Exact simultaneous confidence intervals for logical selection of a biomarker cut‐point

Abstract

This article proposes four new principles for logical biomarker cut‐point selection methods to adhere to: subgroup sensibility, sensitivity, specificity, and target monotonicity. At every cut‐point value, our method gives confidence intervals not only for the efficacy at that cut‐point value, but also efficacies in the marker‐positive and marker‐negative subgroups defined by that cut‐point. These confidence intervals are given simultaneously for all possible cut‐point values. Using Alzheimer's disease (AD) and type 2 diabetes (T2DM) as examples, we show our method achieves the four principles. Our method strongly controls familywise type I error rate (FWER) across both levels of multiplicity: the multiplicity of having marker‐positive and marker‐negative subgroups at each cut‐point, and the multiplicity of searching through infinitely many cut‐points. This is in contrast to other available methods. The confidence level of our simultaneous confidence intervals is in fact exact (not conservative). An application (app) is available, which implements the method we propose.

Overview of modern approaches for identifying and evaluating heterogeneous treatment effects from clinical data

“…18,25 Model complexity and control of data over-fitting What is the size of the covariate space that the method can handle? Some methods assume a very lowdimensional setting with the focus on a single continuous biomarker, for example, STEPP by Bonetti and Gelber, 26 and methods for a simultaneous inference for selected biomarker cut-point points in Han et al 27 At the other extreme, some methods can handle the number of predictors exceeding the sample size or growing as a function of sample size. 25 What is the complexity of the ''model space'' for a given HTE evaluation method?…”