Metabolomics in rheumatic diseases: The potential of an emerging methodology for improved patient diagnosis, prognosis, and treatment efficacy

Abstract

Metabolomics belongs to the family of “-omics” sciences, also comprised of genomics, transcriptomics, and proteomics, all of which share the advantage of a non-targeted approach for identifying biomarkers and profiling the patient. This means that they do not require a preliminary knowledge of the substances to be studied. Moreover, even small quantities of biological fluids or tissues may be utilized for analysis. Metabolomic procedure has become feasible only recently with the advent and accessibility of new high-throughput technologies, including mass spectrometry and nuclear magnetic resonance. The methodology generally involves three defining steps: 1) the acquisition of experimental data, 2) the multivariate statistical analysis, and 3) the projection of the acquired information (profiles) to construct the patient map. Metabolomic analysis has been applied to several disorders: as far as rheumatic diseases are concerned, a few studies have focused on rheumatoid arthritis, spondyloarthritis, systemic lupus erythematosus, and osteoarthritis. Both murine models and clinical data have shown the potential of this novel tool to contribute to deciding a diagnosis, discriminate between patients based on disease activity, and even predict the response to a particular treatment. The present review fully reports these findings and offers a critical view of the challenges still to be met.

Introduction

Metabolomics is a novel tool for studying biological systems that allows for the analysis of the specific response of organisms to environmental stimuli. This new approach belongs to the family of “-omics” sciences, which is also comprised of genomics, transcriptomics, and proteomics, and is appealingly characterized by being “data driven” without using predetermined models. The starting point is experimental data arranged in large matrices by the use of multivariate statistical analysis. Thus, it is possible to study the correlation between the levels of the metabolites to directly derive a profile or signature of the specific disease.

Through the study of metabolomics it is possible to define the current status of the system, as a result of environmental pressure and genetic potential, that is not otherwise predictable by genomic and/or proteomic characterizations. For this reason, regardless of the sequential relationship in which the “-omics” are introduced (Fig. 1), metabolomics provides information directly in line with the new approach of Systems Biology that considers living systems as dynamic and complex and assumes that their behavior originates from interactions such that it is difficult to make predictions when exclusively considering the properties of individual parts [1], [2], [3]. This approach can drive the diagnostic algorithm of several pathologies, including chronic rheumatic diseases. This is a crucial point because, presently, most of them do not dispose of specific diagnostic biomarkers and therefore, metabolomics might provide the opportunity to map the patient on their specific pathologic pathway. The fundamental rationale in metabolomics is that perturbations caused by a disease in a biological system will lead to correlated changes in concentrations of certain metabolites. While in some cases, for example congenital metabolic diseases, it may be possible to identify a single, robust diagnostic metabolite, there are many others (including rheumatic disorders) where the perturbations involve the activation of multiple pathways. In such cases, by using multivariate statistics on descriptive experimental profiles obtained by Nuclear Magnetic Resonance (NMR) or Mass Spectrometry (MS), it may be possible to describe patterns of changes and biomarkers that are highly discriminatory for the perturbation and/or disease state [4], [5].

Furthermore, the metabolomic approach may also be used to map the patients over time by simply using the spectroscopic profile and with no need to identify specific substances: one spectrum is associated to one system condition. Table 1 provides a glossary for the most frequently used words in this field.

Herein, we report an overview of metabolomic methodology detailing the three main steps: 1) the acquisition of experimental data by spectrometric/spectroscopic techniques, 2) multivariate statistical analysis, and 3) the projection of the acquired information to construct the patient map (Fig. 2). We also assemble findings obtained to date on relevant rheumatic diseases by means of this technique. Finally, we provide information on metabolomic indicators in the main rheumatic diseases, discuss the unresolved questions and enlighten upon the potential of metabolomics in this field.