Titel: DNA methylome profiling of peripheral T-cell lymphomas from prospective randomized European clinical trials
ID: W8-003
Art: Invited talk
Redezeit: 15 min
Session: Workshop 8
(Epi-)Genomics and Cancer

Referent: Anja Pfaus (Ulm/DE)

Abstract - Text


Peripheral T-cell lymphomas (PTCL) are a heterogeneous group of rare lymphoid malignancies derived from mature T-cells. With current treatment options, the majority of patients do not achieve remission or experience relapse after completion of therapy, generally with dismal outcome. Mechanisms of progression and relapse remain elusive and predictive biomarkers do not exist. Despite recurrent mutations in genes involved in DNA methylation like DNMT3A, TET2 or IDH2, no genome-wide DNA methylation profiling has been reported yet in an extended patient cohort particularly not from patients treated within clinical trials.

Within the European TransCan project, we aim at a multi-omics characterization including exome (tumor and germline) and RNA sequencing as well as DNA methylation profiling of samples from patients with PTCL treated in prospective randomized European clinical trials. Here, we present first results of the comprehensive characterization of the DNA methylation landscape. To this end, Illumina MethylationEPIC array analysis was conducted in 94 PTCLs tumor samples. Moreover, blood samples taken before treatment initiation were studied from 18 patients representing extreme good and bad responders (n=9 and n=9, respectively).

DNA methylation data was exploited to calculate the biological (epigenetic) age using a range of different algorithms ("epigenetic clocks"). Moreover, DNA methylome profiles of PTCL tumor samples were compared to non-malignant T-cell populations of defined differentiation stages. Unsupervised analysis of malignant and benign samples revealed three distinct epigenetic PTCL subgroups strongly correlating with the biological age. Divergences between chronological (real) and biological (epigenetic) age is a recognized proxy of any kind of pathological condition. Treatment naïve blood samples partly suggested a correlation of DNA methylation and clinical outcome. We show that extreme good therapy responders tend to have a younger biological age predicted by DNA methylation than poor responders.

We conclude that the identification of epigenetic subgroups and biomarkers can become an important step to understand and optimize treatment response in these biologically heterogeneous entities. Our first results from this ongoing study show that DNA methylation analysis in tumor and blood samples before treatment might give a good proxy to estimate the response to therapy and, thus, to guide therapeutic protocols in the future.