Heidelberg DNA-methylation classifier

The Heidelberg DNA-methylation classifier is a tool that uses DNA methylation patterns to identify the tissue of origin of cancer cells. This classifier was developed by researchers at the German Cancer Research Center in Heidelberg, Germany.

DNA methylation is a chemical modification of DNA that can control gene expression. Different tissues have unique DNA methylation patterns, which can be used to identify the tissue of origin of a tumor. The Heidelberg DNA-methylation classifier uses this principle to classify tumors based on their DNA methylation patterns.

The classifier is based on a machine learning algorithm that was trained using DNA methylation data from thousands of tumor samples representing 33 different types of cancer. The algorithm can accurately predict the tissue of origin of a tumor sample with high sensitivity and specificity.

The Heidelberg DNA-methylation classifier is available online as a web tool, and it can be used to analyze DNA methylation data from tumor samples. This tool is particularly useful for cases where the tissue of origin of a tumor is uncertain, which can occur in cases of metastatic cancer or poorly differentiated tumors.

Overall, the Heidelberg DNA-methylation classifier represents a valuable tool for cancer diagnosis and research, and it has the potential to improve our understanding of the molecular mechanisms underlying cancer development and progression.

Lebrun et al. reported a pediatric case of a novel tumor type among the other CNS embryonal tumors classified within the methylation class “CNS Embryonal Tumor with BRD4-LEUTX Fusion-LEUTX Fusion”. The patient was a 4-year girl with no previous history of the disease. For a few weeks, she suffered from headaches, vomiting, and mild fever associated with increasing asthenia and loss of weight leading to a global deterioration of health. MRI brain examination revealed a large, grossly well-circumscribed tumoral mass lesion located in the left parietal lobe, contralateral hydrocephalus, and midline shift. Microscopic examination showed a highly cellular tumor with a polymorphic aspect. The majority of the tumor harbored neuroectodermal features composed of small cells with scant cytoplasm and hyperchromatic nuclei associated with small “medulloblastoma-like” cells characterized by syncytial arrangement and focally a streaming pattern. Tumor cells were diffusely positive for Synaptophysin, CD56, INI1, and SMARCA4 associated with negativity for GFAP, OLIG2, EMA, BCOR, LIN28A, and MIC-2. Additional IHC features included p53 protein expression in more than 10% of the tumor's cells and very interestingly, loss of H3K27me3 expression. The Heidelberg DNA-methylation classifier classified this case as “CNS Embryonal Tumor with BRD4:LEUTX Fusion”. RNA-sequencing analyses confirmed the BRD4 (exon 13)-LEUTX (exon 2) fusion with no other molecular alterations found by DNA sequencing. The case report confirmed that a new subgroup of CNS embryonal tumor with high aggressive potential, loss of H3K27me3 protein expression, BRDA4-LEUTX fusion, named “Embryonal CNS tumor with BRD4-LEUTX fusion”, has to be considered in the new CNS WHO classification ¹⁾.