PhD Scientific Days 2023

Molecular Sciences - Posters K

Molecular Subgroup Classification in Diffuse Large B-cell Lymphoma: First Results of the DLBCL Molecular Profiling Program in Hungary

Text of the abstract

Introduction
Two-thirds of diffuse large B-cell lymphoma (DLBCL) patients are cured with first-line treatment, but the remaining cases develop poorly treatable relapsed/refractory DLBCL, partly due to the underlying molecular heterogeneity. New classification systems based on multiplex genetic variation could lead to better classification methods, improved risk stratification and therapy selection in the future.
Aims
The aim of our research group is to investigate the applicability and prognostic significance of the new molecular classification algorithms in DLBCL in a prospective observational study.
Method
We have developed the 'SU-DLBCL Predictor' gene panel, which allows the analysis of 251 genes, 4 translocation breakpoints (BCL2, BCL6, MYC, PD-L1) and genome-wide copy number variations. During the pilot phase of the project, twenty-four DLBCL patients' diagnostic tissue samples were analysed. After library preparation with a custom SureSelect XTHS2 (Agilent, USA) panel, samples were sequenced on the NextSeq2000 (Illumina, USA) platform and data were evaluated using our own bioinformatics pipeline and the LymphGen classification algorithm.
Results
A median of 19 (min: 9, max: 71) coding variants were identified in the 24 samples analysed, most frequently affecting the KMT2D (42%), CARD11 (42%), HIST1H1E (33%), FAT4 (38%), and TP53 (46%) genes. BCL6 translocation was detected in two patients, while copy number alterations were highly prevalent as amplifications affecting MYC (25%), BCL2 (42%), BCL6 (46%) genes and deletions of the TP53 gene (21%). Based on these abnormalities, 4% of cases were classified as N1, 8% as MCD, 12% as A53, 4% as MCD/A53, 13% as BN2, 8% as EZB, 13% as ST2, and 38% as Other using the LymphGen algorithm.
Conclusion
We are the first in our country to successfully optimize a testing algorithm for the molecular classification of DLBCL, with preliminary results for 24 samples showing overlap with literature data both in terms of variation frequencies and subgroup distribution. Subsequently, patient samples will be analysed using the 'SU-DLBCL Predictor' gene panel.
Funding
Funding for this research was provided by NKFIH KDP-1022882, EFOP-3.6.3-VEKOP-16-2017-00009, ÚNKP-21-3-II-SE-24, ÚNKP-22-5-SE-7, ÚNKP-22-3-II-SE-72, H2020-739593, K21_137948, TKP2021-EGA-24, TKP2021-NVA-15, FK20_134253, the MTA Bolyai programme BO/125/22 and Elixir Hungary.