Serum Metabolite Profile By Targeted Lipomics Approach to Evaluate Drug Resistance and Prognosis in Myeloma Patients

ObjectiveMultiple Myeloma (MM) is a plasma cell tumor that is prone to occur in elderly people. Drug resistance and repeated recurrence are important factors affecting the survival of MM patients, but the mechanism is unclear. Aims To systematic study the difference of metabolites before and after treatment in MM, patients through targeted ultra-high performance liquid chromatography tandem mass spectrometry (UPLC-MS), in order to analyze abnormal metabolites related to MM treatment response, and prognosis, and provide ideas for precise treatment of MM. Methods After received bortezomib (BTZ) / daratumumab (Dara)-based regimens, 46 cases of MM patients were divided into responding MM (rMM) and non- responding MM (nrMM) group. The serum lipid metabolites of 46 pairs of MM patients and healthy controls (HC) before and after treatment were detected, which were associated with the main clinical features of M protein, symptoms. The serum fatty acid transposase (CD36) and hypoxia inducible factor-1α (HIF-1α) were measured using ELISA. Result sBetween MM and HC groups, 121 differential metabolites were screened out, which were mainly enriched in sphingolipid, linoleic acid and glycerophospholipid metabolism pathways. There were 32 common differential metabolites associated with IgG and IgA type M proteins，including 4 kinds of cholesterol ester (CE), 2 kinds of ceramide (Cer), 3 kinds of fatty acid (FA), hexosylceramide (HexCer), lactosylceramides (LacCer) (d18:2/16:0), 2 kinds of lysophosphatidylcholine (LysoPC) (20:4), lysophosphatidylethanolamine (LysoPE) (18:1), phosphatidylethanolamine (PE) (O-16:0/22:6), 8 kinds of phosphatidylcholine (PC), 2 kinds of sphingomyelin (SM) and 7 kinds of triglycerides (TG). Among them, 5 metabolites including CE (20:3), PE (O-16:0/22:6), LacCer (d18:2/16:0), LysoPC (20:4) and TG (16:0/18:1/18:1) were associated with at least one of the symptoms such as hypercalcemia, renal insufficiency and anemia symptoms. These differential metabolites may be associated with the activity or pathogenesis of MM. Of 32 differential metabolites, 15 metabolites including 4 kinds of CE, LacCer[d18:2/16:0], LysoPE[18:1], PE[O-16:0/22:6], 2 kinds of PC and 6 kinds of TG were showed statistical differences (P < 0.050) between MM and rMM groups, which might be used to evaluate the treatment response with high sensitivity and specificity (AUC > 0.7, P < 0.050). Eleven differential metabolites including LacCer [d18:2/16:0], LysoPE [18:1], Cer (d16:1/22:0), 6 kinds of PC and 2 kinds of SM showed statistical differences (P < 0.050) between rMM and nrMM groups, which can be used to evaluate the resistance of RRMM patients to BTZ/Dara with high sensitivity and specificity (AUC>0.7, P < 0.050). Among them, fifth differential metabolites including LacCer[d18:2/16:0], LysoPE[18:1], PC[O-18:0/22:4] and 2 kinds of TG may be related resistance to Dara-based regimens. Survival analysis displayed that the decrease in LacCer (d18:2/16:0), SM (d18:2/14:0), and SM (d18:1/20:1) was correlated with poor OS (P < 0.050), so as the reduction in HexCer (d16:1/22:0), LacCer (d18:2/16:0), LysoPE (18:1), PC (O-18:0/18:1), PC (P-16:0/16:1), SM (d18:2/14:0), and SM (d18:1/20:1) with inferior PFS (P < 0.050). Conclusion The above screened metabolites maybe serve as biomarkers for evaluating disease activity, treatment response, and drug resistance. Its related antagonists may provide important evidence for metabolic targeted therapy and recovery of treatment sensitivity in MM in the future. Keywords Multiple myeloma; Lipomics; Activity; Treatment response; Drug resistance; Prognosis.