Tissue imaging and serum lipidomic profiling for screening potential biomarkers of thyroid tumors by matrix-assisted laser desorption/ionization-Fourier transform ion cyclotron resonance mass spectrometry

Changes in serum lipidome and in tissue lipidome are associated with cancer. In this study, tissue mass spectrometry imaging (MSI) and serum lipid profiling by matrix-assisted laser desorption/ionization-Fourier transform ion cyclotron resonance mass spectrometry (MALDI-FTICR MS) were performed to investigate significantly changed lipids in both tumor (malignant thyroid cancer (MTC) and benign thyroid tumor (BTT)) tissues and sera. Y-scatterplots of variable importance in the projection (VIP) values vs. fold change values indicate that change trends in the levels of ten lipids (i.e., phosphatidylcholine (PC)(34:1), PC(36:1), PC(38:6), phosphatidic acid (PA) (36:2), PA(36:3), PA(38:3), PA(38:4), PA(38:5), PA(40:5), and sphingomyelin (SM)(34:1)) in both tissues and sera from MTC patients, BTT patients, and normal individuals are significantly associated with these three types of pathophysiological status. In order to examine their diagnostic ability, 289 serum samples from 124 MTC patients, 43 BTT patients, and 122 normal controls were randomly divided into the training set and validation set. A biomarker of PC(34:1) exhibited excellent diagnostic ability to differentiate both MTC and BTT patients from normal individuals, with an area under the receiver operating characteristic (ROC) curve value of 0.984, a sensitivity of 96.4 %, and a specificity of 92.7 %. A panel which included PA(36:3) and SM(34:1) could distinguish between MTC and BTT, with an area under receiver operating characteristic curve (AUC) of 0.961, a sensitivity of 87.8 %, and a specificity of 92.9 %. It is worth noting that a panel consisting of PC(34:1), PA(36:3), and SM(34:1) could differentiate MTC patients from both BTT patients and normal individuals, with an AUC of 0.841, a sensitivity of 86.6 %, and a specificity of 75.5 %.