BACKGROUND: Platform trials allow adding new experimental treatments to an on-going trial. This feature is attractive to practitioners due to improved efficiency. Nevertheless, the operating characteristics of a trial that adds arms have not been well-studied. One controversy is whether just the concurrent control data (i.e. of patients who are recruited after a new arm is added) should be used in the analysis of the newly added treatment(s), or all control data (i.e. non-concurrent and concurrent).
METHODS: We investigate the benefits and drawbacks of using non-concurrent control data within a two-stage setting. We perform simulation studies to explore the impact of a linear and a step trend on the inference of the trial. We compare several analysis approaches when one includes all the control data or only concurrent control data in the analysis of the newly added treatment.
RESULTS: When there is a positive trend and all the control data are used, the marginal power of rejecting the corresponding hypothesis and the type one error rate can be higher than the nominal value. A model-based approach adjusting for a stage effect is equivalent to using concurrent control data; an adjustment with a linear term may not guarantee valid inference when there is a non-linear trend.
CONCLUSIONS: If strict error rate control is required then non-concurrent control data should not be used; otherwise it may be beneficial if the trend is sufficiently small. On the other hand, the root mean squared error of the estimated treatment effect can be improved through using non-concurrent control data.
Contemp Clin Trials Commun. 2018 Aug 24;12:1-8
[PMID:
30182068]
Stat Med. 2018 Sep 20;37(21):3056-3077
[PMID:
29869347]
Cir Esp. 2019 Feb;97(2):65-70
[PMID:
30686474]
N Engl J Med. 2017 Jul 6;377(1):62-70
[PMID:
28679092]
Stat Med. 2011 Dec 30;30(30):3475-87
[PMID:
21953285]
Control Clin Trials. 2003 Oct;24(5):506-22
[PMID:
14500050]
World J Surg Oncol. 2018 Nov 30;16(1):230
[PMID:
30501634]
Clin Trials. 2011 Aug;8(4):390-7
[PMID:
21610005]
Stat Methods Med Res. 2020 Jun;29(6):1715-1727
[PMID:
31074333]
J Chronic Dis. 1987;40 Suppl 2:17S-24S
[PMID:
3667863]
Stat Med. 2019 Aug 15;38(18):3305-3321
[PMID:
31115078]
BMC Med Res Methodol. 2006 Mar 02;6:8
[PMID:
16509997]
BMJ Open. 2018 Feb 10;8(2):e018320
[PMID:
29440155]
J Biopharm Stat. 2012;22(4):758-72
[PMID:
22651113]
Stat Methods Med Res. 2016 Apr;25(2):716-27
[PMID:
23242385]
ERJ Open Res. 2018 May 08;4(2):
[PMID:
29761108]
Clin Trials. 2020 Jun;17(3):273-284
[PMID:
32063029]
J Biopharm Stat. 2007;17(6):1227-38
[PMID:
18027228]
J Biopharm Stat. 2005;15(4):575-91
[PMID:
16022164]
BMC Med Res Methodol. 2017 Dec 04;17(1):159
[PMID:
29202708]
BMJ Qual Saf. 2017 Nov;26(11):919-928
[PMID:
28947635]
Methods Inf Med. 2014;53(6):501-10
[PMID:
25396221]
Biostatistics. 2018 Apr 1;19(2):199-215
[PMID:
29036330]
Stat Methods Med Res. 2017 Dec;26(6):2853-2868
[PMID:
26546254]
Interact Cardiovasc Thorac Surg. 2009 Sep;9(3):494-9
[PMID:
19509097]
J Med Syst. 2019 Feb 12;43(3):74
[PMID:
30756197]
J R Soc Med. 2018 Apr;111(4):143-144
[PMID:
29648508]
BMC Med. 2018 Feb 28;16(1):29
[PMID:
29490655]
Indian J Pharmacol. 2010 Aug;42(4):201-7
[PMID:
20927243]
Biom J. 2020 Mar;62(2):361-374
[PMID:
31265159]
Pharm Stat. 2018 Mar;17(2):182-197
[PMID:
29266692]
Clin Trials. 2016 Jun;13(3):358-66
[PMID:
26908536]
Trials. 2015 Apr 22;16:179
[PMID:
25897686]
J Biopharm Stat. 2020;30(1):69-88
[PMID:
31017843]
BMC Med Res Methodol. 2007 Nov 03;7:46
[PMID:
17980042]
