Abstract
Idiopathic multicentric Castleman disease (iMCD) is a rare and deadly hematologic illness involving episodic disease flares with polyclonal lymphoproliferation, systemic inflammation, and multiple organ system dysfunction. A cytokine storm involving interleukin(IL)-6 is believed to drive disease pathogenesis in some patients. However, only 34% of patients were found to respond to anti-IL-6 therapy with siltuximab in its registrational clinical trial; no other FDA approved treatments exist for iMCD. With the 5- and 10-year mortality rates reported as 35% and 60%, respectively, there is a clear need for additional treatment options. However, the development of next generation therapeutics is challenging as the etiology, pathological cell types, and signaling pathways involved in iMCD are largely unknown. To identify pathophysiological mechanisms and cellular drivers of iMCD, we applied cutting edge single-cell RNA-sequencing (scRNA-seq) technology to investigate bulk peripheral blood mononuclear cells (PBMCs) isolated from an iMCD patient at two distinct stages of disease activity. The first sample was collected during a short remission period following the patient's first disease flare (partial remission) (clinical symptom: fatigue; laboratory tests: hemoglobin 11.2 g/dL, platelets: 225,000/µL; albumin 4.2 g/dL, creatinine 0.73 mg/dL) and a second sample was collected at the start of his second flare (flare 2) (clinical symptoms: fatigue, fever, night sweats and fluid accumulation; laboratory tests: hemoglobin 12.9 g/dL, platelets: 122,000/µL; albumin 2.3 g/dL, creatinine 1.48 mg/dL). We utilized the Cellranger pipeline (10x Genomics, v.2.1.0) for aggregation of single-cell transcriptomes and Loupe Cell Browser (10x Genomics, v.2.1.0) for initial analysis of 20,135 recovered cells from partial remission (16,283 means reads/cell, 799 median genes/cell) and 19,322 recovered cells in flare 2 (17,327 reads/cell, 823 median genes/cell). Initial analyses of clusters revealed changes in the composition and frequency of immune cell subsets between the two samples. Plasmablasts (identified as expressing CD19, CD27, CD38, CD79a, CD79b) increased 7-fold in number during flare 2 with 28 cells in partial remission and 216 cells in flare 2. Similarly, monocyte and macrophage cell populations increased in frequency from 9% of all PBMCs in the partial remission sample to 15% of all PBMCs in the flare 2 sample. Conversely, CD8+ T cell frequency in the dataset decreased from 22% of the partial remission sample to 13% in flare 2. Interrogation of gene expression profiles of immune cell clusters identified highly activated CD8+ T cells which increased in frequency during flare 2 and are characterized by an inflammatory gene signature including expression of perforin and granzyme. Additionally, inflammatory gene signatures within the myeloid cell compartment during flare were identified, including elevated expression of S100 family members. S100 proteins are implicated in the pathogenesis of a number of autoimmune diseases and contribute to immune cell migration, chemotaxis, and leukocyte invasion. To our knowledge this is the first application of cutting edge single-cell sequencing technology to PBMCs obtained from an iMCD patient in flare and remission. Our observations support a role for both T and B cell activation in iMCD flare and lead us to hypothesize that CD8+ T cells may have left circulation and migrated to sites of active inflammation during this patient's disease flare. This dataset demonstrates involvement of multiple immune cell populations and inflammatory gene programs during disease flare in this patient and provides a novel resource for understanding gene expression and cell population changes in Castleman disease.
Disclosures
Fajgenbaum: Janssen Pharmaceuticals, Inc.: Research Funding.
An optimized workflow for single-cell transcriptomics and repertoire profiling of purified lymphocytes from clinical samples