The controversial role of tumor necrosis factor α in fibrotic diseases

Results of in vitro and in vivo experiments concerning the role of TNFα in fibrosis are in part contradictory and do not allow definite conclusions about the role of TNFα as a profibrotic mediator or an antifibrotic cytokine. The majority of in vitro studies show antifibrotic effects of TNFα, in that it suppresses the production of collagen, reduces the expression of TIMPs, and stimulates the release of MMPs, thereby preventing the accumulation of ECM. The intracellular signaling pathways that mediate these effects have been identified and include activation of JunB and c-Jun. However, under certain circumstances, TNFα might have profibrotic effects in vitro. Myofibroblasts in particular, which are thought to play a central role in fibrotic disorders such as SSc by releasing large amounts of ECM proteins, might react to TNFα differently from resting fibroblasts. While the in vitro studies overall favor TNFα as an antifibrotic cytokine, different in vivo animal studies with antagonists of TNFα and mice lacking TNFRI or both TNFα receptors demonstrated that inhibition of TNFα signaling can prevent fibrosis. However, studies on TNFα-transgenic mice yielded different results, with most suggesting a profibrotic role of TNFα. In our opinion, these differences in the results between in vitro and in vivo studies might be explained by the inflammatory component in animal models of experimental fibrosis (Figure 1). Most of these models depend critically on a strong inflammatory component at the initial stages as a trigger for the later development of fibrosis. As a major proinflammatory stimulus, TNFα has been shown to play an important role in the activation of inflammatory cells. Inflammatory conditions and cellular inflammatory infiltrates are absent in vitro, and TNFα can exert its direct effects on the matrix-producing cells. In our view, a likely scenario is that the direct antifibrotic effects of TNFα on fibroblasts might be outweighed in experimental models of fibrosis by its important role in driving inflammation. Thus, to predict the effects of anti-TNFα treatments in human fibrotic diseases, the key question is whether inflammation triggers and perpetuates the development of fibrosis in humans. The role of inflammation in human fibrotic diseases is less clear than was initially thought; it might differ between the specific organs and might also depend on other parameters such as disease duration. For example, in SSc skin fibrosis, inflammatory infiltrates are thought to play an important role in very early disease stages, but are rarely seen in later stages of the disease. This might predict that anti-TNFα treatments could be a promising antifibrotic strategy for very early inflammatory skin fibrosis in SSc, while it might even be deleterious for later noninflammatory stages of skin fibrosis. Accordingly, the effects on lung fibrosis might be different from those on skin fibrosis. Inflammation is thought to play a more important role in interstitial pulmonary fibrosis associated with SSc, with recurrent alveolitis triggering progression of fibrosis during the course of the disease. If this concept holds true, patients with pulmonary fibrosis and alveolitis might benefit from anti-TNFα treatments, while the effects on lung fibrosis might again be deleterious in patients with no evidence of inflammation/alveolitis. However, it must be emphasized that this concept is theoretical, because so far, no controlled study has looked at the effects of TNFα in inflammatory phases of SSc. Taken together, the results of available molecular and cellular studies do not allow definite conclusions about the role of TNFα in fibrotic diseases. More experimental studies are necessary, with a particular focus on fibrotic animal models that are independent of inflammation as a trigger for the fibrotic process. Cellular studies have lacked a complete description of the transcriptom after TNF or anti-TNF treatment, which would provide a more comprehensive analysis and, together with a proteomics study, could elucidate the contradictory findings reported so far instead of analyzing only selective and expected pathways. Open-label uncontrolled studies with TNFα inhibitors in SSc appear to be promising. However, from the existing literature, there are concerns that treatment with TNFα antagonists could lead to progression of fibrosis. Therefore, TNFα antagonists should not be used in daily clinical practice for the treatment of patients with fibrotic diseases until these concerns are cleared. To elucidate the definitive clinical effects of TNFα antagonists in fibrotic diseases, placebo-controlled trials need to be performed. Patients with inflammatory stages of the fibrotic disease are most likely to respond, while patients with noninflammatory stages of fibrosis might even show deleterious effects. Regarding safety issues, these trials need to focus on fibrosis outcomes in addition to the conventional safety issues such as infection and malignancy, and they need to be designed with sufficient statistical power to exclude potentially harmful effects. © 2008, American College of Rheumatology.