Marginal zone lymphoma transdifferentiated to histiocytic sarcoma

Histiocytic sarcoma (HS) is a rare aggressive, but poorly understood haematopoietic neoplasm.1 A substantial number of HS are evolved from other preceding or concurrent, genealogically different, lymphoma/leukaemias through a still unknown mechanism(s).2 The clonal relatedness of these tumours was infrequently suggested by finding the same immunoglobulin or T-cell receptor genes rearrangements.3 Recent landmark studies discovered frequently mutated genes in the RAS-MAPK pathway in HS in both de novo tumours and in tumours with antecedent haematologic malignancies,4, 5 including follicular lymphoma or chronic lymphocytic leukaemia.2 Few cases of histiocytic sarcoma in association with preceding marginal zone lymphoma (MZL) have been published in English literature.6, 7 Here, we present a case of a 53-year-old woman who presented initially with MZL and later with HS, which was shown to be clonally related to the MZL. In October 2018, the patient presented with extensive adenopathy. Excisional biopsy of a cervical node showed MZL (Fig 1A) mainly in the form of a perifollicular infiltrate of small B cells with scattered larger immunoblasts. The lymphoma cells were positive for CD20, PAX5, and BCL2 and were negative for CD5, CD10, CD163, BCL6, LMO2, and cyclin-D1. No rearrangements in BCL2, BCL6, or MALT1 were present. Clonality study identified a clonal B-cell population (Fig 2A). She received six cycles of bendamustine and rituximab. While she initially showed significant clinical improvement, a subsequent positron-emission tomography (PET) scan showed extensive disease. She received six cycles of R-CHOP with no significant avail. Magnetic resonance imaging showed a large left paracervical spinal mass infiltrating the paravertebral muscles as well as neural foramina resulting in brachial plexopathy. Biopsy from the cervical lesion showed HS (Fig 1B) showing variably sized from intermediate to large cells with frequent giant cell formation with hyperchromatic nuclear chromatin and one or more prominent nucleoli. Mitotic figures including atypical forms were readily detected. Large areas of tumour necrosis with associated neutrophilic infiltration and single-cell apoptosis were identified. Rare multi-nucleated giant cells were present of which some showed features of haemophagocytosis. Neoplastic cells were positive for CD4, CD45, CD68 (PG-M1 clone), CD163, lysozyme, and partially for S100 and negative for CD1a, CD3, CD15, CD20, CD21, CD23, CD30, MUM1, PAX5, cytokeratin, SOX10 and p53. She was started on cyclophosphamide, etoposide and vinblastine. Despite receiving chemotherapy, her condition progressively deteriorated. She died in June 2020. The evolution of a low-grade B-cell lymphoma to a tumour with a seemingly different cellular lineage has been well recognized and while in many cases a biological connection was mostly assumed, in a few instances an unequivocal molecular relatedness was shown.8 We decided to examine the presence of any clonal relationship between the MZL and HS in this patient. B-cell receptor clonality using the BIOMED-2 protocol was performed. In three separate primer mixes, identical immunoglobulin heavy (IgH) chain gene rearrangements were identified in MZL and HS. Of note, the HS showed the virtual disappearance of the peak in mix C, possibly due to somatic hypermutation. Additional peaks in IgH primer mix B and C appeared, possibly due to preferential amplification of IgH rearrangements of a few stromal B-cells (Fig 2A). Sequencing the IgH locus confirmed a clonal relationship between the MZL and HS with a very high rate of somatic hypermutation in both specimens (Fig 2B). The MZL and HS clonal sequences contained identical V-(D)-J gene segments and virtually identical CDR3 junction sequences. The presence of a few missense changes confirmed an ongoing hypermutation process. It is not uncommon even in sporadic HS to find a clonal immunoglobulin receptor gene rearrangement. Whether there is an underdiagnosed clonal B-cell population in these instances, is not known and would be worth studying further. Several studies have previously illustrated the same clonal B-cell receptor rearrangement between mainly an indolent B-cell lymphoma and subsequent HS.9 While many of the haematopoietic cells are considered to be committed to one specific lineage, there is a great level of lineage plasticity between lymphoid and myeloid cells.10 We next evaluated the mutational profiles of both MZL and HS in an attempt to discover a possible clonal evolution or possibly an alternative divergent differentiation from a common ancestor. Next-generation sequencing using a panel of over 500 genes was performed on both MZL and HS with an average read depth for the MZL of 1053× and for the HS of 1205×. The gene-sequencing results showed a similar core mutational profile with the addition of several new mutations in HS (Table I). As illustrated in Table I, of two major gene mutations with known oncogenic roles in lymphomas detected in HS, that is, BCL10 and CREBBP, only CREBBP was present in the indolent lymphoma.11 Variant allele fraction (VAF) of this shared mutation was very similar and low in both samples. Interestingly, the deletion of MAP3K1 (p.Arg930Thrfs*10) detected in MZL was not present in HS. As this change was only present in a small subclone (VAF: 5·2%), this might have been deleted after treatment. BCL10 mutation, on the other hand, was a new addition showing a relatively high VAF in the HS and is presumably one of the major driving genetic changes for the latter tumour. BCL10 is an apoptotic regulatory molecule identified through its direct involvement in a subset of marginal zone lymphomas.12 Overexpression of C/EBPα/β or underexpression of PAX5, a deciding transcription factor for B cell lineage, can, at least in vitro, lead to transdifferentiation of a B cell tumour to a more myeloid tumour such as HS.13 HS, here in this patient, was negative for any B-cell immunomarkers, including PAX5. As both cytogenetic and molecular studies confirmed the transdifferentiation of MZL to the subsequent HS, whatever the main driving force was for these changes had led to the underexpression of PAX5. Interestingly, we did not detect any gene mutations in MAPK and PI3K signalling pathways, two of the most altered pathways in HS.4 While we identified other relatively common changes such as CCND3 or CREBBP mutations in this tumour, the presence of multiple mutations in linker histone gene HIST1H1 E might suggest a mutually exclusive route to HS tumourigenesis in our case. HIST1H1 mutations are relatively common in low-grade non-Hodgkin lymphoma (NHL), especially follicular lymphoma.14 Six different HIST1H1 E mutations were observed in HS with a relatively high VAF indicating their probable role in tumourigenesis of HS. It has recently been shown that disruption of linker histone H1 protein can unlock the expression of stem cell genes that are normally silenced in mature lymphoma.15 Whether this genetic change works as a driving force for possible lineage infidelity needs more comprehensive research. In summary, this represents a rare case of MZL transdifferentiated to HS and presents the first case where the initial tumour and transdifferentiated tumours were sequenced using an extensive NGS panel. This highlighted possible pathogenic mutations in BCL10 and HIST1H1, suggesting possible mechanisms for the pathogenesis of HS, both de novo and in the context of transdifferentiation. This opens new avenues for the study of this rare haematological malignancy. AS: study design, acquisition, assembly, analysis and interpretation of data, drafting of the manuscript, critical revision of the manuscript for important intellectual content. JMD: acquisition, interpretation of data, drafting of the manuscript. PAM: study design, analysis, and interpretation of data, drafting of the manuscript. RMT: acquisition and interpretation of data, drafting of the manuscript, critical revision of the manuscript for important intellectual content. PS: acquisition, analysis, and interpretation of molecular data, drafting of the manuscript, critical revision of the manuscript for important intellectual content. The study was approved by the “Research Ethics Board” of the University Health Network and conducted in compliance with the Declaration of Helsink. None. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.