Subtype-specific differences in the genomic landscape of cutaneous T cell lymphomas. (October 2021)

• Record Type: Journal Article
• Authors: Park, Joonhee; Daniels, Jay; Ringbloom, Kimberly; Martinez-Escala, Maria Estela; Yang, Jingyi; Snowden, Caroline; Zhang, Yancong; Mo, Samuel; Pro, Barbara; Sosman, Jeffrey; Guitart, Joan; Choi, Jaehyuk
• Is Part Of: European journal of cancer. Volume 156(2021)Supplement 1
• Page Start: S2
• Publication Date: 2021-10
• Subjects: Cancer -- Periodicals; Neoplasms -- Periodicals; Cancer -- Périodiques; Cancer; Tumors; Electronic journals; Periodicals; Electronic journals; 616.994
• DOI: 10.1016/S0959-8049(21)00628-6
• Languages: English
• ISSNs: 0959-8049

Abstract

Abstract : Introduction: Cutaneous T cell lymphomas (CTCLs) represent a diverse group of lymphomas of the skin-homing CD4+ T cell. In early-stage disease, mycosis fungoides (MF), disease is limited to the skin and draining lymph nodes and does not involve the blood. In contrast, Sézary syndrome (SS) is characterized by erythroderma and leukemic involvement. Though immune markers have suggested that MF and SS are distinct entities, over-representation of SS in existing next-generation sequencing data has limited comparisons of the genomics of these diseases. To overcome this gap, we performed DNA-sequencing in a large cohort of CTCL cases to understand genetic relationships between disease subtypes. Materials and methods: We performed whole genome sequencing or whole exome sequencing on clinically annotated CTCL samples with MF, SS or a third group which we refer to as leukemic MF. This third group initially presented with skin-limited disease and were diagnosed as MF. At diagnosis, blood involvement was not observed. However, over time this group developed lymphadenopathy, erythroderma, and leukemic involvement. For each sample, somatic copy number variants (SCNVs) and somatic single nucleotide variants (SSNVs) were identified via previously published pipelines in the lab. Then, the frequency of driver events was compared across clinical subtypes. Results: First, we analyzed SSNVs. The majority of driver point mutations were not specific to one disease subtype, with the exception of two mutations more frequent in MF (JUNB and MAPK1). Analysis of the context of SSNVs showed similarities and differences in the mutational processes driving disease. MF, leukemic MF, and SS all showed enrichment in a UV mutational signature. Patients with leukemic involvement specifically showed enrichment of an aging signature, while MF showed enrichment in an alkylating signature. Analysis of SCNVs showed dramatic subtype-specific differences. Samples with leukemic disease (whether originally diagnosed as SS or MF) showed highly recurrent chromosome arm level deletions of 17p and 10q, and amplifications of 17q. At the gene level, many of driver SCNVs were shared across stages, including deletions of ARID1A, ZEB1, and CDKN2A. However, several genes including those located on broad SCNVs (such as TP53 on 17p) were significantly more altered in SS and leukemic MF compared to skin-limited MF. Conclusions: Together, our data suggests that there are genetic similarities between MF and SS including in driver point mutations, a UV mutational signature, and several driver genes deletions. However, there are certain alterations that are hallmarks of leukemic disease. These observations hold true for samples that are initially diagnosed with MF and later develop leukemic disease. This data raises the possibility that some or all of these alterations are key for leukemic progression, and that perhaps MF and SS exist as a disease spectrum rather than distinct entities.