BRAF^V600E vs cell of origin: what governs LCH?

In this issue of Blood, Shi et al use immunophenotyping and single-cell RNA sequencing to identify differentially expressed genes (DEGs) in various populations of circulating mononuclear cells from pediatric patients with Langerhans cell histiocytosis (LCH).¹ 

LCH is an inflammatory myeloid neoplasia that is characterized by MAPK mutations, most frequently BRAF^V600E. Shi et al answer some of the biologic and clinical mysteries of LCH. What is the effect of the BRAF^V600E mutation on DEGs in various cell populations? Is there a gene dosage effect? Do the answers to these questions explain the clinical manifestations of LCH?

LCH is an inflammatory myeloid neoplasia with an incidence of 4 to 8 cases per million children, a rate that is similar to pediatric Hodgkin’s lymphoma and acute myeloid leukemia.² Patients are classified as being at a high risk for death if they have bone marrow, spleen, or liver involvement. Biopsies reveal clonal pathologic CD207⁺CD1A⁺ dendritic cells (DCs; Langerhans cells [LCs]) and inflammatory stroma.² Mutually exclusive somatic mutations of MAPK pathway genes occur in ∼85% of cases.² These drive myeloid differentiation, senescence, resistance to apoptosis, and dysfunctional migration of LCs.^3-5 Single-agent MAPK inhibition yields nearly universal responses. However, relapse almost always occurs following cessation of therapy, and precursor cells persist among peripheral blood mononuclear cells.^6-8 

Thus, the cell of origin in LCH is an active area of investigation and where the work of Shi et al assumes importance. They performed a subgroup analysis of 14 patients with diverse disease characteristics: 4 in the clinical high-risk group with cell-free circulating (cf)BRAF^V600E levels >3.1%, 4 with a low cfBRAF^V600E level (<3.1%) but high-risk (bone marrow, spleen, or liver involvement) and low-risk clinical status, and 6 BRAF^V600E-negative patients.¹ Activation of MAPK genes occurred in circulating mononuclear cells of all patients. When analyzing patients by mutation status, unique sets of genes and transcription factors were activated in the different cell populations. An intriguing discovery was the decreased frequency of plasmacytoid DCs, which was significantly associated with disease severity. Analysis of gene expression in patients before and after BRAF inhibitor therapy revealed decreased markers of inflammation and altered cellular metabolism genes (see figure).

Prior transcriptomic analysis of lesional LCs led to the hypothesis that they arise from myeloid DC precursors rather than epidermal LCs.^3,9 Recurrent mutations in the MAPK pathway genes in LCs led to the misguided myelomonocytic DC precursor model of LCH: aberrant MAPK activation in myeloid DC precursors at critical stages of differentiation was believed to determine the extent of disease.² In this model, acquisition of BRAF^V600E in stem cells leads to high-risk disease, whereas acquisition of BRAF^V600E in committed tissue-resident DC precursors leads to low-risk LCH.^2,3

A study of single-cell sequencing defined 14 LCH progenitor cell subsets within the lesions of all clinical LCH subtypes.¹⁰ These subsets were compared with monocytes, macrophages, and T lymphocytes in the same lesion. The 2 least differentiated subsets had the least specialized pattern of gene expression, with high levels of CD1A and cell proliferation genes involved with DNA replication and cell cycle regulation. The more differentiated subsets had higher expression of cytokine signaling, chemotaxis, and interferon signaling. Analysis of transcription regulation found associations with cancer in the least differentiated group and immune-related genes in the most differentiated groups. One of the hypotheses generated from these results was that clinical manifestation of LCH depends on the mix of less and more differentiated cells within the lesion.¹⁰ 

Shi et al investigated the molecular alterations in the circulating mononuclear phagocytes in LCH patients by a modified version of single-cell tagged reverse transcription. Transcription profiles identified 11 clusters, including CD14⁺ classical monocytes, CD16⁺⁺ nonclassical monocytes, CD14⁺CD16⁺ intermediate monocytes, a minor cDC1 cluster, a major cDC2 cluster, plasmacytoid DCs, B cells, plasma cells, and immature progenitors. Analysis of differentially expressed genes from the cfBRAF^V600E mutational level groups revealed widespread activation of the MAPK pathway, with diverse aberrations in the expression levels of RAS-MAPK-ERK pathway–related genes across the circulating myeloid compartment and the expression levels of genes varying by cfBRAF^V600E mutational group and cell subset.

Universal activation of the MAPK pathway in all of the circulating monocyte populations of LCH patients can be due to a bone marrow precursor mutation leading to cell-intrinsic activation of this pathway across all of these lineages or it could be due to cell-extrinsic effects. By looking at a lineage that should not harbor a MAPK-activating mutation, which of these 2 options is correct should be revealed. Any differential expression of MAPK-related signaling in these cells would suggest that cell-extrinsic effects are happening and are dependent on cell type. Although hypothesized, the investigators did not provide evidence that the only cells harboring the BRAF^V600E mutation expressed MAPK DEGs. Simultaneous detection of DNA mutations and RNA transcripts at a same single cell could address this problem in future studies.

One caveat of their study is that the overall number of sequenced cells is small. Moreover, the control cells were not evenly distributed among the different cell types. Hence, it is not clear whether the overall transcriptomic landscape is affected by the level of BRAF^V600E, especially given the intriguing observation of widespread MAPK activation across all cell types. Another confounding finding is the relatively large number of high-risk multisystem LCH patients without cfBRAF^V600E. It needs to be determined whether this was due to the heterogeneity of LCH and racial diversity or to presence of additional MAPK mutations that were not tested.

This study significantly advances our understanding of LCH immunopathogenesis, indicating that the oncogenic BRAF^V600E mutation may lead to activation of the MAPK pathway in LCH lesions, as well as in circulating mononuclear phagocytes, the presumed precursors of tissue LCH cells. However, the data leave unanswered questions, many of which will require additional functional validation using in vitro and in vivo models. How does one reconcile the data from that study with previous results indicating that multisystem and single-lesion diseases are derived from cells with different maturational status? Will patients with neurodegenerative LCH have the same composition of peripheral mononuclear phagocytes? Are the observations made in that study a common theme in histiocytic disorders, such as Erdheim-Chester disease and juvenile xanthogranuloma?