Modeling clinical outcome using multiple correlated functional biomarkers: A Bayesian approach

Abstract

In some biomedical studies, biomarkers are measured repeatedly along some spatial structure or over time and are subject to measurement error. In these studies, it is often of interest to evaluate associations between a clinical endpoint and these biomarkers (also known as functional biomarkers). There are potentially two levels of correlation in such data, namely, between repeated measurements of a biomarker from the same subject and between multiple biomarkers from the same subject; none of the existing methods accounts for correlation between multiple functional biomarkers. We propose a Bayesian approach to model a clinical outcome of interest (e.g., risk for colorectal cancer) in the presence of multiple functional biomarkers while accounting for potential correlation. Our simulations show that the proposed approach achieves good performance in finite samples under various settings. In the presence of substantial or moderate correlation, the proposed approach outperforms an existing approach that does not account for correlation. The proposed approach is applied to a study of biomarkers of risk for colorectal neoplasms and our results show that the risk for colorectal cancer is associated with two functional biomarkers, APC and TGF-α, in particular, with their values in the region between the proliferating zone and the differentiating zone of colorectal crypts.

Potentials and pitfalls of clinical peptidomics and metabolomics

“…Human perception of complex coherences may be very limited, and for the acceptance of peptidomic or metabolomic data at the patient's bedside it seems essential to provide information condensed enough to enable a prima vista understanding of a marker model, as well as simple application for the clinician. For these "final" interpretation steps, new statistical concepts and computational options are emerging: For the modelling of clinical outcome on the basis of correlated functional biomarkers, a new Bayesian approach Review article: Medical intelligence was suggested by Long et al [78]. For the incorporation of replicate information into correlation estimates, Zhu et al [66] proposed the R-based CORREP package, which can be applied not only to genetic data, but also to metabolite data [65].…”

Modeling biomarker ratios with gamma distributed components

“…A more general model for the GB2 distribution was proposed by Yee (2015) as part of the VGAM framework. Regarding the ratio of correlated gamma distributed variables, which is of particular interest in biomarker research (Long et al, 2016), no regression modeling strategy for E[U/V |X] exists (to the best of our knowledge). In fact, although there are several well-established results on the probability density function (p.d.f.)…”