P1257: MOLECULAR CHARACTERIZATION AND CLONAL EVOLUTION OF MYCOSIS FUNGOIDES

Abstract

Background: Mycosis fungoides (MF) is the most common primary cutaneous T-cell lymphoma with a frequency of over 50% of all cases. MF has an indolent progression over several years through different stages. Symptoms evolve from a non-aggressive “low-risk” (LR) clinical form to an aggressive and treatment-resistant “high-risk” (HR) form. Approximately 20% of patients with MF evolve to a HR MF with transformed cells (tMF), which is associated with a poor prognosis. Aims: At the molecular level, the genomic landscape of MF remains poorly studied. Furthermore, the mechanisms that determine the progression to HR and tMF are still unknown. Methods: We performed whole exome sequencing (WES) of 70 samples from 48 patients at the University Hospital Center (CHU) of Lille, France, between 2007 and 2018. Some patients have multiple sequential samples over several years, with or without progression from LR to HR. Sequencing was performed on the Illumina NovaSeq 6000 system at the Broad Institute (BI) Genomics Platform. The bioinformatics analysis was performed on the Terra platform (BI), which includes tools for the analysis of genomic data. Copy number variations (CNVs) are detected with ModelSegments software in Tumor-Only mode and common large and focal CNVs in the cohort are highlighted by GISTIC2.0. Single nucleotide variants (SNVs) were detected with the CGA WES Characterization Pipeline (Getz lab, BI). Significantly mutated genes were determined by MutSig2CV. The mutational signature was performed with SignatureAnalyzer software. Subsequently ABSOLUTE software is used to calculate the cancer cells fraction of each variant and CNVs in order to follow the clonal evolution of sequential samples. Results: Somatic mutation analysis at the time of diagnosis detected genes involved in T-cell oncogenesis, by order of frequency: the transcription factor JUNB, the MAP kinase member MAPK1, the JAK/STAT member STAT5A and the epigenetic regulators of DNA SUZ12 and TET2. Mutational signatures were represented by the COSMIC SBS1, SBS7, and SBS15, characterized by aging, ultraviolet (UV) exposure, and defective DNA mismatch repair and microsatellite instability, respectively, suggesting their role in the mutational process of MF oncogenesis. The analysis of the CNVs revealed focal deletions affecting genes involved in ER stress TMEM259 (cytoband 19p13.3), the tumor suppressor gene TP53 (17p13.1), the transcriptional repressor gene ZEB1 (10p11.22) which inhibits interleukin 2 IL-2 gene expression, the TNFAIP3 gene (6q16.3) which inhibits nuclear factor kappa B (NF-kB), and the CDKN2A and MTAP genes (9p21.3) which are frequently co-deleted in cancers. The single focal amplification identified a small cytoband in 10p15.1 that carries the genes encoding the interleukin 2 receptor alpha (IL2RA) and interleukin 15 receptor alpha (IL15RA) which have a central role in T cell survival and proliferation. Clonal evolution analysis established that del10p11.22, amp10p15.1 and triplication of chromosome 7 are present only in HR stage, suggesting their role in the tumor progression. Overall survival analysis shows that amp10p15.1 and del10p11.22 – both associated with IL2 pathway - are significantly associated with reduced survival. Summary/Conclusion: WES data from MF identified key CNVs events of oncogenesis and clonal evolution of MF that could serve as predictive markers in clinical practice. Our findings clarify MF genetics and provide novel insights into the disease progression.