Rédiger un PREreview

Primary resistance to anti-PD-1 therapy in metastatic melanoma remains a clinical challenge. This study reanalyzed the GSE168204 dataset to elucidate molecular mechanisms of resistance and response, incorporating batch correction to address limitations in our prior preprint[6]. RNA-seq data from 25 melanoma biopsies (9 responders, 16 non-responders) were analyzed using DESeq2 with surrogate variable analysis[9,10]. We identified 3,247 differentially expressed genes, revealing a "cell cycle shield" signature in non-responders characterized by upregulation of CDK1, CCNB1, E2F1, and HSP90AA1 enriched for proliferation and DNA repair pathways, suggesting immune evasion through rapid tumor growth. Responders exhibited upregulation of EP300, CREBBP, FCGR2B, and histone genes enriched for chromatin organization and systemic lupus erythematosus pathways, indicating immune activation and autoimmune-like transcriptional programs. Notably, batch correction reversed the roles of EP300 and FCGR2B from non-responders to responders[6], highlighting their context-dependent functions in immune engagement. The "cell cycle shield" suggests targeting CDK1 or HSP90AA1 to overcome resistance[13,14], while the SLE signature may serve as a response biomarker reflecting immune activation states[7]. Validation in larger cohorts and experimental models is needed to translate these findings into personalized immunotherapy strategies.