization method for covariate-adaptive randomization. Scand J Stat2024; 51: 861–890.
50.
MaWLiPZhangLX, et al.A new and unified family of covariate adaptive randomization procedures and their properties. J Am Stat Assoc2024; 119: 151–162.
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
