A comparative study of restricted randomization procedures for multiarm trials with equal or unequal treatment allocation ratios
Yevgen Ryeznik
Department of Mathematics, Uppsala University, Uppsala, Sweden
Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden
Search for more papers by this authorCorresponding Author
Oleksandr Sverdlov
Early Development Biostatistics, Novartis Institutes for Biomedical Research, East Hanover, NJ, USA
Correspondence
Oleksandr Sverdlov, Early Development Biostatistics, Novartis Institutes for Biomedical Research, East Hanover, NJ, USA.
Email: [email protected]
Search for more papers by this authorYevgen Ryeznik
Department of Mathematics, Uppsala University, Uppsala, Sweden
Department of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden
Search for more papers by this authorCorresponding Author
Oleksandr Sverdlov
Early Development Biostatistics, Novartis Institutes for Biomedical Research, East Hanover, NJ, USA
Correspondence
Oleksandr Sverdlov, Early Development Biostatistics, Novartis Institutes for Biomedical Research, East Hanover, NJ, USA.
Email: [email protected]
Search for more papers by this authorAbstract
Randomization designs for multiarm clinical trials are increasingly used in practice, especially in phase II dose-ranging studies. Many new methods have been proposed in the literature; however, there is lack of systematic, head-to-head comparison of the competing designs. In this paper, we systematically investigate statistical properties of various restricted randomization procedures for multiarm trials with fixed and possibly unequal allocation ratios. The design operating characteristics include measures of allocation balance, randomness of treatment assignments, variations in the allocation ratio, and statistical characteristics such as type I error rate and power. The results from the current paper should help clinical investigators select an appropriate randomization procedure for their clinical trial. We also provide a web-based R shiny application that can be used to reproduce all results in this paper and run simulations under additional user-defined experimental scenarios.
Supporting Information
| Filename | Description |
|---|---|
| sim7817-sup-0001-Ryeznik_Sverdlov_REVIED_Supplemental_Appendix.docxWord 2007 document , 2.6 MB |
Figure S1.1:Momentum of Probability Mass vs. Allocation Step for the 1:1:1:1 Allocation Figure S1.2:Momentum of Probability Mass vs. Allocation Step for the 2:1:1:2 Allocation Figure S1.3:Momentum of Probability Mass vs. Allocation Step for the 37:21:21:21 Allocation Figure S2.1:Forcing Index vs. Allocation Step for the 1:1:1:1 Allocation Figure S2.2:Forcing Index vs. Allocation Step for the 2:1:1:2 Allocation Figure S2.3:Forcing Index vs. Allocation Step for the 37:21:21:21 Allocation Figure S3.1:Heat Map Plot of the “Overall” Balance/Randomness Performance Index G for the 1:1:1:1 Allocation Figure S3.2:Heat Map Plot of the “Overall” Balance/Randomness Performance Index G for the 2:1:1:2 Allocation Figure S3.3:Heat Map Plot of the “Overall” Balance/Randomness Performance Index G for the 37:21:21:21 Allocation Figure S4.1:Distributions of Treatment Allocation Proportions for the 1:1:1:1 Allocation Figure S4.2:Distributions of Treatment Allocation Proportions for the 2:1:1:2 Allocation Figure S4.3:Distributions of Treatment Allocation Proportions for the 37:21:21:21 Allocation Figure S5.1:Average Standard Deviation of the Allocation Proportions vs. Average Forcing Index for the 1:1:1:1 Allocation Figure S5.2:Average Standard Deviation of the Allocation Proportions vs. Average Forcing Index for the 2:1:1:2 Allocation Figure S5.3:Average Standard Deviation of the Allocation Proportions vs. Average Forcing Index for the 37:21:21:21 Allocation Figure S6.1:Simulated Unconditional Treatment Randomization Probabilities for DBCD, MaxEnt, and MWUD for the 1:1:1:1 Allocation and Sample size up to 200 Figure S6.2:Simulated Unconditional Treatment Randomization Probabilities for DBCD, MaxEnt, and MWUD for the 2:1:1:1 Allocation and Sample size up to 200 Figure S6.3:Simulated Unconditional Treatment Randomization Probabilities for DBCD, MaxEnt, and MWUD for the 37:21:21:21 Allocation and Sample size up to 200 |
Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
REFERENCES
- 1Rosenberger WF, Lachin J. Randomization in Clinical Trials: Theory and Practice. 2nd ed. New York: Wiley; 2015.
- 2Dumville JC, Hahn S, Miles JNV, Torgerson DJ. The use of unequal randomisation ratios in clinical trials: a review. Contemp Clin Trials. 2006; 27(1): 1-12.
- 3Peckham E, Brabyn S, Cook L, Devlin T, Dumville J, Torgerson DJ. The use of unequal randomisation ratios in clinical trials—an update. Contemp Clin Trials. 2015; 45(Pt A): 113-122.
- 4Sverdlov O, Rosenberger WF. On recent advances in optimal allocation designs in clinical trials. J Stat Theory Pract. 2013; 7(4): 753-773.
10.1080/15598608.2013.783726 Google Scholar
- 5Biedermann S, Dette H, Zhu W. Optimal designs for dose-response models with restricted design spaces. J Am Stat Assoc. 2006; 101(474): 747-759.
- 6Dette H, Bretz F, Pepelyshev A, Pinheiro J. Optimal designs for dose-finding studies. J Am Stat Assoc. 2008; 103(483): 1225-1237.
- 7Bretz F, Dette H, Pinheiro J. Practical considerations for optimal designs in clinical dose-finding studies. Stat Med. 2010; 29(7-8): 731-742.
- 8Miller F, Guilbaud O, Dette H. Optimal designs for estimating the interesting part of a dose-effect curve. J Biopharm Stat. 2007; 17(6): 1097-1115.
- 9Hu F, Rosenberger WF. The Theory of Response-Adaptive Randomization in Clinical Trials. New York: Wiley; 2006.
10.1002/047005588X Google Scholar
- 10Berger VW, Ivanova A, Knoll MD. Minimizing predictability while retaining balance through the use of less restrictive randomization procedures. Stat Med. 2003; 22(19): 3017-3028.
- 11Zhao W, Weng Y, Wu Q, Palesch Y. Quantitative comparison of randomization designs in sequential clinical trials based on treatment balance and allocation randomness. Pharm Stat. 2012; 11(1): 39-48.
- 12Atkinson AC. Selecting a biased-coin design. Stat Sci. 2014; 29(1): 144-163.
- 13Baldi Antognini A, Zagoraiou M. Balance and randomness in sequential clinical trials: the dominant biased coin design. Pharm Stat. 2014; 13(2): 119-127.
- 14Baldi Antognini A, Rosenberger WF, Wang Y, Zagoraiou M. Exact optimum coin bias in Efron's randomization procedure. Stat Med. 2015; 34(28): 3760-3768.
- 15Berger VW, Bejleri K, Agnor R. Comparing MTI randomization procedures to blocked randomization. Stat Med. 2016; 35(5): 685-694.
- 16Selmaj K, Li DK, Hartung HP, et al. Siponimod for patients with relapsing-remitting multiple sclerosis (BOLD): an adaptive, dose-ranging, randomised, phase 2 study. Lancet Neurol. 2013; 12(8): 756-767.
- 17Sun H, Dodick DW, Silberstein S, et al. Safety and efficacy of AMG 334 for prevention of episodic migraine: a randomised, double-blind, placebo-controlled, phase 2 trial. Lancet Neurol. 2016; 15(4): 382-390.
- 18Verma A, Jiang CY, Betts TR, et al. Approaches to catheter ablation for persistent atrial fibrillation. N Engl J Med. 2015; 372(19): 1812-1822.
- 19Mayer RJ, Van Cutsem E, Falcone A, et al. Randomized trial of TAS-102 for refractory metastatic colorectal cancer. N Engl J Med. 2015; 372(20): 1909-1919.
- 20Powers SW, Coffey CS, Chamberlin LA, et al. Trial of amitriptyline, topiramate, and placebo for pediatric migraine. N Engl J Med. 2017; 376(2): 115-124.
- 21Ivanova A. A play-the-winner-type urn design with reduced variability. Metrika. 2003; 58(1): 1-13.
- 22Hu F, Zhang LX. Asymptotic properties of doubly adaptive biased coin designs for multitreatment clinical trials. Ann Stat. 2004; 32(1): 268-301.
- 23Zhao W, Weng Y. Block urn design—a new randomization algorithm for sequential trials with two or more treatments and balanced or unbalanced allocation. Contemp Clin Trials. 2011; 32(6): 953-961.
- 24Zhao W. Mass weighted urn design—a new randomization algorithm for unequal allocations. Contemp Clin Trials. 2015; 43: 209-216.
- 25Kuznetsova OM, Tymofyeyev Y. Brick tunnel randomization for unequal allocation to two or more treatment groups. Stat Med. 2011; 30(8): 812-824.
- 26Kuznetsova OM, Tymofyeyev Y. Wide brick tunnel randomization—an unequal allocation procedure that limits the imbalance in treatment totals. Stat Med. 2014; 33(9): 1514-1530.
- 27Berger VW. Selection Bias and Covariate Imbalances in Randomized Clinical Trials. New York: Wiley; 2005.
- 28Hu F, Rosenberger WF, Zhang LX. Asymptotically best response-adaptive randomization procedures. J Stat Plan Inference. 2006; 136(6): 1911-1922.
- 29Klotz JH. Maximum entropy constrained balance randomization in clinical trials. Biometrics. 1978; 34(2): 283-287.
- 30Efron B. Forcing a sequential experiment to be balanced. Biometrika. 1971; 58(3): 403-417.
- 31Titterington DM. On constrained balance randomization for clinical trials. Biometrics. 1983; 39(4): 1083-1086.
- 32Bazaraa MS, Sherali HD, Shetty CM. Nonlinear Programming: Theory and Algorithms. New York: Wiley; 1993.
- 33Salama I, Ivanova A, Qaqish B. Efficient generation of constrained block allocation sequences. Stat Med. 2008; 27(9): 1421-1428.
- 34Zhao W, Ramakrishnan V. Generalizations of Wei's urn design to unequal allocations in sequential clinical trials. Contemp Clin Trials Commun. 2016; 2: 75-79.
- 35Kuznetsova OM, Plamadeala Johnson V. Approaches to expanding the two-arm biased coin randomization to unequal randomization while preserving the unconditional allocation ratio. Stat Med. 2017; 36(16): 2483-2498.
- 36Wei LJ, Smythe RT, Smith RL. K-treatment comparison with restricted randomization rules in clinical trials. Ann Stat. 1986; 14(1): 265-274.
- 37Smith RL. Properties of biased coin designs in sequential clinical trials. Ann Stat. 1984; 12(3): 1018-1034.
- 38Kuznetsova OM. Brick tunnel randomization and the momentum of the probability mass. Stat Med. 2015; 34: 4031-4056.
- 39Heritier S, Gebski V, Pillai A. Dynamic balancing randomization in controlled clinical trials. Stat Med. 2005; 24(24): 3729-3741.
- 40Zhao W. Letter to the Editor—A better alternative to the inferior permuted block design is not necessarily complex. Stat Med. 2016; 35: 1736-1738.
- 41Hu Y. Generalized Efron's biased coin design and its theoretical properties. J Appl Probab. 2016; 53(2): 327-340.
- 42Baldi Antognini A, Giovagnoli A. A new ‘biased coin design’ for the sequential allocation of two treatments. Appl Stat. 2004; 53(4): 651-664.
- 43Kuznetsova OM, Tymofyeyev Y. Expansion of the modified Zelen's approach randomization and dynamic randomization with partial block supplies at the centers to unequal allocation. Contemp Clin Trials. 2011; 32(6): 962-972.
- 44Kuznetsova OM, Tymofyeyev Y. Preserving the allocation ratio at every allocation with biased coin randomization and minimization in studies with unequal allocation. Stat Med. 2012; 31(8): 701-723.
- 45Proschan M, Brittain E, Kammerman L. Minimize the use of minimization with unequal allocation. Biometrics. 2011; 67(3): 1135-1141.
- 46Kaiser LD. Dynamic randomization and a randomization model for clinical trials data. Stat Med. 2012; 31(29): 3858-3873.
- 47Kennes LN, Cramer E, Hilgers RD, Heussen N. The impact of selection bias on test decisions in randomized clinical trials. Stat Med. 2011; 30(21): 2573-2581.
- 48Tamm M, Cramer E, Kennes LN, Heussen N. Influence of selection bias on the test decision—a simulation study. Methods Inf Med. 2012; 51(2): 138-143.
- 49Tamm M, Hilgers RD. Chronological bias in randomized clinical trials arising from different types of unobserved time trends. Methods Inf Med. 2014; 53(6): 501-510.
- 50Rückbeil MV, Hilgers RD, Heussen N. Assessing the impact of selection bias on test decisions in trials with a time-to-event outcome. Stat Med. 2017; 36(17): 2656-2668.
- 51Uschner D, Hilgers RD, Heussen N. The impact of selection bias in randomized multi-arm parallel group clinical trials. PLoS One. 2018; 13(1):e0192065.
- 52Hilgers RD, Uschner D, Rosenberger WF, Heussen N. ERDO—a framework to select an appropriate randomization procedure for clinical trials. BMC Med Res Methodol. 2017; 17(1): 159.
- 53Galbete A, Rosenberger WF. On the use of randomization tests following adaptive designs. J Biopharm Stat. 2016; 26(3): 466-473.
- 54Rosenberger WF, Sverdlov O. Handling covariates in the design of clinical trials. Stat Sci. 2008; 23(3): 404-419.
