MOLECULAR HIGH GRADE (MHG) GENE EXPRESSION PROFILE IN DLBCL IS ENRICHED AMONG PATIENTS WITH EARLY TREATMENT FAILURE

Résumé fourni par la source

Diffuse large B cell lymphoma (DLBCL) is biologically and clinically heterogenous, with variable response to standard treatment. We previously described a subgroup of DLBCL with a GCB-like gene expression signature similar to that of Burkitt lymphoma and enriched for ‘double hit’ FISH status. This molecular high grade (MHG) subset of patients was identified from the REMoDL-B trial using Illumina WG-DASL data and showed an inferior response to R-CHOP,1 confirmed in our population based cohort.2 The MaPLe trial (Molecular Profiling for Lymphoma) is a multicentre prospective observational cohort study for the collection of tumour samples from patients with DLBCL, with almost 3000 samples now collated. The aim of this analysis was to characterise lymphomas that progressed early according to gene expression profiling (GEP) status, including MHG and cell of origin (COO), as well as double hit FISH status and targeted sequencing. Patients who progressed within 6 months of MaPLe study baseline visit were identified. To allow assessment of biological variables which may contribute to treatment failure, each case was matched to another patient who had not progressed within 6 months based on age and IPI score at diagnosis. Samples with sufficient material were interrogated using the HTG EdgeSeq Pan B-Cell Lymphoma Panel and classified according to MHG and COO status. Targeted sequencing of 191 gene/region targets using TWIST library prep, sequenced on an Illumina HiSeq was also performed and FISH for double hit status is underway. Thirty nine eligible patients were identified as ‘early progressors’ and matched to 39 controls. HTG data was available and passed QC metrics in 64 patients (32 from each group). In total, 18/64 (28%) were classified as MHG, with 12/18 of these (67%) represented within the early progressor group. This is significantly higher than the prevelance of MHG previously demonstrated in REMoDL-B and population based datasets, reflecting the case selection within this study. ABCs were evenly distributed between groups (11/21 vs 10/21) while GCBs and unclassifiable samples were slightly more prevalent in the control group (7/18 vs 11/18 and 2/7 vs 5/7 respectively). Relevant mutations, previously described to be associated with ABC, GCB and MHG, were demonstrated within the GEP subgroups, however there was no significant association between individual or subtype-specific mutations and progression. The results of this study confirm that the MHG gene expression profile can be reproduced using HTG EdgeSeq Pan B-Cell Lymphoma Panel data. MHG was more frequently identified in the early progressor cohort, suggesting that MHG GEP contributes to poor response to conventional R-CHOP treatment for DLBCL. Prospective identification of this group for testing more effective treatment strategies is a priority. 1. Sha et al. J Clin Oncol. 2019;37(3):202-212 2. Painter et al. BJH 2019;185(4):781-784 The research was funded by: Bloodwise (Blood Cancer UK) Keywords: Diagnostic and Prognostic Biomarkers, Aggressive B-cell non-Hodgkin lymphoma, Pathology and Classification of Lymphomas Conflicts of interests pertinent to the abstract P. Johnson Consultant or advisory role: Takeda, Bristol-Myers Squibb, Novartis, Genmab, InCyte, Morphosys, Kymera Research funding: Janssen, Epizyme

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.