PS02.161: METABOLIC MARKERS OF OESOPHAGEAL SQUAMOUS CELL CARCINOMA: A SYSTEMATIC REVIEW AND POOLED ANALYSIS

Background Oesophageal squamous cell carcinoma (ESCC) is a major global health burden associated with poor survival. Patients suffering from this condition typically present at an advanced stage due to non-specific early symptoms. An increasing emphasis on early detection has led to the investigation of novel biomarkers of ESCC. This systematic review aims to define existing biomarkers of ESCC and to identify perturbations in associated metabolic pathways. Methods A systematic online literature search (1946–January 2018) was performed to identify studies reporting metabolic biomarkers of ESCC. Pooled point estimate of the hierarchal summary ROC curve of analysed metabolites was performed using bivariate meta-analyses. Conduct of this review was in accordance with the recommendations of the Cochrane Library and MOOSE guidelines. Results An online search identified 628 potential relevant studies, of which eight (including a total 508 patients) met inclusion criteria. All publications represented phase-I biomarker discovery studies. Methods used to assess metabolites in patients with ESCC included liquid chromatography-mass spectrometry (n = 6), gas chromatography-mass spectrometry (n = 1) and 1H-nuclear magnetic resonance (n = 1). Sample types utilised were plasma (n = 5), tissue (n = 2) and urine (n = 1). A total of 1670 metabolites were identified as being associated with ESCC. Key metabolic pathways involved included tricarboxylic acid cycle and in pathways of oxidative stress. Metabolites identified increased in ESCC are breakdown products of fatty acid synthesis and glycerophospholipid metabolism e.g.: palmitic acid, oleic acid, LysoPC(24:0), LysoPC(18:2), L-carnitine, branched chain amino acids e.g.: valine, leucine, isoleucine and phenylalanine. Metabolites decreased in ESCC are breakdown products of amino acid metabolism e.g: glutamine, carbohydrates e.g.: glucose and organic acids e.g.: α-ketoglutaric acid oxime. Pooled analysis of findings from five studies showed an area under the receiver-operating characteristic analysis curve of 0.96, sensitivity 88.0% (95% CI 80.8%-92.7%) and specificity 93.5% (95% CI 96.9%- 86.8%). Conclusion This is the first review to define metabolic biomarkers associated with ESCC. Significant variability in identified metabolites is evidence of the broad range of analytical techniques employed and the diversity of metabolic pathways associated with this disease. Notwithstanding, findings provide important insight in to those metabolites associated with ESCC and provide a basis for future studies. Disclosure All authors have declared no conflicts of interest.