Computed tomography-derived body composition analysis in patients with advanced cancer: clinical utility and future...

With weight loss increasingly occurring against a background of obesity across a variety of advanced cancers, there has been increasing interest in computed tomography (CT)-derived body composition analysis. Various imaging software packages and thresholds are commonly in use in CT-derived body composition analysis. This review discusses the current research in field of body composition with emphasis on the information required for such measurements to be taken into routine clinical practice.

CT is widely used for tumour staging in patients with cancer. Over the last decade, this imaging modality has been exploited to make measurements of body composition. Using a common landmark (L3) several different thresholds have been developed to stratify adipose and muscle tissue areas according to age, sex and BMI and their relationship with survival. A significant relationship between CT-derived body composition and clinical outcomes has been shown in different tumour types and geographical locations. However, there is considerable variation with methods, thresholds and muscle groups used for analysis. Therefore, there would appear to be a need to develop reliable methodology and population-specific reference ranges to guide clinical interpretation and enable routine clinical use.

There has been an explosion of clinical research interest in CT-derived body composition analysis. Such body composition analysis provides important host phenotype information which has prognostic value. For CT-derived body composition to be fit for use in routine clinical practice, there is need for universally accepted terminology, software, muscle group selection, prognostic thresholds to standardize such body composition analysis.