Response-adaptive randomization (RAR) dynamically adjusts allocation probabilities of sequentially enrolled patients based on accumulating response information. It has gained increasing attention in clinical trials for its ability to enhance statistical efficiency by accelerating estimation of treatment effects, improve ethical allocation by favoring superior interventions, and maintain robust parameter estimation (e.g. via variance or uncertainty reduction) while accounting for adaptive modifications. While both regulatory agencies and the broader clinical research community are making efforts to promote the application of RAR, its real-world implementation remains limited. In this article, we review the application of RAR in clinical practice from 2015 to 2024, identifying key challenges such as managing diverse patient outcome types, incorporating repeated measurements, and addressing missing data. We categorize various RAR methods and propose practical solutions to these challenges, providing insights and practical references for clinical trial practitioners. Additionally, we discuss the limitations of existing RAR methods and outline potential future research directions. Our review aims to bridge the gap between theory and practice, promoting the broader adoption of RAR in clinical trials while advancing the development of RAR theory and methodology.
BurnettTMozgunovPPallmannP, et al.Adding flexibility to clinical trial designs: an example-based guide to the practical use of adaptive designs. BMC Med2020; 18: 1–21.
7.
GilholmPErgetuEGelbartB, et al.Adaptive clinical trials in pediatric critical care: a systematic review. Pediatr Crit Care Med2023; 24: 738–749.
PallmannPBeddingAWChoodari-OskooeiB, et al.Adaptive designs in clinical trials: why use them, and how to run and report them. BMC Med2018; 16: 1–15.
10.
HuFRosenbergerWF. The theory of response-adaptive randomization in clinical trials. Hoboken, NJ: John Wiley & Sons, 2006.
11.
ThompsonWR. On the likelihood that one unknown probability exceeds another in view of the evidence of two samples. Biometrika1933; 25: 285–294.
12.
GuidanceFD. Adaptive design clinical trials for drugs and biologics. Biotechnol Law Rep2010; 29: 173.
13.
PriceDScottJ. The US Food and Drug Administration’s complex innovative trial design pilot meeting program: progress to date. Clin Trials2021; 18: 706–710.
14.
BjerrumOJLindenHH. European new safe and innovative medicines initiatives: history and progress (through December 2009). In: Translational medicine and drug discovery. Cambridge University Press, 2011, pp.265–288.
15.
RobertsonDSLeeKMLópez-KolkovskaBC, et al.Response-adaptive randomization in clinical trials: from myths to practical considerations. Stat Sci2023; 38: 185.
16.
BothwellLEAvornJKhanNF, et al.Adaptive design clinical trials: a review of the literature and ClinicalTrials.gov. Br Med J Open2018; 8: e018320.
17.
RobertsonDSChoodari-OskooeiBDimairoM, et al.Point estimation for adaptive trial designs I: a methodological review. Stat Med2023; 42: 122–145.
18.
RobertsonDSChoodari-OskooeiBDimairoM, et al.Point estimation for adaptive trial designs II: practical considerations and guidance. Stat Med2023; 42: 2496–2520.
19.
GuglielmettiLArdizzoniEAtgerM, et al.Evaluating newly approved drugs for multidrug-resistant tuberculosis (endTB): study protocol for an adaptive, multi-country randomized controlled trial. Trials2021; 22: 651.
20.
DeliuNDasRMayA, et al.StratosPHere 2: study protocol for a response-adaptive randomised placebo-controlled phase II trial to evaluate hydroxychloroquine and phenylbutyrate in pulmonary arterial hypertension caused by mutations in BMPR2. Trials2024; 25: 680.
21.
KimESHerbstRSWistubaII, et al.The BATTLE trial: personalizing therapy for lung cancer. Cancer Discov2011; 1: 44–53.
22.
BarkerASigmanCKelloffG, et al.I-SPY 2: an adaptive breast cancer trial design in the setting of neoadjuvant chemotherapy. Clin Pharmacol Ther2009; 86: 97–100.
23.
DasRDeliuNToshnerM, et al.Implementing response-adaptive randomisation in stratified rare-disease trials: design challenges and practical solutions. arXiv:241003346, 2024.
24.
HudgensSHowerterAPolekE, et al.Psychometric validation and interpretation of the Nocturia Impact Diary in a clinical trial setting. Qual Life Res2022; 31: 1837–1848.
25.
SimonJAAndersonRABallantyneE, et al.Efficacy and safety of elinzanetant, a selective neurokinin-1,3 receptor antagonist for vasomotor symptoms: a dose-finding clinical trial (SWITCH-1). Menopause2023; 30: 239–246.
26.
HuckvaleKHoonLStechE, et al.Protocol for a bandit-based response adaptive trial to evaluate the effectiveness of brief self-guided digital interventions for reducing psychological distress in university students: the Vibe Up study. Br Med J Open2023; 13: e066249.
27.
VillarSSBowdenJWasonJ. Response-adaptive designs for binary responses: how to offer patient benefit while being robust to time trends?Pharm Stat2018; 17: 182–197.
ZelenM. Play the winner rule and the controlled clinical trial. J Am Stat Assoc1969; 64: 131–146.
30.
WeiLDurhamS. The randomized play-the-winner rule in medical trials. J Am Stat Assoc1978; 73: 840–843.
31.
IvanovaAHobermanS. Higher order response-adaptive urn designs for clinical trials with highly successful treatments. J R Stat Soc C: Appl Stat2015; 64: 175–189.
32.
IvanovaA. A play-the-winner-type urn design with reduced variability. Metrika2003; 58: 1–13.
33.
SutradharBCBiswasABariW. Marginal regression for binary longitudinal data in adaptive clinical trials. Scand J Stat2005; 32: 93–113.
34.
SutradharBCJowaheerV. Analyzing longitudinal count data from adaptive clinical trials: a weighted generalized quasi-likelihood approach. J Stat Comput Simul2006; 76: 1079–1093.
35.
ZhangLXHuFCheungSH. Asymptotic theorems of sequential estimation-adjusted urn models. Ann Appl Probab2006; 16: 340–369.
36.
YuJLaiD. Interim analysis of sequential estimation-adjusted urn models with sample size re-estimation. Can J Stat2021; 49: 1075–1092.
37.
AlettiGGhigliettiARosenbergerWF. Nonparametric covariate-adjusted response-adaptive design based on a functional urn model. Ann Stat2018; 46: 3838–3866.
38.
MulierePPaganoniAMSecchiP. A randomly reinforced urn. J Stat Plan Inference2006; 136: 1853–1874.
39.
CrimaldiILouisPYMinelliIG. An urn model with random multiple drawing and random addition. Stoch Process Their Appl2022; 147: 270–299.
GhigliettiAScaraleMGMiceliR, et al.Urn models for response-adaptive randomized designs: a simulation study based on a non-adaptive randomized trial. J Biopharm Stat2018; 28: 1203–1215.
42.
GhigliettiAPaganoniAM. An urn model to construct an efficient test procedure for response adaptive designs. Stat Methods Appt2016; 25: 211–226.
43.
KadaneJB. Bayesian methods and ethics in a clinical trial design. New York, NY: J. Wiley & Sons, 1996.
KarrisonTGHuoDChappellR. A group sequential, response-adaptive design for randomized clinical trials. Control Clin Trials2003; 24: 506–522.
48.
RosenbergerWFStallardNIvanovaA, et al.Optimal adaptive designs for binary response trials. Biometrics2001; 57: 909–913.
49.
JennisonCTurnbullBW. Group sequential methods with applications to clinical trials. Boca Raton, FL: CRC Press, 1999.
50.
KomakiFBiswasA. Bayesian optimal response-adaptive design for binary responses using stopping rule. Stat Methods Med Res2018; 27: 891–904.
51.
BiswasABhattacharyaR. A covariate-adjusted response-adaptive allocation for a general class of continuous responses. J Stat Theory Pract2016; 10: 852–863.
52.
BiswasADuttaSLahaAK, et al.Response-adaptive allocation for circular data. J Biopharm Stat2015; 25: 830–842.
53.
ZhuJJazićILiuY. A novel framework for Bayesian response-adaptive randomization. In: Pharmaceutical statistics Workshop. Springer International Publishing, 2016, pp.141–155.
54.
FengCHuF. Optimal responses-adaptive designs based on efficiency, ethic, and cost. Stat Interface2018; 11: 99–107.
55.
BandyopadhyayUBhattacharyaR. An optimal three treatment allocation for binary treatment responses. Stat Biopharm Res2018; 10: 287–300.
56.
EiseleJR. The doubly adaptive biased coin design for sequential clinical trials. J Stat Plan Inference1994; 38: 249–261.
57.
EiseleJRWoodroofeMB. Central limit theorems for doubly adaptive biased coin designs. Ann Stat1995; 23: 234–254.
58.
HuFZhangLX. Asymptotic properties of doubly adaptive biased coin designs for multitreatment clinical trials. Ann Stat2004; 32: 268–301.
59.
WangYZhuHLeeJJ. Evaluation of bias for outcome adaptive randomization designs with binary endpoints. Stat Interface2020; 13: 287–315.
60.
YeXHuFMaW. Robustness of response-adaptive randomization. Biometrics2024; 80: ujae049.
61.
YuYXuCZhongJ, et al.Comparison of treatments with ordinal responses in trials with sequential monitoring and response-adaptive randomization. Stat Med2022; 41: 5061–5083.
62.
CaoWZhuHWangL, et al.Doubly adaptive biased coin design to improve Bayesian clinical trials with time-to-event endpoints. Stat Med2024; 43: 1743–1758.
63.
MukherjeeACoadDSJanaS. Covariate-adjusted response-adaptive designs for censored survival responses. J Stat Plan Inference2023; 225: 219–242.
64.
SuPFCheungSH. Response-adaptive treatment allocation for survival trials with clustered right-censored data. Stat Med2018; 37: 2427–2439.
65.
GaoJSuPFHuF, et al.Adaptive treatment allocation for comparative clinical studies with recurrent events data. Biometrics2020; 76: 183–196.
66.
SuPF. Response-adaptive treatment allocation for clinical studies with recurrent event and terminal event data. Stat Med2022; 41: 258–275.
67.
MauYLSuPF. Evaluating response-adaptive randomization procedures for recurrent events and terminal event data using a composite endpoint. Pharm Stat2022; 21: 1167–1184.
BerrySMCarlinBPLeeJJ, et al.Bayesian adaptive methods for clinical trials. Boca Raton, FL: CRC Press, 2010.
70.
ThallPFoxPWathenJ. Statistical controversies in clinical research: scientific and ethical problems with adaptive randomization in comparative clinical trials. Ann Oncol2015; 26: 1621–1628.
71.
GiovagnoliAVerdinelliI. Bayesian adaptive randomization with compound utility functions. Stat Sci2023; 38: 52–67.
72.
BeallJElmJChamberlainJ, et al.An expected score approach to ordinal outcomes in a Bayesian, response adaptive, randomized trial. Stat Biopharm Res2023; 15: 820–825.
MerrellDChanderengTParkY. A Markov decision process for response-adaptive randomization in clinical trials. Comput Stat Data Anal2023; 178: 107599.
75.
VillarSSBowdenJWasonJ. Multi-armed bandit models for the optimal design of clinical trials: benefits and challenges. Stat Sci2015; 30: 199.
76.
VillarSSWasonJBowdenJ. Response-adaptive randomization for multi-arm clinical trials using the forward looking Gittins index rule. Biometrics2015; 71: 969–978.
77.
SmithALVillarSS. Bayesian adaptive bandit-based designs using the Gittins index for multi-armed trials with normally distributed endpoints. J Appl Stat2018; 45: 1052–1076.
78.
WilliamsonSFVillarSS. A response-adaptive randomization procedure for multi-armed clinical trials with normally distributed outcomes. Biometrics2019; 76: 197–209.
79.
AhujaVBirgeJR. Response-adaptive designs for clinical trials: simultaneous learning from multiple patients. Eur J Oper Res2016; 248: 619–633.
80.
AtkinsonACBiswasA. Bayesian adaptive biased-coin designs for clinical trials with normal responses. Biometrics2005; 61: 118–125.
81.
SverdlovORosenbergerWFRyeznikY. Utility of covariate-adjusted response-adaptive randomization in survival trials. Stat Biopharm Res2013; 5: 38–53.
82.
VillarSSBowdenJWasonJ. Multi-armed bandit models for the optimal design of clinical trials: benefits and challenges. Stat Sci2015; 30: 199–215.
83.
RussoDJVan RoyBKazerouniA, et al.A tutorial on Thompson sampling. Found Trends Mach Learn2018; 11: 1–96.
84.
LaterrePFBerrySMBlemingsA, et al.Effect of selepressin vs placebo on ventilator- and vasopressor-free days in patients with septic shock: The SEPSIS-ACT randomized clinical trial. J Am Med Assoc2019; 322: 1476.
85.
AppelLJMichosEDMitchellCM, et al.The effects of four doses of vitamin D supplements on falls in older adults: a response-adaptive, randomized clinical trial. Ann Intern Med2021; 174: 145–156.
86.
ShortNJKantarjianHMLoghaviS, et al.Treatment with a 5-day versus a 10-day schedule of decitabine in older patients with newly diagnosed acute myeloid leukaemia: a randomised phase 2 trial. Lancet Haematol2019; 6: e29–e37.
ZhuHZhuH. Covariate-adjusted response-adaptive designs based on semiparametric approaches. Biometrics2023; 79: 2895–2906.
89.
LiuHLinXHuangX. An oncology clinical trial design with randomization adaptive to both short-and long-term responses. Stat Methods Med Res2019; 28: 2015–2031.
90.
AndererABastaniHSilberholzJ. Adaptive clinical trial designs with surrogates: when should we bother?Manage Sci2022; 68: 1982–2002.
91.
WangH. Response adaptive randomization using surrogate and primary endpoints. PhD Thesis, Virginia Commonwealth University, 2016.
92.
XuZMaTTangL, et al.Bayesian response adaptive randomization design with a composite endpoint of mortality and morbidity. Stat Med2024; 43: 1256–1270.
93.
GranholmALangeTHarhayMO, et al.Effects of sceptical priors on the performance of adaptive clinical trials with binary outcomes. Pharm Stat2024; 23: 728–741.
94.
DasS. A two armed optimal response adaptive randomization for ordinal categorical responses with possible misclassification. Seq Anal2024.
95.
DasSBhattacharyaRBiswasA. An optimal response adaptive design for multi-treatment clinical trials with ordinal categorical outcomes. J Biopharm Stat2021; 31: 809–827.
96.
HuangTLiuZHuF. Longitudinal covariate-adjusted response-adaptive randomized designs. J Stat Plan Inference2013; 143: 1816–1827.
97.
LiangYCarriereKC. Allocating multiple patients at a time in multiple-objective response-adaptive repeated measurement designs. Int J Stat Probab2012; 1: 186.
BiswasADewanjiA. A randomized longitudinal play-the-winner design for repeated binary data. Aust N Z J Stat2004; 46: 675–684.
101.
BiswasADewanjiA. Inference for a RPW-type clinical trial with repeated monitoring for the treatment of rheumatoid arthritis. Biom J2004; 46: 769–779.
102.
AtkinsonACBiswasA. Optimal response and covariate-adaptive biased-coin designs for clinical trials with continuous multivariate or longitudinal responses. Comput Stat Data Anal2017; 113: 297–310.
103.
BiswasAParkEBhattacharyaR. Covariate-adjusted response-adaptive designs for longitudinal treatment responses: PEMF trial revisited. Stat Methods Med Res2012; 21: 379–392.
KimSHCarriereKC. Practical and optimal crossover designs for clinical trials. In: Recent advances in medical statistics (ed C Vargas-De-León), chapter 8. Rijeka: IntechOpen, 2022.
BandyopadhyayUMukherjeeSBiswasA. Adaptive two-treatment three-period crossover design for normal responses. Braz J Probab Stat2020; 34: 291–303.
108.
BandyopadhyayUBiswasAMukherjeeS. A response-adaptive design in crossover trial. Statistics2012; 46: 645–661.
109.
BandyopadhyayUBiswasAMukherjeeS. Adaptive two-treatment two-period crossover design for binary treatment responses. Stat Neerl2007; 61: 329–344.
110.
LiangYCarrièreKC. Multiple-objective response-adaptive repeated measurement designs for clinical trials. J Stat Plan Inference2009; 139: 1134–1145.
111.
LiangYLiYWangJ, et al.Multiple-objective response-adaptive repeated measurement designs in clinical trials for binary responses. Stat Med2014; 33: 607–617.
112.
BandyopadhyayUMukherjeeS. Adaptive crossover design for normal responses. Commun Stat - Theory Methods2015; 44: 1466–1482.
113.
FountzilasETsimberidouAMVoHH, et al.Clinical trial design in the era of precision medicine. Genome Med2022; 14: 101.
114.
LiuXDeliuNChakrabortyT, et al.Thompson sampling for zero-inflated count outcomes with an application to the drink less mobile health study. Ann Appl Stat2025; 19: 1403–1425.
115.
TomkinsSLiaoPKlasnjaP, et al.Intelligentpooling: practical Thompson sampling for mhealth. Mach Learn2021; 110: 2685–2727.
116.
DuanLHuF. Doubly adaptive biased coin designs with heterogeneous responses. J Stat Plan Inference2009; 139: 3220–3230.
117.
SverdlovOChangMDmitrienkoA, et al.Modern adaptive randomized clinical trials: statistical and practical aspects. Boca Raton, FL: CRC Press, 2016.
118.
TsiatisAA. Semiparametric theory and missing data. New York, NY: Springer, 2006.
119.
ChenXLeeKMVillarSS, et al.Some performance considerations when using multi-armed bandit algorithms in the presence of missing data. PLoS One2022; 17: e0274272.
120.
RubinDB. Multiple imputation for nonresponse in surveys. New York, NY: John Wiley & Sons, 2004.
121.
RobinsJMRotnitzkyA. Semiparametric efficiency in multivariate regression models with missing data. J Am Stat Assoc1995; 90: 122–129.
122.
ScharfsteinDORotnitzkyARobinsJM. Adjusting for nonignorable drop-out using semiparametric nonresponse models. J Am Stat Assoc1999; 94: 1096–1120.
123.
LiangKYZegerSL. Longitudinal data analysis using generalized linear models. Biometrika1986; 73: 13–22.
124.
AndersonTW. Maximum likelihood estimates for a multivariate normal distribution when some observations are missing. J Am Stat Assoc1957; 52: 200–203.
125.
LittleRJA. Pattern-mixture models for multivariate incomplete data. J Am Stat Assoc1993; 88: 125–134.
126.
LittleRJARubinDB. Statistical analysis with missing data. 3rd ed. Hoboken, NJ: John Wiley & Sons, 2020.
127.
MaZ. Missing data and adaptive designs in clinical studies. PhD Thesis, University of Virginia, 2013.
128.
RosenbergerWFLachinJM. Randomization in clinical trials: theory and practice. 2nd ed. Hoboken, NJ: John Wiley & Sons, 2016.
129.
BiswasARaoJ. Missing responses in adaptive allocation design. Stat Probab Lett2004; 70: 59–70.
130.
HardwickJOehmkeRStoutQF. Adaptive allocation in the presence of missing outcomes. Comput Sci Stat1998; 30: 219–223.
131.
BarohnRJGajewskiBPasnoorM, et al.Patient assisted intervention for neuropathy: comparison of treatment in real life situations (PAIN-CONTRoLS): Bayesian adaptive comparative effectiveness randomized trial. JAMA Neurol2021; 78: 68.
132.
FaseruBEllerbeckEFCatleyD, et al.Changing the default for tobacco-cessation treatment in an inpatient setting: study protocol of a randomized controlled trial. Trials2017; 18: 379.
133.
BouneffoufDUpadhyaySKhazaeniY. Contextual bandit with missing rewards. arXiv:200706368, 2020.
134.
O’BrienBGreenCEAl-JurdiR, et al.Bayesian adaptive randomization trial of intravenous ketamine for veterans with late-life, treatment-resistant depression. Contemp Clin Trials Commun2019; 16: 100432.
135.
AngusDCBerrySLewisRJ, et al.The REMAP-CAP (randomized embedded multifactorial adaptive platform for community-acquired pneumonia) study. Rationale and design. Ann Am Thorac Soc2020; 17: 879–891.
136.
UpadhyaySYurochkinMAgarwalM, et al.Online semi-supervised learning with bandit feedback. arXiv:201012574, 2020.
BaiZDHuFRosenbergerWF. Asymptotic properties of adaptive designs for clinical trials with delayed response. Ann Stat2002; 30: 122–139.
139.
WilliamsonSFJackoPJakiT. Generalisations of a Bayesian decision-theoretic randomisation procedure and the impact of delayed responses. Comput Stat Data Anal2022; 174: 107407.
140.
GranholmALangeTHarhayMO, et al.Effects of duration of follow-up and lag in data collection on the performance of adaptive clinical trials. Pharm Stat2024; 23: 138–150.
141.
TamuraRNFariesDEAndersenJS, et al.A case study of an adaptive clinical trial in the treatment of out-patients with depressive disorder. J Am Stat Assoc1994; 89: 768–776.
142.
XuJYinG. Two-stage adaptive randomization for delayed response in clinical trials. J R Stat Soc C: Appl Stat2014; 63: 559–578.
143.
LinRThallPFYuanY. A phase I–II basket trial design to optimize dose-schedule regimes based on delayed outcomes. Bayesian Anal2021; 16: 179–202.
144.
LinRThallPFYuanY. An adaptive trial design to optimize dose-schedule regimes with delayed outcomes. Biometrics2020; 76: 304–315.
145.
ZhangYCaoSZhangC, et al.A Bayesian adaptive phase I/II clinical trial design with late-onset competing risk outcomes. Biometrics2021; 77: 796–808.
146.
ZhangYGuoBCaoS, et al.SCI: a Bayesian adaptive phase I/II dose-finding design accounting for semi-competing risks outcomes for immunotherapy trials. Pharm Stat2022; 21: 960–973.
147.
LinRYuanY. Time-to-event model-assisted designs for dose-finding trials with delayed toxicity. Biostatistics2020; 21: 807–824.
148.
ZhangYZangY. CWL: a conditional weighted likelihood method to account for the delayed joint toxicity–efficacy outcomes for phase I/II clinical trials. Stat Methods Med Res2021; 30: 892–903.
149.
ZhouYLinRLeeJJ, et al.TITE-BOIN12: a Bayesian phase I/II trial design to find the optimal biological dose with late-onset toxicity and efficacy. Stat Med2022; 41: 1918–1931.
150.
LancewickiTSegalSKorenT, et al.Stochastic multi-armed bandits with unrestricted delay distributions. In: International conference on machine learning. PMLR, 2021, pp.5969–5978.
151.
ShiLWangJWuT. Statistical inference on multi-armed bandits with delayed feedback. In: International conference on machine learning. PMLR, 2023, pp. 31328–31352.
152.
Even-DarEMannorSMansourY. Action elimination and stopping conditions for the multi-armed bandit and reinforcement learning problems. J Mach Learn Res2006; 7: 1079–1105.
153.
JinIHLiuSThallPF, et al.Using data augmentation to facilitate conduct of phase I–II clinical trials with delayed outcomes. J Am Stat Assoc2014; 109: 525–536.
154.
EmmanuelTMaupongTMpoelengD, et al.A survey on missing data in machine learning. J Big Data2021; 8: 140.
155.
ThomasTRajabiE. A systematic review of machine learning-based missing value imputation techniques. Data Technol Appl2021; 55: 558–585.
156.
ButcherNJMonsourAMewEJ, et al.Guidelines for reporting outcomes in trial reports: the consort-outcomes 2022 extension. J Am Med Assoc2022; 328: 2252–2264.
157.
DigglePJTaylor-RobinsonD. Longitudinal data analysis. In: Handbook of epidemiology. New York, NY: Springer New York, 2024, pp.1–34.
158.
DeliuNWilliamsJJChakrabortyB. Reinforcement learning in modern biostatistics: constructing optimal adaptive interventions. Int Stat Rev2024; 93: 385–424.
159.
YiYWangX. A Markov decision process for response adaptive designs. Econom Stat2023; 25: 125–133.
160.
DaiJGraduPHarshawC. Clip-OGD: an experimental design for adaptive neyman allocation in sequential experiments. In: Advances in neural information processing systems. Volume 36. NeurIPS, 2023, pp.32235–32269.
161.
ShiLWeiWWangJ. Using surrogates in covariate-adjusted response-adaptive randomization experiments with delayed outcomes. In: Advances in neural information processing systems. Vol. 37. Curran Associates, Inc., 2024, pp.71085–71121.
162.
NorwoodPP. Advancements in adaptive randomization for high-cost experiments. Raleigh, NC: North Carolina State University, 2023.
