Rédiger un PREreview

Altered lipid metabolism is an emerging hallmark of cancer. In glioblastoma, sufficient lipids are key to driving rapid cancer proliferation. SREBF1, a key regulator of fatty acid synthesis, recently gained therapeutic interest. SREBF1 methylation is implicated in metabolic diseases with currently unknown effects in cancer, including glioblastoma. This study characterizes SREBF1 methylation in two independent, population-scale glioblastoma cohorts. Unsupervised clustering of SREBF1 methylation sites reveals two predominant clusters with contrasting levels of cancer stemness (P < 0.01). The high-stemness cluster is younger in epigenetic age (P < 0.05) and higher in the neuronal-stromal composition (P < 0.05). Independent of age, sex, and MGMT subtype, the genome-wide epigenetic landscapes of the clusters were distinct at the global (P < 5e-8) and single-nucleotide resolution (all Bonferroni P < 0.05). Genes involved in neuronal specification and organization were hypo-methylated in the high-stemness cluster (all Bonferroni P < 0.05), which also showed worse overall survival (hazard ratios ≥ 1.25). Taken together, this study offered novel insights into SREBF1 methylation, warranting future research into the epigenetics of the lipid metabolism machinery.