The primacy of IL-6 in IPS?

In this issue of Blood, Varelias and colleagues demonstrate a central role of interleukin (IL)-6 in idiopathic pneumonia syndrome (IPS) after allogeneic stem cell transplantation (SCT) in both mice and humans.¹ 

IPS is an acute lung injury of noninfectious etiology developing after allogeneic SCT.²  IPS develops as a result of an inflammatory cascade involving conditioning toxicity, cytokine and chemokine production, and an influx of alloreactive donor T cells, neutrophils, and monocytes/macrophages. Although multifactorial etiologies of IPS have been proposed, studies in murine models demonstrate donor T-cell dependency of IPS development, suggesting that IPS includes lung manifestations of acute graft-versus-host disease (GVHD).

Varelias et al have previously shown that absence of interferon (IFN)-γ signaling in host promotes development of IPS.³  In this issue of Blood, they further demonstrate that donor T cell–derived IFN-γ inhibits IL-6 production from host lung parenchyma.¹  Inhibition of IFN-γ signaling in host resulted in dysregulated production of IL-6 from lung parenchymal cells and induced T helper (Th)17–cell differentiation of donor T cells. Importantly, dysregulation of IL-6 was most predominant in the lung, leading to preferential expansion of alloantigen-specific Th17 cells within the lung. Th17-related cytokines promote migration of both T cells and macrophages into the lung and induce IPS (see figure). A central role of the IL-6/IL-17 axis in IPS development was confirmed by using cytokine-deficient mice or by the administration of cytokine-blocking antibodies. The authors also demonstrated that IL-6 was dysregulated in patients developing IPS after allogeneic SCT.

The article by Varelias et al significantly improves our understandings of IPS and GVHD on several points. Emerging evidence indicates the important roles of donor Th17 cells in GVHD, particularly in the skin and lung, although there are still controversies. The present study demonstrates a complex interaction between Th1 and Th17 responses involved in the development of lung-specific GVHD.¹  Moreover, Th17 cells display plasticity and acquire function of Th1 cells last.⁴  Indeed, Varelias et al identified TNFα-producing Th17 cells by using an IL-17 fate map reporter model that could identify progenies of IL-17A–producing cells. Further studies to determine the roles of such inflammatory Th17 cells (as well as T cytotoxic [Tc] 17 cells) in the context of GVHD and graft-versus-leukemia (GVL) are warranted because a better understanding of the complex interplay between Th subsets and cytokines that might differentially affect GVHD and GVL has important clinical implications.

Although the lung is thought to be a target organ of GVHD, IPS does not necessarily develop together with classical acute GVHD involving the skin, liver, and gastrointestinal (GI) tract. This study by Varelias et al demonstrates that the lung-specific inflammatory environment caused by dysregulated production of IL-6 from lung parenchymal cells is associated with the induction of IPS.¹  We also have demonstrated that Th17-attracting chemokines were upregulated preferentially in the lung during acute GVHD.⁵  Taken together, these results suggest that allogeneic T-cell responses induce a cytokine/chemokine environment favoring differentiation and migration of Th17 cells to the lung and making the lung vulnerable to Th17-mediated GVHD.

There are few established therapies for IPS, and this study by Varelias et al suggests that calcineurin inhibitors may not be the best prophylactic agents against IPS, with their ability to suppress IFN-γ production while preserving IL-6 production. Pathophysiology of IPS has been explored in several mouse models in which TNFα levels are increased in bronchoalveolar lavage and inhibition of TNFα significantly mitigates experimental IPS.⁶  On the basis of these results, TNF inhibitors have been evaluated for the treatment of IPS.⁷  This study by Varelias et al suggests that IL-6 inhibition suppresses both IL-17 and TNFα and thus represents a novel strategy to prevent or treat IPS.¹ 

Serum levels of various inflammatory cytokines are upregulated early after SCT. Among these, IL-6 is the major detectable pathogenic cytokine. The authors’ group recently reported a promising result of a phase 1/2 trial of anti-IL-6R monoclonal antibody tocilizumab given on day −1 for GVHD prophylaxis in combination with cyclosporine and methotrexate.⁸  Treatment of B-cell malignancies by CD19 chimeric antigen receptor–modified T cells induces severe cytokine-release syndrome (CRS), mostly within a week after infusion.⁹  Although multiple inflammatory cytokines are upregulated in CRS, tocilizumab alone induces rapid and complete reversal of the symptoms. Given the similarity of clinical manifestations among CRS, acute GVHD, and IPS, IL-6 may be a key cytokine to initiate such a posttransplant inflammatory syndrome and a promising target to intervene.

This study will pave the way to develop a novel therapeutic strategy for IPS. However, there are many questions remaining. First, why is the lung most sensitive to the absence of IFN-γ? Second, where and by which cells are IPS-causing alloreactive T cells primed? The authors’ group has previously shown that host-derived nonhematopoietic antigen-presenting cells (APCs) prime alloreactive T cells in the GI tract and mesenteric nodes in a model in which GI GVHD is most dominant.¹⁰  However, the lung appears to be the major site of T-cell expansion in this IPS model. It is thus possible that alloreactive T cells that cause organ-specific GVHD are primed at local sites. If so, what type of nonhematopoietic cells within the lung act as APCs? Lastly, more human data that confirm the roles of the IL-6/IL-17 axis shown in animal data are needed because IL-6 is often dysregulated in patients early after SCT, irrespective of development of GVHD or IPS, and addition of tocilizumab did not reduce serum levels of IL-17 in a clinical study.⁸  Nonetheless, it will be exciting to evaluate whether tocilizumab could prevent or improve clinical IPS.