This article was originally published here
Pharmaceutical statistics. December 21, 2021. doi: 10.1002 / pst.2187. Online ahead of print.
New technologies for new biomarkers have transformed the field of precision medicine. However, in applications such as liquid biopsy for the early detection of tumors, the misclassification rates of next-generation sequencing and other technologies have become an inevitable feature of biomarker development. Since initial experiments are usually limited to technological choices and specific application parameters, a statistical method that can project performance measures of other scenarios with different rates of misclassification would be very useful for planning biomarker development. and future testing. In this paper, we describe an approach based on an extended version of simulation extrapolation (SIMEX) to project the performance of measured biomarkers with varying misclassification rates due to different technological or application parameters when experimental results fail. are available only from a specific setting. Through simulation studies for logistic regression and proportional hazards models, we show that our proposed method can be used to project biomarker performance with good precision when switching from one technology to another or from one technology to another. framework of application. Similar to the original SIMEX model, the proposed method can be implemented with existing software in a simple way. An example of data analysis is also presented using a lung cancer data set and performance measures for two biomarkers based on a panel of genes. The results demonstrate that it is possible to infer the potential implications of using a range of technologies or application scenarios for biomarkers with limited human trial data.