Restricted Mean Survival Time for a Randomized Study with Survival Outcome
Pub. online: 31 March 2025
Type: Statistical Data Science
Open Access
Open Access
Received
21 June 2024
21 June 2024
Accepted
17 March 2025
17 March 2025
Published
31 March 2025
31 March 2025
Abstract
When comparing two survival curves, three tests are widely used: the Cox proportional hazards test, the logrank test, and the Wilcoxon test. Despite their popularity in survival data analysis, there is no clear clinical interpretation especially when the proportional hazard assumption is not valid. Meanwhile, the restricted mean survival time (RMST) offers an intuitive and clinically meaningful interpretation. We compare these four tests with regards to statistical power under many configurations (e.g., proportional hazard, early benefit, delayed benefit, and crossing survivals) with data simulated from the Weibull distributions. We then use an example from a lung cancer trial to compare their required sample sizes. As expected, the CoxPH test is more powerful than others when the PH assumption is valid. The Wilcoxon test is often preferable when there is a decreasing trajectory in the event rate as time goes. The RMST test is much more powerful than others when a new treatment has early benefit. The recommended test(s) under each configuration are suggested in this article.
Supplementary material
Supplementary MaterialThe R function to compute simulated TIE and statistical power.
References
Cox DR (1972). Regression models and life-tables. Journal of the Royal Statistical Society, Series B, Methodological, 34(2): 187–202. https://doi.org/10.1111/j.2517-6161.1972.tb00899.x
Harrington DP, Fleming TR (1982). A class of rank test procedures for censored survival data. Biometrika, 69(3): 553–566. https://doi.org/10.1093/biomet/69.3.553
Jiang T, Cao B, Shan G (2020). Accurate confidence intervals for risk difference in meta-analysis with rare events. BMC Medical Research Methodology, 20(1): 98. https://doi.org/10.1186/s12874-020-00954-8
Liao JJ, Liu GF, Wu WC (2020). Dynamic RMST curves for survival analysis in clinical trials. BMC Medical Research Methodology, 20(1): 218. https://doi.org/10.1186/s12874-020-01098-5
Lu X, Zhang Y, Tang Y, Bernick C, Shan G (2025). Conversion to Alzheimer’s disease dementia from normal cognition directly or with the intermediate mild cognitive impairment stage. Alzheimer’s & Dementia, 21(1): e14393. https://doi.org/10.1002/alz.14393
Lu Y, Tian L (2021). Statistical considerations for sequential analysis of the restricted mean survival time for randomized clinical trials. Statistics in Biopharmaceutical Research, 13(2): 210–218. https://doi.org/10.1080/19466315.2020.1816491
Peto R, Peto J (1972). Asymptotically efficient rank invariant test procedures. Journal of the Royal Statistical Society. Series A. General, 135(2): 185. https://doi.org/10.2307/2344317
Royston P, Parmar MK (2013). Restricted mean survival time: an alternative to the hazard ratio for the design and analysis of randomized trials with a time-to-event outcome. BMC Medical Research Methodology, 13: 152. https://doi.org/10.1186/1471-2288-13-152
Shan G (2020). Two-stage optimal designs based on exact variance for a single-arm trial with survival endpoints. Journal of Biopharmaceutical Statistics, 30(5): 797–805. https://doi.org/10.1080/10543406.2020.1730869
Shan G (2021). Optimal two-stage designs based on restricted mean survival time for a single-arm study. Contemporary Clinical Trials Communications, 21: 100732. https://doi.org/10.1016/j.conctc.2021.100732
Shan G (2022). Randomized two-stage optimal design for interval-censored data. Journal of Biopharmaceutical Statistics, 32(2): 298–307. https://doi.org/10.1080/10543406.2021.2009499
Shan G, Zhang H (2019). Two-stage optimal designs with survival endpoint when the follow-up time is restricted. BMC Medical Research Methodology, 19: 74. https://doi.org/10.1186/s12874-019-0696-x
Shan G, Zhang Y, Tang Z, Ding A, Wu S (2025). Disease progression trajectory curves to estimate saved time in Alzheimer’s disease trialsitle. Contemporary Clinical Trials, 151: 107814. https://doi.org/10.1016/j.cct.2025.107814
Takiguchi Y, Moriya T, Asaka-Amano Y, Kawashima T, Kurosu K, Tada Y, et al. (2007). Phase II study of weekly irinotecan and cisplatin for refractory or recurrent non-small cell lung cancer. Lung Cancer, 58(2): 253–259. https://doi.org/10.1016/j.lungcan.2007.06.004
Tian L, Fu H, Ruberg SJ, Uno H, Wei LJ (2018). Efficiency of two sample tests via the restricted mean survival time for analyzing event time observations. Biometrics, 74(2): 694–702. https://doi.org/10.1111/biom.12770
Uno H, Wittes J, Fu H, Solomon SD, Claggett B, Tian L, et al. (2015). Alternatives to hazard ratios for comparing the efficacy or safety of therapies in noninferiority studies. Annals of Internal Medicine, 163(2): 127–134. https://doi.org/10.7326/M14-1741
Zhang Y, Li Y, Song S, Li Z, Lu M, Shan G (2024). Predicting conversion time from mild cognitive impairment to dementia with interval-censored models. Journal of Alzheimer’s Disease, 101(1): 147–157. https://doi.org/10.3233/JAD-240285
