Detecting NET using Methylation-based biomarkers and the novel IMPRESS technology (A16)

Introduction: Due to current limitations, diagnosis of neuroendocrine tumors (NETs) is frequently delayed. Addressing this issue is crucial, necessitating the identification of new, adequate biomarkers. Our previous work highlighted the methylome as a promising source of biomarkers, enabling differentiation of NET vs. non-NET samples, and distinguishing based on tissue of origin (TOO).

Aim(s): We aim to (i) create a biomarker panel of general NET and tissue-specific differentially methylated CpGs (DMCs) and (ii) validate it in fresh frozen tissue from pancreatic (PNET), lung (LNET) and small intestine (SINET) NETs using our IMPRESS (Improved Methylation Profiling using Restriction Enzymes and smMIP Sequencing) technology.

Materials and methods: General NET and tissue-specific DMCs were identified via two differential methylation analyses. DMCs located in restriction sites and with a minimum methylation difference of 0.16 were selected for IMPRESS. This technology combines methylation-sensitive restriction enzymes (MSREs) and single-molecule Molecular Inversion Probes (smMIPs) to detect hypermethylated CpGs. For all selected DMCs, smMIPs were designed with MIPGEN software. smMIPs are now being validated in 23 PNET, 63 LNET, 8 SINET and 60 normal adjacent tissue samples.

Results: Our biomarker panel comprises 2304 DMCs (1152 general NET and 384 DMCs for each tissue type included). Validation of our panel is ongoing to define DMCs with the highest discriminatory efficiency to predict the presence of a NET and TOO using a classifier model. The results of ongoing analyses will be shown at the ENETS conference.

Conclusion: We have set up a biomarker panel that is now being validated in tissue using IMPRESS and are building the adjoint classifier model.

Keywords: neuroendocrine tumor, dna methylation, biomarker selection, impress technology