New approach targets key traffickers in the fibrosis network Free

In this issue of Blood, Newton et al¹ report that sclerotic cutaneous graft-versus-host disease (GVHD) shares dysregulated gene expression with scleroderma and that human antibody neutralization of epiregulin ameliorates the diseases.

Sclerotic cutaneous GVHD (SclGVHD) is an autoimmune scleroderma/systemic sclerosis (SSc)–like syndrome.^2,3 SclGVHD pathogenesis is triggered by alloreactive T cells and perpetuated with subsequent dysregulations among T cells, B cells, dendritic cells (DCs), macrophages, and fibroblast cells, leading to excessive extracellular matrix (ECM) production and fibrosis.² Transforming growth factor β (TGF-β) produced by CSF1R⁺ macrophages was found to augment fibrosis in murine models of SclGVHD, and anti-CSF1R blocking monoclonal antibody (mAb) reduced macrophage and fibrosis and ameliorated SclGVHD.⁴ However, a recent phase 2 clinical trial with anti-CSF1R mAb (axatilimab) showed that skin and fascia had only partial response in most patients compared to complete responses seen in other organs⁵; in addition, 36% of axatilimab-treated patients discontinued treatment due to adverse events or progression of diseas.⁵ Thus, further understanding of pathogenesis and developing effective therapeutics for SclGVHD are still in need.

Odell et al⁶ had demonstrated a critical role of epiregulin (EREG)–epidermal growth factor (EGFR) signaling between DC3 and fibroblasts in maintaining elevated ECM production by fibroblasts in fibrotic tissues of murine models and human SSc, even though EREG is just 1 of 7 cell-surface EGFR ligands. They observed that accumulation of EREG⁺ DCs in the skin and lung fibrotic tissues was correlated with disease severity in patients with SSc.⁶ EREG⁺ DCs are proinflammatory DC3 cells derived from CD14⁺ myeloid precursors in peripheral blood under granulocyte-macrophage colony–stimulating factor (GM-CSF) influence, which is different from conventional DC2 cells derived from myeloid precursors under FLT3 influence.⁷ Odell et al⁶ unraveled a cellular circuit in which type-I interferon (IFN-I) induces EREG expression in DC3, and EREG-mediated EGFR activation triggers expression of multiple fibrotic ECM genes in fibroblasts. On the contrary, EREG⁺ DC3 also activates NOTCH signaling in the fibroblast; in turn, fibroblast-derived NOTCH ligands bind NOTCH receptors on EREG⁺ DC3 to maintain EREG expression.⁶ Ex vivo culture of skin and lung explants from patients with SSc showed that EREG inhibition with mAb reversed human skin and lung fibrosis, and the inhibition appeared to be more effective for the skin.⁶ However, whether similar mechanisms exist in SclGVHD was not studied.

The clinical manifestations and histopathology of SclGVHD are similar to SSc,² but their similarities at cellular and molecular levels have not been studied. In the current study, Newton et al integrated single-cell RNA sequencing data from the skin of patients with SclGVHD with publicly available data of SSc, morphea (localized scleroderma), and healthy controls to compare disease markers. Increased frequencies of DC3 with upregulated expression of EREG were observed with all 3 types of fibrotic skin diseases.¹ Further analysis of fibroblast subsets showed that both SSc and SclGVHD had higher frequencies of myofibroblast that expressed higher levels of tenascin-C (TNC), in contrast, morphea had higher frequencies of inflammatory fibroblast that expressed higher levels of fibronectin-1 (FN1).¹ Both TNC and FN1 are endogenous TLR4 ligands; although TNC augments fibrotic collagen production, FN1 augments the production of CCL2 and interleukin-6, which promote cell migration, adhesion, and matrix assembly.⁸ These indicate that pathogenic fibroblasts in SclGVHD are similar to those in SSc at cellular and molecular levels; these also suggest a cellular circuit consisting of EREG-TNC/FN1-TLR4 signaling among DC3 cells, myofibroblasts, and inflammatory fibroblast/macrophages.

SclGVHD is an important manifestation of chronic GVHD, a sequala of acute GVHD triggered by alloreactive T cells.² Besides skin, chronic GVHD targets also include gut, liver, lung, and salivary/lacrimal gland,² and tissue- and disease stage–specific pathogenesis may exist in the different target organs.² It is of interest that anti-EREG treatment appeared to be more effective for reducing fibrosis in the skin than in lung in murine SSc models.⁶ This may explain why anti-CSF1R treatment that reduced TGF-β–producing macrophages had only partial response in the skin in most patients compared to complete responses seen in other organs.⁵ The observations that DC3 can differentiate from blood CD14⁺ myeloid precursors under GM-CSF influence⁷ and that EREG from inflammatory DC3 plays an important role in maintaining fibrosis⁶ may also provide new explanation to some clinical observations. For example, patients given peripheral blood transplant tend to develop more severe SclGVHD,⁹ and this may result from GM-CSF–producing alloreactive T helper 1/TC1 cells augmenting the differentiation and expansion of DC3 in GVHD target skin tissues. Antibodies from donor B cells augmented the persistence of SclGVHD in murine models,¹⁰ and this may result from antibody stimulation of IFN-I production that augments DC3 production of EREG. Thus, understanding how alloreactive T- and B-cell subsets regulate DC3 and macrophage differentiation/expansion, as well as their secretion of EREG, TGF-β, or other fibrotic factors in different GVHD target organs at different disease stages, may help develop novel approaches to prevent induction or treatment of fibrosis in different organs; and targeting DC3 and inhibition of EREG may be particularly effective for SclGVHD.

To promote clinical translation of EREG inhibition for treating fibrotic diseases, using humanized transgenic mouse platform for human antibody drug discovery, Newton et al successfully developed a high-affinity human EREG-neutralizing antibody, called EREG hNAb, which specifically inhibits human EREG.¹ The efficacy and safety of EREG hNAb in treating fibrotic diseases were first tested with a murine model of SSc with human EREG knockin mice, and the treatment reduced skin thickness and tissue expression of profibrotic inflammatory genes Tnc and Il6.¹ Second, EREG hNAb were tested with ex vivo culture of skin explant from patients with SclGVHD. EREG hNAb treatment did not affect fibrosis marker expression in healthy skin or its viability while reducing TLR4- and EGFR-associated biomarkers in SclGVHD skin.¹ These suggest that EREG hNAb can be a strong potential therapeutic candidate for treating patients.

In summary, this report by Newton et al provides novel insights into the pathogenesis of fibrosis in SclGVHD and a potential novel therapy with EREG hNAb for the disease.

Conflict-of-interest disclosure: The author declares no competing financial interests.