Longitudinal Plasma Lipidomics Reveals Distinct Signatures Following Surgery in Patients with Glioblastoma

Background: Glioblastoma is an aggressive brain tumor that invariably recurs despite treatment, partly due to metabolic adaptations, including altered lipid metabolism. This study investigates plasma lipidomic profiles in patients with glioblastoma to explore their potential as a liquid biopsy for disease monitoring. Methods: Plasma samples were collected from 36 patients with histopathologically confirmed IDH wild-type glioblastoma at four treatment stages: pre-surgery (n = 36), post-surgery (n = 32), pre-radiation (n = 28), and post-radiation (n = 17). Untargeted lipidomics analysis was performed using liquid chromatography-high resolution mass spectrometry (LC-HR-MS/MS). Results: Plasma lipidomic signatures differed significantly across treatment stages. Specifically, the lipidomic profile prior to surgery was statistically distinct from those at subsequent stages, demonstrating an increased compound abundance of numerous lipids that are decreased at subsequent stages, including linoleic acid (fold-change 2.58, p = 4.21 × 10−11), behenic acid (fold-change 2.09, p = 9.3 × 10−10), and linolenic acid (fold-change 4.44, p = 5.83 × 10−6). Random forest modeling could predict pre-surgical samples with 85.7% accuracy. Conclusions: Plasma lipidomics shows promise as a potential liquid biopsy approach for monitoring glioblastoma treatment but future studies will need to examine these findings in larger and well-controlled cohorts.