Enrichment designs with a continuous biomarker require the estimation of a threshold to determine the subpopulation benefitting from the treatment. This article provides the optimal allocation for inference in a two-stage enrichment design for treatment comparisons when a continuous biomarker is suspected to affect patient response. Several design criteria, associated with different trial objectives, are optimized under balanced or Neyman allocation and under equality of the first two empirical biomarker’s moments. Moreover, we propose a new covariate-adaptive randomization procedure that converges to the optimum with the fastest available rate. Theoretical and simulation results show that this strategy improves the efficiency of a two-stage enrichment clinical trial, especially with smaller sample sizes and under heterogeneous responses.
LutySECarterJDMcKenzieJM, et al. Randomised controlled trial of interpersonal psychotherapy and cognitive-behavioural therapy for depression. Br J Psychiatry2007; 190: 496–502.
2.
SteingrimssonJABetzJQianT, et al. Optimized adaptive enrichment designs for three-arm trials: learning which subpopulations benefit from different treatments. Biostatistics2021; 22: 283–297.
3.
StallardN. Adaptive enrichment designs with a continuous biomarker. Biometrics2023; 79: 9–19.
4.
SpencerAHarbronCManderA, et al. An adaptive design for updating the threshold value of a continuous biomarker. Stat Med2016; 35: 4909–4923.
5.
BlangeroYRabilloudMEcochardR, et al. A Bayesian method to estimate the optimal threshold of a marker used to select patients’ treatment. Stat Methods Med Res2020; 29: 29–43.
6.
Baldi AntogniniAFrieriRZagoraiouM. New insights into adaptive enrichment designs. Statistical Papers 2023. DOI:10.1007/s00362-023-01433-0
7.
SimonNSimonR. Adaptive enrichment designs for clinical trials. Biostatistics2013; 14: 613–625.
8.
SimonRMaitournamA. Evaluating the efficiency of targeted designs for randomized clinical trials. Clin Cancer Res2004; 10: 6759–6763.
Russek-CohenESimonR. Evaluating treatments when a gender by treatment interaction may exist. Stat Med1997; 16: 455–464.
11.
FollmannD. Adaptively changing subgroup proportions in clinical trials. Stat Sin1997; 7: 1085–1102.
12.
ThaseME. Using biomarkers to predict treatment response in major depressive disorder: evidence from past and present studies. Dialogues Clin Neurosci2014; 16: 539–544.
13.
StrawbridgeRYoungACleareA. Biomarkers for depression: recent insights, current challenges and future prospects. Neuropsychiatr Dis Treat2017; 13: 1245–1262.
14.
EllenbogenKAGoldMRMeyerTE, et al. Primary results from the smartdelay determined av optimization: a comparison to other AV delay methods used in cardiac resynchronization therapy (smart-av) trial. Circulation2010; 122: 2660–2668.
15.
SteinKEllenbogenKGoldM, et al. Smartdelay determined av optimization: a comparison of AV delay methods used in cardiac resynchronization therapy (smart-av): rationale and design. Pacing Clin Electrophysiol2010; 33: 54–63.
FrieriRRosenbergerWFlournoyN, et al. Design considerations for two-stage enrichment trial. Biometrics2022. DOI:10.1111/biom.13805.
18.
ZhuHHuFZhaoH. Adaptive clinical trials to detect interaction between treatment and a dichotomous biomarker. Can J Stat2013; 41: 525–539.
19.
LinZFlournoyNRosenbergerW. Inference for a two-stage enrichment design. Ann Stat2021; 49: 2697–2720.
20.
JiangWFreidlinBSimonR. Biomarker-adaptive threshold design: a procedure for evaluating treatment with possible biomarker-defined subset effect. J Natl Cancer Inst2007; 99: 1036–1043.
21.
WasonJMJakiT. A review of statistical designs for improving the efficiency of phase II studies in oncology. Stat Methods Med Res2016; 25: 1010–1021.
22.
AtkinsonACBiswasA. Adaptive biased-coin designs for skewing the allocation proportion in clinical trials with normal responses. Stat Med2005; 24: 2477–2492.
23.
AtkinsonAC. Optimum biased coin designs for sequential clinical trials with prognostic factors. Biometrika1982; 69: 61–67.
24.
AtkinsonAC. Optimum designs for two treatments with unequal variances in the presence of covariates. Biometrika2015; 102: 494–499.
25.
AzrielDRinottYPoschM. Optimal designs for the development of personalized treatment rules. Scand J Stat2023; 50: 797–824.
26.
RosenbergerWFLachinJL. Randomization in Clinical Trials: Theory and Practice. New York: John Wiley & Sons, 2002.
27.
HuFHuYMaZ, et al. Adaptive randomization for balancing over covariates. WIRES Comput Stat2014; 6: 288–303.
28.
ZagoraiouM. Choosing a covariate-adaptive randomization procedure in practice. J Biopharm Stat2017; 27: 845–857.
Baldi AntogniniAFrieriRZagoraiouM, et al. The efficient covariate-adaptive design for high-order balancing of quantitative and qualitative covariates. Stat Pap2022. DOI:10.1007/s00362-022-01381-1.
31.
EfronB. Forcing sequential experiments to be balanced. Biometrika1971; 58: 403–417.
32.
Baldi AntogniniAZagoraiouM. On the almost sure convergence of adaptive allocation procedures. Bernoulli2015; 21: 881–908.
33.
MaZHuF. Balancing continuous covariates based on kernel densities. Contemp Clin Trials2013; 34: 262–269.
34.
CandelMJJMVan BreukelenGJP. Sample size calculation for treatment effects in randomized trials with fixed cluster sizes and heterogeneous intraclass correlations and variances. Stat Methods Med Res2015; 24: 557–573.
35.
Van BreukelenGJPCandelMJJM. Efficient design of cluster randomized trials with treatment-dependent costs and treatment-dependent unknown variances. Stat Med2018; 34: 37: 3027–3046.
36.
MeynSPTweedieRL. Stability of markovian processes I: criteria for discrete time Markov chains. Adv Appl Probab1992; 24: 542–574.
