Every few days, we will be publishing layman’s abstracts of new articles from our prestigious portfolio of journals in statistics. The aim is to highlight the latest research to a broader audience in an accessible format.
A flexible multi‐domain test with adaptive weights and its application to clinical trials. Pharmaceutical Statistics. 2020; 19: 315– 325. https://doi.org/10.1002/pst.1993, , .
When a clinical trial is being designed to evaluate a new investigational drug, its complexity often depends on the nature of the underlying disease. Some diseases manifest themselves in different ways in different patients. For example, in Mucopolysaccharidosis type I (MPS I), a rare lysosomal storage disorder, α-L-iduronidase enzyme deficiency leads to excessive accumulation of glycosaminoglycans in numerous tissues resulting in cardiac disease, respiratory disease, joint stiffness, developmental delay and other deleterious manifestations. In MPS I, an investigational treatment may ameliorate multiple clinical domains, and, therefore, evaluation of the treatment is suited to a multi-domain outcome. A single domain endpoint may fail to provide the whole picture of treatment efficacy.
In a clinical trial, investigators often administer a new drug to patients randomly assigned to receive treatment or placebo, so that the two groups may be compared in terms of efficacy and safety. To evaluate whether the new drug is more efficacious compared to placebo across multiple outcomes, this paper proposes a flexible statistical test and illustrates its application in a clinical trial.
Conventional tests, such as the O’Brien and Wei-Lachin tests, use a weighted sum of outcome-specific test statistics. This proposed test differentiates itself from these conventional tests in that the weights, instead of fixed, are selected adaptively via a data-driven algorithm. Simulations showed that the proposed test controls the type I error rate and has increased power over the O’Brien and Wei-Lachin tests in scenarios reflective of clinical trial settings. Data from a clinical trial in a rare lysosomal storage disorder were used to illustrate the properties of the proposed test. As a strategy of combining statistical evidence from multiple outcomes, the proposed test is flexible and readily applicable to a variety of clinical trial scenarios.