Serum Free Unsaturated Fatty Acids: Biomarkers Of Early Detection Of Chronic Diseases

Background and Objectives:Chronic diseases have become the main cause of death worldwide, which serious threat to people’s life and health. Early detection is one of the most effective ways to improve the survival rate and prognosis. Metabolomics combines with advanced statistical analysis method are now widely used for identifying biomarkers related to diseases. Monitoring fluctuations of certain metabolites in biofluids is useful for early screening and detection diseases. In this study, we employed FTICR MS to quantify the levels of human serum free fatty acids (FFAs) with different diseases. The SAS and SPSS software were performed to explore FFAs biomarkers or biomarker panels for early detection of diseases.Methods:First, we used MALDI-FTICR MS in negative mode to detect serum FFAs from 339 samples, including 161 healthy controls,118 patients with hyperglycemia and 60 patients without hyperglycemia. To make sure the data quality, multiple point internal standard calibration curves and stability were measured. Base on the data, we next constructed chip-based direct-infusion nanoESI-FTICR MS (CBIDnanoESI(-)-FTICR MS) to measure the unsaturated fatty aicds from 2861 serum samples, including lung diseases, breast diseases, colorectal diseases, gastric diseases, pancreatic diseases, thyroid diseases and healthy controls. The reliability for the FFAs analysis was validated through its linearity, limit of detection (LOD), stability, precision and spike-and-recovery. The subjects were randomly assigned to the training or validation set. In addition, we also investigated the effect of icotinib hydrochloride on FFAs levels during the treatment for 8 advanced-stage LC patients. The Mann-Whitney U test was performed to compare FFAs between different groups. The FFAs with levels exhibited significant differences were submitted to receiver operating characteristic (ROC) curves analysis to assess the diagnostic accuracy.Results:The results of hyperglycemia patients indicate that the levels of serum FFAs are closely associated with diabetes. In this study, the levels of C16:0, C18:3, C18:2, C18:1, C18:0, C20:4 and C22:6 in hyperglycemia patients are significantly higher than those in healthy controls, and most of them displayed an increased tendency along with the elevation of fasting blood glucose (FBG) levels in hyperglycemic patients. The ROC curves show that the panel including C18:3, C18:2, C18:1, C20:4 and C22:6 has excellent diagnostic performance for distinguishing healthy subjects from the patients with hyperglycemia or patients without hyperglycemia, with AUC> 0.97, sensitivity> 90% and specificity> 91%. Concerning the cancer or benign diseases patients study, the levels of serum unsaturated FFAs were presented down-regulation in cancer patients in both training and validation set. ROC results indicate that FFAs panels have high diagnostic accuracy for discriminating cancer or benign diseases, and show potential for discriminating early-stage cancers. For example, a combination of C18:2/C18:1 and C18:3/C18:1 is able to discriminate early-stage pancreatic cancer (AUC of 0.912, sensitivity of 86.7% and specificity of 88.6%) from non-cancer subjects (pancreatitis plus healthy controls), a combination of C16:1, C18:3, C18:2, C18:1, C20:4 and C22:6 displayed high diagnostic performance for distinguishing early-stage lung cancer from non-cancer subjects (benign lung diseases plus healthy controls), with AUC of 0.933, sensitivity of 84.2% and specificity of 89.1%, and a combination of C16:1, C18:3, C18:2, C20:4 and C22:6 is able to discriminate early-stage gastric cancer, colorectal cancer and breast cancer, with AUC> 0.82, sensitivity> 80% and specificity> 72%, which are greatly higher than those of clinically used serum biomarker CA 19-9 and CEA. During the investigation about the effect of icotinib hydrochloride on FFAs levels, we found significant changes in monounsaturated fatty acids (MUFA) and polyunsaturated fatty acids (PUFA) levels along with post-dose time points (weeks), which might predict the disease progression in advance.Conclusions:In this study, we employed high-sensitivity and high-throughput MALDI/NanoESI-FTICR MS to analyze complex biological samples without cumbersome preparation steps. FFAs panels could be promising novel biomarkers for early detection of cancer, with excellent diagnostic performance. These panels could be new biomarkers for monitoring the therapeutic efficacy of icotinib and predicting the progression of the disease. Therefore, it is a promising strategy for screening chronic diseases by its early detection and monitoring the efficiency of medication.