DNA methylation-based prediction of response to immune checkpoint inhibition in metastatic melanoma

A novel approach of reference-free deconvolution of large-scale DNA methylation data enabled to develop a machine learning classifier based on CpG sites, specific for Latent Methylation Components (LMC), that allowed for patient allocation to prognostic clusters. DNA methylation data was processed using reference-free analyses (MeDeCom) and reference-based computational tumor deconvolution (MethylCIBERSORT, LUMP). These results demonstrate that LMC-based segregation of large-scale DNA methylation data is a promising tool for classifier development and treatment response estimation in cancer patients under targeted immunotherapy. 65 specimens of melanoma skin metastasis were analyzed using the Illumina Infinium MethylationEPIC BeadChip all patients belong to AJCC Stage IV