Back to basics: explaining sample size in outcome trials, are statisticians doing a thorough job?

Abstract

Time to event outcome trials in clinical research are typically large, expensive and high-profile affairs. Such trials are commonplace in oncology and cardiovascular therapeutic areas but are also seen in other areas such as respiratory in indications like chronic obstructive pulmonary disease. Their progress is closely monitored and results are often eagerly awaited. Once available, the top line result is often big news, at least within the therapeutic area in which it was conducted, and the data are subsequently fully scrutinized in a series of high-profile publications. In such circumstances, the statistician has a vital role to play in the design, conduct, analysis and reporting of the trial. In particular, in drug development it is incumbent on the statistician to ensure at the outset that the sizing of the trial is fully appreciated by their medical, and other non-statistical, drug development team colleagues and that the risk of delivering a statistically significant but clinically unpersuasive result is minimized. The statistician also has a key role in advising the team when, early in the life of an outcomes trial, a lower than anticipated event rate appears to be emerging. This paper highlights some of the important features relating to outcome trial sample sizing and makes a number of simple recommendations aimed at ensuring a better, common understanding of the interplay between sample size and power and the final result required to provide a statistically positive and clinically persuasive outcome.

A phase II open-label randomized study to assess the efficacy and safety of selumetinib (AZD6244 [ARRY-142886]) versus capecitabine in patients with advanced or metastatic pancreatic cancer who have failed first-line gemcitabine therapy

“…The Type I and Type II error were adjusted to be less constrained, so that the targeted treatment benefit was appropriate while the sample size remained reasonable [14]. The study was powered on a hypothesized treatment effect HR of 0.5, meaning that with 38 events a HR of 0.66 (the critical value, corresponding P< 0.2) or lower would have resulted in a positive signal for further development of selumetinib monotherapy in this patient population [15]. Therefore, recruitment of a minimum of 64 patients was required.…”

Influence of statistician involvement on reporting of randomized clinical trials in medical oncology

“…The involvement of a qualified statistician in the design, analysis, and reporting of RCTs can potentially address these deficiencies [6][7][8]. However, the degree of the statistician's involvement is difficult to ascertain in many RCT reports.…”

The role of the minimum clinically important difference and its impact on designing a trial

“…This fact was also described in Hung and O'Neill [11] and for survival studies, in Carroll [12]. By construction, the chance of observing a difference at least this big when the true effect is equal to the MCID is 90%.…”

“…The latter, in our opinion, should be the exception instead of a general expectation. The need to clearly distinguish between an expectation on the true (unknown) treatment effect and a requirement on the observed treatment effect when determining the sample size for a study is discussed in Carroll [12].…”