Multimodal (Bio)Markers and Risk of Obesity – A Comprehensive Scoping Review

TL;DR

This review explores combining multiple biomarkers and factors to better predict and prevent obesity, emphasizing the need for a comprehensive approach.

Contribution

The paper provides a comprehensive scoping review on multimodal biomarkers for obesity risk prediction.

Findings

Obesity risk prediction requires a multimodal approach integrating biomarkers from multiple domains.

Current research highlights the feasibility of using multiomics and behavioral data for risk stratification.

Machine learning and AI are increasingly used to interpret complex obesity-related data.

Abstract

Obesity has been associated with several chronic diseases, especially noncommunicable ones and related comorbidities. Despite international efforts to decrease the prevalence of obesity, the number of persons struggling with this ailment is not decreasing. An important aspect is obesity prevention, including the early detection of the risk, i.e. whether an individual is likely to develop obesity, to allow for early risk stratification and countermeasure initiation. However, obesity is a complex and multifactorial complication, and many factors appear to play a role, including age, sex, diet, physical activity (PA), psychological and emotional status, genetic make-up, epigenetics, and gut microbiota. One isolated biomarker, therefore, could not enable optimal risk stratification and prognosis for the individual; rather, a combined set or multimodal approach to tackle risk prediction is demanded. Such a multimodal interpretation would integrate biomarkers from various domains, such as more classical markers (insulin, leptin), multiomics (e.g. genetics, epigenomics, transcriptomics, proteomics, and metabolomics), behavioral attributes (dietary, PA, and sleep patterns, and smoking status), psychological traits (mental health status, depression, and eating disorders), and gut–microbiota (composition and diversity) into a combined interpretation, also employing more advanced interpretation tools, such as machine learning and artificial intelligence. In this scoping review, we aimed to summarize the current state of the art in this area, highlighting the progress and novel approaches in combating obesity, and focusing on the feasibility and effectiveness of such biomarkers and their application within clinical trials. In addition, we outline potential future steps and recommendations for future approaches.

Figures