Lipidomic profiling as a biomarker for prostate cancer diagnosis and response to enzalutamide (Enza).
Sara Bleve,Francesco Ravera,Silvia Rodrigues,Mohamed Omar,Emilio Francesco Giunta,Filippo Pederzoli,Hubert Pakula,Amelia Altavilla,Nicole Brighi,Giorgia Gurioli,Maria Concetta Cursano,Chiara Casadei,Cristian Lolli,Giuseppe Schepisi,David M. Nanus,Pier Vitale Nuzzo,Ugo De Giorgi,Massimo Loda
DOI: https://doi.org/10.1200/jco.2024.42.4_suppl.217
IF: 45.3
2024-02-01
Journal of Clinical Oncology
Abstract:217 Background: Metabolic rewiring, particularly reprogramming of lipid metabolism, is a hallmark of prostate cancer (PC). Upregulated lipogenesis plays a paramount role in PC pathogenesis, progression, and treatment resistance, especially in therapies targeting the androgen receptor (AR) pathway. Thus, we aimed to study whether lipidomic profiling can distinguish men with PC from those with benign prostate lesions (BL), as well as predict the response to Enza in metastatic castration-resistant PC (mCRPC). Methods: Plasma samples were collected from 14 men with BL and 13 pts with localized hormone-sensitive PC prior to the tissue biopsy, and 54 with mCRPC prior to Enza as first-line treatment. Plasma lipidomic profiling was performed by liquid chromatography and electrospray ionization-tandem mass spectrometry (LC-MS/MS). Lipids were analyzed with an untargeted approach using MultiQuantTM 3.0.3. Multiple comparisons were made by two-way ANOVA with Tukey’s multiple comparison test. A machine learning approach was used to assess the correlation between the lipid profile and the prediction of radiographic progression-free survival (rPFS) defined based on RECIST 1.1/PCWG-3 criteria. Kaplan-Meier survival curves were generated for each selected lipid species independently associated with rPFS adjusting for clinicopathological factors by a multivariate Cox regression analysis. Results: LC-MS/MS quantified a total of 1839 plasma lipids species comprising 17 classes. Ten lipid classes showed a higher expression in PC compared to BL, and only Phosphatidylethanolamine (PE) was significantly different between the two groups (p < 0.05). At the level of lipid species, 69 were found to be significantly dysregulated in pts with PC compared to pts with BL: 26 species belonging to Ceramides, Phosphatidylethanolamine, Phosphatidylglycerol, Phosphatidylinositol were upregulated, whereas 43 belonging to Cholesterol esters, Diacylglycerides, Phosphatidylcholine, and Triacylglycerides were downregulated. After a median follow-up time of 27.9 months, 24/54 (44%) pts with mCRPC progressed on Enza. Median progression-free survival was 44.1 months (95% CI, 41.1 -NA months). The baseline levels of 10 lipids were individually significantly associated with rPFS. Two lipids belonging to the phosphatidylinositol (PI) class, PI.14.0_16.0 and PI.20.3_20.3, were independent predictors of rPFS (p < 0.05) when they were modeled with PS ECOG score, BMI, baseline PSA levels, Gleason score, and the number of metastatic sites in a multivariable Cox regression. The results highlight the potential role of the PI class in modulating the AR pathway. Conclusions: Lipidomic profiling can differentiate between BL and PC. Furthermore, lipidomic profiling correlates with response to Enza in men with mCRPC. Further studies are needed to externally validate the diagnostic and predictive role of lipid signatures in PC.
oncology