EpigenOMic Determinants of the Neuroendocrine Phenotype As Biomarkers for Noninvasive Diagnosis of Neuroendocrine Neoplasms

GEP-NENs are a heterogeneous group of diseases that encompasses relatively indolent and more aggressive tumors, sometimes with mixed exocrine/endocrine components (MINENs). Due to persistent uncertainties in diagnosis and treatment, prognosis of the mixed and more aggressive forms remains poor. Non-invasive tests would enable timely identification and monitoring over time. A better understanding of their biology and of the phenotypic differentiation process may pave the way for novel therapies. Differentiation follows, and can be inferred from, changes in the epigenetic landscape that shape transcriptional programs. Among epigenetic markers, DNA methylation is particularly well suited for non-invasive detection as it can be measured in circulating tumor DNA (ctDNA). New technologies like Oxford Nanopore Technologies (ONT) enable simultaneous analysis of DNA sequence and methylation, as we recently showed by Magi et al Nat Comms Biol 2022. Previous studies identified distinct neuroendocrine epi-transcriptomic landscapes but were conducted on tumor specimens (Yachida et al Cancer Discov 2022), impeding the discrimination of signals specifically generated within neuroendocrine tumor cells from the noise due to surrounding nontumoral or exocrine tumoral cells. This is now potentially solved by methods allowing single-cell sequencing directly from frozen or paraffin-embedded patient samples.

The main expected outcome is the development of a novel liquid biopsy assay for the detection and monitoring of GEP-NENs and MiNENs. To this end, we will exploit the ability of Oxford Nanopore Technologies (ONT) sequencing to simultaneously yield sequencing and methylated DNA profiles. Importantly, assessment of cell of origin from ctDNA requires the comparison of methylation and fragmentomics signals with previously generated maps of reference cells and tissues (Katsman et al 2022 Genome Biology). For rare tumors like NENs, these maps are not available in the literature; the few studies that have characterized the epigenomic features of NENs are likely contaminated by the surrounding stroma, so epigenetic signals (including both DNA methylation and tumor-specific transcription factor binding sites, essential for fragmentomics) may be uninformative for detection in the blood.