An adaptive design for updating the threshold value of a continuous biomarker

Abstract

Potential predictive biomarkers are often measured on a continuous scale, but in practice, a threshold value to divide the patient population into biomarker ‘positive’ and ‘negative’ is desirable. Early phase clinical trials are increasingly using biomarkers for patient selection, but at this stage, it is likely that little will be known about the relationship between the biomarker and the treatment outcome. We describe a single-arm trial design with adaptive enrichment, which can increase power to demonstrate efficacy within a patient subpopulation, the parameters of which are also estimated. Our design enables us to learn about the biomarker and optimally adjust the threshold during the study, using a combination of generalised linear modelling and Bayesian prediction. At the final analysis, a binomial exact test is carried out, allowing the hypothesis that ‘no population subset exists in which the novel treatment has a desirable response rate’ to be tested. Through extensive simulations, we are able to show increased power over fixed threshold methods in many situations without increasing the type-I error rate. We also show that estimates of the threshold, which defines the population subset, are unbiased and often more precise than those from fixed threshold studies. We provide an example of the method applied (retrospectively) to publically available data from a study of the use of tamoxifen after mastectomy by the German Breast Study Group, where progesterone receptor is the biomarker of interest.

Cumulative use of salivary markers with an adaptive design improves detection of periodontal disease over fixed biomarker thresholds

“…Adaptive-threshold designs have gained interest in various fields of medicine as they may bring on increased power over fixed thresholds [17]. In the limits of the present study, we conclude that the cumulative use of salivary P. gingivalis, IL-1b, and MMP-8 levels with an adaptive design brings additional power over fixed thresholds of single biomarkers in detecting periodontitis.…”

“…The CRS method supports the idea of assigning study population-based thresholds with a more personalized approach, instead of using universal thresholds [17]. It was observed that, there are four-to nine-fold differences between the levels of given thresholds in studies using different methods for these biomarker determinations [10][11][12][13][14].…”

Point estimation following two‐stage adaptive threshold enrichment clinical trials

“…Acknowledging that different patients may require different care has led to trial designs that incorporate assessment of treatment effects in different subsets of the population. Most statistical methodologies for such designs focus on hypothesis testing . In this paper, we have considered point estimation following an adaptive threshold enrichment clinical trial.…”

Continuous threshold models with two‐way interactions in survival analysis

“…They utilize the Bayesian method to estimate the threshold of a biomarker which is not normally distributed but follows a Student‐ t distribution and a mixture of Dirichlet processes. Spencer et al (2016) create a design which can learn from the biomarker and make adjustments on the threshold optimally using the combination of Bayesian prediction and the generalized linear model. This design aims to figure out the subpopulation for which the treatment is effective and estimate the threshold parameter.…”