Co-Occurrence of Clonally Related Follicular Lymphoma and Histiocytic Sarcoma

Abstract

Histiocytic sarcoma (HS) is an extremely rare and aggressive malignancy thought to be derived from mononuclear phagocytic cells. HS may occur as primary tumor or may be clonally related to a hematologic malignancy. Rare cases of HS have been reported in patients with follicular lymphoma (FL). Prior reports have demonstrated a common clonal origin of FL and HS in many cases via demonstration of identical BCL2/IGH [t(14;18)] translocations in both malignancies. As this is likely the earliest event in FL pathogenesis, the mechanisms by which HL may arise from FL remain unknown. A greater understanding of how a B cell malignancy can transdifferentiate into a myeloid one would yield insights into the fundamental underlying biology. From a patient with long-standing history of FL and subsequent diagnosis of HS, we collected tumor biopsies from a tumor site harboring both FL and HS. Tumor biopsies were subjected to standard histopathological assays, as well as targeted hybrid-capture DNA sequencing of 164 genes commonly mutated in hematologic malignancies and cytogenetic assays for the BCL2 rearrangement (Figure 1). Additionally, we performed single cell RNA sequencing (scRNA-seq) using the 10X chromium followed by single cell long read sequencing using Nanopore Oxford Technologies (Figure 1). Two tumor populations were evident in the scRNAseq data - FL tumor cells, identified by homogeneity of BCR light chain variable gene expression and expression of FL marker genes, and a separate cluster of cells expressing CD68, CD163 and S100A8, characteristic for HS cells. Gene expression profiles of the two tumor populations revealed high expression of the targetable “don´t eat me” signal CD47 on both populations. HS also expressed BCL2, hinting at a common clonal origin of FL and HS neoplasms. Indeed, fluorescent in situ hybridization (FISH) studies demonstrated the presence of t(14;18) in both the FL and HS tumor cells. To understand if similar transcription factors (TFs) were active in the two malignancies, we inferred the top 10 enriched TF for each tumor subpopulation and found that while TFs like IRF8 and REL were enriched in FL cells, HS cells showed a higher expression of MYC, STAT2 and CEBPD. Bulk targeted hybrid-capture DNA sequencing of the tumor sample revealed mutations in genes frequently altered in FL ( e.g.CREBBP, BIRC3, and PIM1) as well as in a gene commonly mutated in HS ( MAP2K1). To gain deeper insight into the clonal relationship between the two malignancies within the same tumor, we conducted long read sequencing on the barcoded single cell-cDNA. We are currently analyzing the data in two ways. First, we are searching for the mutations identified in bulk sequencing now in the single cell long read data from the FL and HS cells, using microenvironmental immune cells from the same tumor as a negative control. We have confirmed that these genes are all expressed in the FL, HS, and immune cells, providing transcripts for us to analyze for mutations. Second, we are employing an unbiased approach for identifying genetic variants, again comparing FL and HS cells. This data will be available by the annual meeting. By applying single cell genomics analyses to a combined FL/HS biopsy, our study aims to understand the evolutionary relationship of HS arising from FL, particularly whether the HS arose from an early dormant clone (branched evolution) or from a late, prevalent Fl clone (linear evolution). Evidence of the former would strengthen the view that common progenitor cells (CPCs) in FL are capable of differentiating into myeloid and lymphoid malignancies.