Flexible trial design in practice - stopping arms for lack-of-benefit and adding research arms mid-trial in STAMPEDE: a multi-arm multi-stage randomized controlled trial

Table 1 Design parameters by trial stage for original research arms

Trial Stage	Type	Primary outcome measure	Target hazard ratio	Power	One-sided Significance level	Critical HR	Trigger events¹(control)
1	Activity	FFS	0.75	95%	0.500	1.00	114
2	Activity	FFS	0.75	95%	0.250	0.92	215
3	Activity	FFS	0.75	95%	0.100	0.89	334
4	Efficacy	OS	0.75	90%	0.025	-	400
Overall	-	-	-	83%	0.013	-	-

¹ Number of control arm events provoking analyses for the original research comparisons with a 2:1 allocation ratio in favour of control. The required number of events would be different for an equal allocation ratio.
Key: FFS = failure-free survival; OS = overall survival; HR = hazard ratio.
Note: these parameters are used for each pairwise comparison of research vs control.

ISSN: 1745-6215