The estimand framework, which was first introduced in 2019, is a relatively new topic in the drug development and clinical trial analysis realm. The goal of this concept is to create a clear and consistent mechanism which describes what clinical trials are studying and how it will be analyzed so that all interested parties (e.g., pharmaceutical companies, regulatory authorities, physicians, patients, etc.) can more easily decipher the benefits and risks of a treatment. To date, this has meant that estimands are most commonly used in Phase IIb/Phase III registrational studies and focus on communicating about a drug’s efficacy. However, as discussed during this year’s Joint Statistics Meetings, the use of estimands is broadening. Below are items to further consider when designing your next clinical trial.

1) When should you use estimands?
Throughout the drug development process, we are constantly gaining knowledge on the proposed treatment which can inform its risk/benefit profile. Given this, it has been suggested by a variety of regulatory & industry speakers that estimands be considered for any study including in the early phases of development as they may help with writing clinical study reports and lead to clearer communication of analysis aims & study results.

Examples include:
a) Early phase studies having estimands oriented around safety or pharmacokinetic objectives.
b) Registrational studies continuing their focus on efficacy objectives.
c) Late phase/post-approval studies having multiple estimands focusing on a mix of both efficacy & safety objectives.

2) Is it a sensitivity analysis or an analysis for a new estimand?
Sensitivity analysis have been previously defined by researchers as “a method to determine the robustness of an assessment by examining the extent to which results are affected by changes in methods, models, values of unmeasured variables, or assumptions” with the aim of identifying “results that are most dependent on questionable or unsupported assumptions” [1,2]. This definition allows for a large variety of analyses to be conducted under the sensitivity analysis banner. However, when we take into consideration the estimand framework, keep in mind that changing the set of assumptions used for a sensitivity analysis could inadvertently change the research question being answered. In these cases, we are no longer conducting a sensitivity analysis but instead are performing an analysis on a secondary or supplemental estimand. Careful attention will be needed when planning additional analyses to ensure that they are still aligned with the study’s aims.

3) How should intercurrent events (ICEs) be managed?
Intercurrent events are an optional element in the estimand framework which leads to a variability in how they are addressed. Some statisticians will argue that if a study defines the treatment properly by considering common events as part of treatment, then there will be no ICEs to consider during the conduct of the study. While no ICEs may be a viable option, it is more likely that your study will have ICEs that must be accounted for to assess the treatment’s effect more accurately.

Since the introduction of estimands, the answer to “Why are estimands necessary?” has remained constant – to allow for clearer communication about benefits/risks of a potential treatment to the relevant stakeholders. However, as illustrated here, the number of questions as to how & when to use them has grown considerably. Early consultation with statisticians about your study protocol and/or program development can help resolve these questions for you and provide for a smoother analysis and regulatory review of your proposed treatments.

[1] Thabane, L., Mbuagbaw, L., Zhang, S. et al. A tutorial on sensitivity analyses in clinical trials: the what, why, when and how. BMC Med Res Methodol 13, 92 (2013). https://doi.org/10.1186/1471-2288-13-92

[2] Schneeweiss S: Sensitivity analysis and external adjustment for unmeasured confounders in epidemiologic database studies of therapeutics. Pharmacoepidemiol Drug Saf. 2006, 15 (5): 291-303. 10.1002/pds.1200.

Scott Mollan, Associate Director, Biostatistics, has over 17 years of experience in clinical and non-clinical statistics across the CRO & pharmaceutical industry. Having led studies of all phases (pilot, pivotal, post-market, phases I-IV) and assisting clients during both the pre-submission phase and FDA approval via the NDA/PMA process (2 NDAs/10 PMAs), he has a wealth of experience to draw upon to support clients. Armed with graduate degrees in business and statistics, Mr. Mollan has been able to leverage his understanding of the clinical trial process via a diverse range of indications to publish on the medical device trial process, lung cancer diagnostics, and women’s health while similarly offering industry presentations on missing data analysis strategies and the use of adaptive trial designs within medical devices studies.