Abstract 1108: Tissue and circulating DNA methylation profiling can recapitulate SCLC heterogeneity to enable identification of new biomarkers of clinical relevance

Abstract Background: Small-cell lung cancer (SCLC), an aggressive lung cancer type, presents challenges in management due to limited treatment options. In the CASPIAN study (NCT03043872) assessing durvalumab (D) ± tremelimumab (T) and platinum-based chemotherapy in extensive-stage SCLC, improved survival with D + chemotherapy was revealed. Ongoing research exploring SCLC subtypes, treatment responses, epigenetic modifications like DNA methylation, and association with treatment outcomes, is offering insights into tailored treatment and disease progression. Recent studies highlighted the promise of liquid biopsies for sensitive detection and molecular subtyping of SCLC, but the lack of a comprehensive investigation comparing matched tissue and liquid biopsy in SCLC limits a thorough assessment of their effectiveness. Methods: DNA was extracted from 175 FFPE samples in CASPIAN and TWIST methylome panel analysis, which covers 3.98M CpG sites, was performed to characterize their methylation profiles at promoters and CpG islands. DNA methylation patterns were compared with corresponding RNA-seq gene expression data, and association of methylation with somatic alterations and expression of key SCLC genes was evaluated. A total of 146 CASPIAN serum samples were used to capture ctDNA methylation profiles with T7-MBD-seq methodology. Results: We uncovered 4 major tissue methylation-based clusters through unsupervised clustering of the top 5% most variable promoter methylation regions. Three clusters were enriched with neuroendocrine (NE) features while one was non-NE driven. Notably, these clusters showed high correlation with RNA-based subtypes and offered distinct insights into clinical outcomes. By integrating promoter methylation levels with other multi-omics datasets, we revealed biological features of each subtype and discovered novel methylation biomarkers, such as ZNF717 and REC8, which were negatively associated with corresponding gene expressions, suggesting that methylation-based profiling could enhance and even expand the current heterogeneity characterization. Other than tissue methylation, ctDNA tumor fractions predicted using copy number alterations served as a prognostic signature for patient outcomes in SCLC. High concordance between ctDNA and tissue methylation provided more confidence in personalized treatment opportunities based on SCLC heterogeneity for this otherwise difficult-to-treat malignancy. Conclusions: Tissue methylation-based heterogeneity can reveal clinically relevant biomarkers for SCLC, and these clusters are crucial for understanding disease mechanisms and enabling personalized treatment. Additionally, liquid biopsies offer a non-invasive method for subtyping SCLC, potentially improving personalized treatment feasibility. Citation Format: Siyu Liu, Mingchao Xie, Alexandra Clipson, Simon Pearce, Brychan Manry, Steven Criscione, Megan Callahan, Priti Chugh, Annabel Peachey, Ana Ortega-Franco, Steven Hill, Florent Mouliere, Dominic Rothwell, James Hadfield, Caroline Dive, Yashaswi Shrestha. Tissue and circulating DNA methylation profiling can recapitulate SCLC heterogeneity to enable identification of new biomarkers of clinical relevance [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 1108.