A post hoc evaluation of a sample size re-estimation in the Secondary Prevention of Small Subcortical Strokes study

Abstract

Background/Aims The use of adaptive designs has been increasing in randomized clinical trials. Sample size re-estimation is a type of adaptation in which nuisance parameters are estimated at an interim point in the trial and the sample size re-computed based on these estimates. The Secondary Prevention of Small Subcortical Strokes (SPS3) study was a randomized clinical trial assessing the impact of single vs. dual antiplatelet therapy and control of systolic blood pressure to a higher (130-149 mmHg) vs. lower (<130 mmHg) on recurrent stroke risk in a two-by-two factorial design. A sample size re-estimation was performed during the SPS3 study resulting in an increase from the planned sample size of 2500 to 3020, and we sought to determine the impact of the sample size re-estimation on the study results. Methods We assessed the results of the primary efficacy and safety analyses with the full 3020 patients and compared them to the results that would have been observed had randomization ended with 2500 patients. The primary efficacy outcome considered was recurrent stroke, and the primary safety outcomes were major bleeds and death. We computed incidence rates for the efficacy and safety outcomes and used Cox proportional hazards models to examine the hazard ratios for each of the two treatment interventions (i.e. the antiplatelet and blood pressure interventions). Results In the antiplatelet intervention, the hazard ratio was not materially modified by increasing the sample size, nor did the conclusions regarding the efficacy of mono vs. dual-therapy change: there was no difference in the effect of dual vs. mono therapy on the risk of recurrent stroke hazard ratios (n=3020 (95% CI): 0.92 (0.72, 1.2) p=0.48; n=2500 (95%CI): 1.0 (0.78, 1.3), p=0.85). With respect to the blood pressure intervention, increasing the sample size resulted in less certainty in the results, as the hazard ratio for higher vs. lower systolic blood pressure target approached, but did not achieve, statistical significance with the larger sample (n=3020 (95% CI): 0.81 (0.63, 1.), p=0.089; n=2500 (95% CI): 0.89 (0.68, 1.17), p=0.40). The results from the safety analyses were similar with 3020 and 2500 patients for both study interventions. Other trial-related factors, such as contracts, finances and study management were impacted as well. Conclusions Adaptive designs can have benefits in randomized clinical trials, but do not always result in significant findings. The impact of adaptive designs should be measured in terms of both trial results, as well as practical issues related to trial management. More post-hoc analyses of study adaptations will lead to better understanding of the balance between the benefits and the costs.

Follow up after sample size re-estimation in a breast cancer randomized trial for disease-free survival

“…Although the statistical and clinical trial literature has many examples of work to theoretically evaluate sample size re-estimation methods and performance through simulation, there have been few examples from real-world applications to detail the implementation and impact of adaptations, particularly from trials utilizing blinded sample size re-estimation [1, 2, 25]. McClure et al provide a detailed review of their re-estimation procedure and trial results from a trial of secondary prevention of small subcortical stroke.…”

“…Evaluation of sample size re-estimation procedures, following the completion of a clinical trial, has been sparsely reported in the literature [1, 2]. Based on the acknowledgment of uncertainty in the information used as a basis for sample size determination at the design stage, various methods of re-evaluation of sample size have been developed [3–8].…”

“…The 2010 draft guidance from the US Food and Drug Administration provides review and recommendations for adaptive clinical trial designs, and implications of these in the regulatory setting [16]. While SSRE methods have been discussed widely, implementation of these has not been widespread, particularly among non-pharmaceutical-sponsored studies [2] and reporting on them is even more rare [1]. Empiric evaluation of SSRE methods is necessary to more fully understand their merits and limitations, and to refine these procedures for more efficient design.…”

Application of sample size re-estimation in clinical trials: A systematic review

“…Consequently, the clinical trial sponsor has a number of factors to consider regarding SSR application [ 12 ]. The use of AD, including SSR, can be beneficial in many circumstances, but the benefit may not be applicable in all cases, meaning that factors such as the statistical impact and the trial cost need to be considered [ 13 ].…”

Response‐adaptive treatment allocation for survival trials with clustered right‐censored data

“…The fundamental idea of a response‐adaptive design is to use the responses of the patients already enrolled in the clinical study to skew the treatment allocation probabilities to reach the predefined objective, such as a reduction of the number of patients who receive the inferior treatment. Because of its attractive properties, adaptive design is currently considered a viable option in many areas of clinical research and mainstream drug development . For example, in the field of global health, the appalling fatality rate of the West African Ebola epidemic in 2014 triggered more discussions of the benefits of an adaptive design for related clinical studies to lower the mortality rate …”