Inhibitor of PD-1/PD-L1: a new approach may be beneficial for the treatment of idiopathic pulmonary fibrosis

Potential risk factors for IPF include genetic mutations, viral infections, lifestyle choices, environmental exposures, and occupational hazards [24]. Smoking, air pollution, and occupational exposure are common risk factors for IPF and LC [25, 26]. The shared pathogenesis of IPF and LC [17, 27] includes uncontrolled cell proliferation [28] and interference with intercellular transmission [29, 30]. Studies have shown that certain cells (e.g. myofibroblasts, cancer-associated fibroblasts [CAFs]) and associated growth factors are equally involved in the disease development of IPF and LC [31]. Of these, myofibroblasts are particularly important in IPF because of their proliferative capacity and their ability to be activated under high fibrotic conditions. These abilities enable myofibroblasts to secrete and deposit ECM proteins, leading to lung fibrosis [32]. TGF-β1 induces the conversion of epithelial cells from epithelial-to-mesenchymal transition (EMT) to fibroblasts/myofibroblasts, thereby promoting fibrous proliferation. Furthermore, alveolar type II epithelial cells have been observed to differentiate into fibroblasts/myofibroblasts and produce the ECM under TGF-β1 stimulation [33].

CAFs are key cellular actors in LC, especially in non-small cell LC (NSCLS) [34]. Several hypotheses have been proposed regarding the cellular origin of CAF in tumors [16], as follows: resident fibroblasts differentiate into CAFs during tumor development under the coordination by certain signaling pathways and cancer-associated adipocytes, which exist in tumor stroma, possibly originating from circulating progenitor cells [35]; bone marrow MSCs and hematopoietic stem cells [36]; epithelial cells via EMT [37]; and vascular endothelial cells are critical to tumor angiogenesis via endothelial-to-mesenchymal transformation [38]. Therefore, TGF-β not only regulates the major molecules that promote fibrosis signaling but also promotes LC progression and cancer cell mitosis. In addition to the mesenchymal phenotype, cell transformation [27, 39] represents a common association between pulmonary fibrosis and carcinogenesis.

Abnormal activation of the following common signaling pathways (Fig. 1) has been reported to occur in both pulmonary fibrosis and lung cancer:

PD-1 is an immunosuppressive molecule mainly expressed in activated T cells and can be enhanced through stimulation with tumor necrosis factor [57]. PD-L1 is the main ligand of PD-1 and is widely expressed in T cells, B cells, dendritic cells, macrophages, and other tissues [58]. Blocking negative regulatory signaling then reactivates T-cell activity and promotes tumor cell killing ability, which is mainly achieved by blocking PD-1 binding to PD-L1 [59]. ICIs primarily exert their antitumor effects by blocking tumor cell escape, helping in repositioning the immune cells and engaging in cytotoxic killing [60]. In particular, PD-1/PD-L1 inhibitors block PD-1 on T cells or PD-L1 on the surface of tumor cells, leading to antitumor effects [61]. At present, PD-1/PD-L1 inhibitors have been approved for the treatment of patients with advanced NSCLC, hepatocellular carcinoma, urothelial carcinoma, metastatic melanoma, and metastatic colorectal cancer [62]. The clinical benefit of PD-1/PD-L1 inhibitors is usually predicted by immunohistochemical determination of PD-L1 expression levels [63]; high expression (> 1 or > 50% in NSCLC) is associated with better response rates [64, 65]. Given the similar pathogeneses and pathway mechanisms of IPF and cancer, we wondered whether ICIs could benefit patients with IPF.

The methods frequently employed to detect PD-1/PD-L1 expression in human specimens include enzyme-linked immunosorbent assay (ELISA) [66], flow cytometry [67], and immunofluorescence [23]. The results of these studies are summarised in Table 1.

One study reported significantly elevated levels of PD-L1 expression in the serum of IPF patients [66], while another study showed the opposite, with no change in PD-L1 expression in the peripheral blood of IPF patients [68]. However, although no alterations were observed in PD-L1 expression in peripheral blood, PD-1 levels in circulating CD4 + T cells and lung tissue in IPF patients were observed to be higher than in age-matched healthy controls in another previous study [67]. Another study observed increased expression of PD-1 and PD-L1 in CD4 + T cells in the peripheral blood of patients [69]. In addition to the determination of PD-L1 expression, studies have also observed changes in the expression of soluble PD-L1 (sPD-L1), with one study showing significantly higher sPD-L1 levels in the serum of IPF patients without surgical biopsy than in healthy controls [66], and another study observing a similar presentation [70]. Given the similarity between IPF and NSCLC, a previous study used lung tissue from patients with lung fibrosis-related cancers as a sample to explore changes in PD-L1 expression [71]. Another previous study showed that PD-L1 was highly expressed on the cell membranes of alveolar and bronchial epithelial cells in lung tissue samples from patients with IPF and pulmonary fibrosis [72]. Alternatively, other studies reported that the PD-L1 expression was not significantly altered in fibroblasts and myofibroblasts in the lung tissues of patients but was overexpressed in alveolar macrophages [66]. Asai et al. reported a significantly higher proportion of PD-1⁺ICOS⁺Tfh cells in patients with IPF [73]. Li et al. observed elevated levels of PD-L1 in lung tissue, alveolar lavage fluid, and serum from patients [74]. Ahmadvand et al. came to the same conclusion for lung tissue and reported that PD-1 was highly expressed in T cells [75].

We noticed that in this table, Bonham et al. [68]found that the expression level of PD-1 did not increase as it did in other studies, which may be related to the fact that the samples used were fresh peripheral blood PBMC, the control group was age-sex matched, and the average fluorescence intensity of PD-1 expression was used in data analysis. Celada et al. [67] and Wang et al. [69] reported the percentage of PD-1⁺CD4⁺ T cells, while Asai et al. [73] reported the proportion of PD-1⁺ICOS⁺Tfh cells in peripheral blood. These results suggest that PD-1/PD-L1 is involved in the process of pulmonary fibrosis. Nonetheless, all studies published to date have been retrospective in design and have small sample sizes; therefore, further research is required. Prospective cohort studies can be designed for further exploration, with as large a sample size as possible, and care should be taken to eliminate bias and confounding factors in the study. Prospective cohort studies can be designed to explore further, with as large a sample size as possible, in addition to being aware of bias and confounding factors in the studies.

The potential role of the PD-1/PD-L1 immune checkpoint in IPF was observed, and a series of preclinical studies were initiated in the hope that this new anti-tumor drug could bring a promising future to the treatment of IPF. The role of PD-1/PD-L1 in the pathogenesis of IPF has been explored through in vivo and in vitro experiments. Table 2 summarizes the results of these studies in animal models of IPF.

In three studies on mouse models of lung fibrosis, elevated levels of PD-L1 expression in lung tissue lesions were observed [23, 76, 77]. Furthermore, aberrant PD-1 expression was observed in a mouse model of pulmonary fibrosis [67, 76, 78]. Celada et al. observed elevated levels of PD-1 in CD4⁺ T cells from the lung tissue of fibrotic mice [67], whereas Cui et al. observed this phenomenon in CD8⁺ T cells [76]. In addition to the elevated expression of PD-1 observed in the lung tissue of fibrotic mice, Wang et al. found elevated expression of this indicator in CD4 + T cells of the peripheral blood of mice [78]. However, none of the above studies measured the specific expression levels of PD-1.

Based on research on human tissue transplantation, increasingly more immunodeficient mice have been used as preclinical animal models over recent years [66, 79‐81]. Some studies have also used humanized mouse models for PD-1/PD-L1 immune checkpoint related studies [21, 76, 82]. Geng et al. used NOD⁻SCID⁻IL2Rgc^−/− (NSG) mice to establish pulmonary fibrosis models and found a significant regulatory effect of PD-L1 on invasive fibroblasts [21]. Additionally, an elevation in PD-L1 levels was observed in a humanized mouse model [76].

In addition to the abovementioned mouse models of bleomycin-induced lung fibrosis, transgenic techniques, and silica can also be utilized to establish mouse models of pulmonary fibrosis for the evaluation of PD-1/PD-L1 expression. Cui et al. detected elevated levels of PD-L1 expression in the lung tissue of IL-6 knockout mice, which is consistent with the results observed for PD-L1 in bleomycin-modelled lung fibrosis mice [76]. Zhao et al. found abnormal levels of expression of immune checkpoint molecules on these cells in a mouse model of silica-induced lung fibrosis, ultimately leading to systemic immune dysregulation [83]. Li et al. observed the same phenomenon as in BLM-induced pulmonary fibrosis in a mouse model of paraquat-induced pulmonary fibrosis, with upregulation of PD-L1 expression in lung tissue [74]. In summary, the aberrant expression levels of PD-1 and PD-L1 in certain lung cell types in a mouse model of pulmonary fibrosis suggest that PD-1/PD-L1 may play a role in the process of pulmonary fibrosis.

We observed differences in the animal models used in these studies. BLM-induced mouse pulmonary fibrosis model is the most commonly used animal model for the study of IPF, which can demonstrate the pathogenesis of human IPF [84], but it has certain limitations because its fibrosis resolves spontaneously after 28 days [85] The use of Silica in the lungs also causes persistent fibrosis, but is more similar to silicosis in humans than pulmonary fibrosis [85]. Systemic administration of paraquat causes fibrosis-like changes in the lungs of animals, but other organs (e.g., liver, kidney, etc.) also fail, increasing mortality rates [86]. Humanized mouse models can better elucidate the mechanism of human fibroblasts inducing pulmonary fibrosis in vivo, but the main issue is the availability of cells from human patients with IPF [87]. Although different modeling methods were used, the results showed increased expression levels of PD-1/PD-L1, suggesting that PD-1/PD-L1 was involved in the process of pulmonary fibrosis. Therefore, the researchers conducted further studies.

Current studies have shown that the fibrotic process is a combination of multiple cells, growth factors, and fibrosis-associated proteins, and their interactions determine the outcome of fibrosis [88]. The role of PD-1/PD-L1 in IPF can be explored from two perspectives.

Mechanistic experiments first recognized that PD-1⁺CD4⁺ T cells can promote pulmonary fibrosis and TGF-β production mainly through IL-17A, where T helper 17 (Th17) cells are the main CD4⁺ T cell subset expressing TGF-β. In an in vivo animal experiment, blocking PD-1 expression significantly reduced the expression of STAT3, IL-17A, and TGF-β on Th17 cells, thereby reducing the production of type I collagen by fibroblasts and attenuating lung fibrosis [67]. Similar results were observed in another study [78]. In addition, a study confirmed that maintaining the functional homeostatic balance between Th17/Treg cells is an important part of regulating autoimmunity and treating cancer [89]. Tregs can up-regulate the expression of TGF-β [90], and IL-6 allows Treg to regain the characteristics of Th17 cells [91]. Elevated IL-6 levels were also observed in patients with IPF. Therefore, we hypothesized that elevated IL-6 leads to the subpopulation conversion of Treg to Th17 cells, which increases the production of IL-17 and TGF-β and promotes lung fibrosis (Fig. 2A). Another study indicated that an increase in PD-1/PD-L1 inhibited CD4⁺ T cell differentiation to Treg cells, thereby reversing the effect of TGF-β on Treg cell differentiation. PD-1 promoted type I collagen production in myofibroblasts while decreasing the proliferation of myofibroblasts co-cultured with CD4⁺ T cells [69] (Fig. 2A). Upregulation of PD-L1 expression on fibroblasts has also been observed in studies on lung fibrosis, especially in invasive fibroblasts, which can promote lung fibrosis [21, 23, 76, 77]. We next explored the mechanism of the involvement of PD-L1 in IPF in fibroblasts.

Research suggests that high expression of PD-L1 on lung fibroblasts can be regulated through the p53 and FAK pathways. Knockdown of PD-L1 in invasive fibroblasts attenuates the severity of lung fibrosis in lung fibroblasts and mice [21]. Evidence also supports that p53 expression levels in myofibroblasts are significantly lower than those in normal lung fibroblasts [92, 93]. PD-L1 knockdown in IPF fibroblasts and targeting of PD-L1 by FAK inhibition or PD-L1-neutralizing antibodies blunts invasion and attenuates fibrosis [21]. In addition, the acquisition of p53 function could sensitize lung fibroblasts to apoptosis and prompt macrophages to clear apoptotic myofibroblasts, thereby attenuating pulmonary fibrosis symptoms in a mouse model of pulmonary fibrosis [93]. Therefore, PD-L1 on lung myofibroblasts may lead to IPF by inhibiting the p53 pathway, leading to myofibroblast resistance to apoptosis, and inhibiting phagocytosis by macrophages, leading to overproliferation of myofibroblasts (Fig. 2B). Recently research suggests that knockdown of PD-L1 in primary human lung fibroblasts significantly attenuated the expression of myofibroblast markers induced by TGF-β. TGF-β could induce PD-L1 to interact with Smad3, suggesting that PD-L1 may act as a cofactor for Smad3 to promote the process of lung fibrosis. PD-L1 knockdown attenuated TGF-β-induced GSK3β phosphorylation/inhibition and β-catenin upregulation, suggesting PD-L1-mediated GSK3β/β-catenin signaling in FMT [23]. Another study also showed a relation between the GSK3β/β-catenin signaling pathway and the progression of IPF [90]. Thereby, PD-L1 in lung fibroblasts was upregulated to promote the development of IPF through the Smad3 pathway and the β-catenin pathway (Fig. 2B).

Autophagy is also involved in the development of IPF. Inhibition of autophagy can induce myofibroblast proliferation and ECM deposition, leading to pulmonary fibrosis [94, 95]. The application of autophagy activators can attenuate the extent of pulmonary fibrosis [96]. It was found that anti-PD-L1 monoclonal antibody down-regulated the PI3K/AKT/mTOR pathway to induce autophagy, thereby inhibiting TGF-β1-induced lung fibroblast invasiveness and ECM deposition [77]. Inhibition of PI3K inhibits AKT phosphorylation and TGF-β-induced proliferation of fibroblasts as well as expression of markers in myofibroblasts [97]. Further, mTOR acts as a downstream molecule of the PI3K/AKT pathway, and inhibition of the PI3K/AKT pathway inhibits this target, upregulates myofibroblast autophagy and attenuates lung fibrosis [98] (Fig. 2C).

The mechanism of action of the PD-1/PD-L1 axis in IPF is not yet fully elucidated and further exploration is needed. We summarize the results of current PD-1/PD-L1 inhibitors in IPF studies below to explore the feasibility of targeting PD-1/PD-L1 for the treatment of IPF.

The majority of studies revealed a reduction in the extent of lung tissue lesions with anti-PD-L1 mAb in mouse models of pulmonary fibrosis [21, 67, 76, 77, 83]. Cui et al. observed that pulmonary fibrosis was reduced in a mouse fibrosis model after blocking PD-L1, thus providing a potential treatment strategy for patients with pulmonary fibrosis [76]. By establishing a humanized severe combined immunodeficiency IPF model, Geng et al. suggested that targeting PD-L1 during the early and late stages might significantly reduce the invasion of IPF fibroblasts and could not only eliminate the progression of pulmonary fibrosis but also reverse it [21]. Lu et al. found that anti-Pd-L1mAb ameliorated lung tissue disorganization and collagen deposition in mice with pulmonary fibrosis while demonstrating that it possesses the ability to activate autophagy in pulmonary fibrosis [77]. Zhao et al. conducted a similar study on silica-induced pulmonary fibrosis in mice and discovered that PD-1/PD-L1 signal inhibition significantly improved silicosis [83]. Current research is not limited to single-drug studies of PD-1/PD-L1 pathway inhibitors, and some researchers are interested in combination therapy. A study combined pirfenidone with a PD-L1 blocker to explore its treatment effects in mouse models of pulmonary fibrosis and LC [99]. Combined therapy with pirfenidone and PD-L1 blockers promoted immune cell invasion and inhibited tumor outgrowth while improving the prognosis of mice. This combination treatment regimen reduced lung fibrosis in mice. Some studies have also pointed out that human MSCs contribute to immune regulation, which has been used in several preclinical studies on IPF treatment [100]. The efficacy of ICIs in the aforementioned mouse models of pulmonary fibrosis demonstrates that PD-L1 inhibitors have a promising effect on alleviating pulmonary fibrosis.

The researchers also designed in vitro experiments to determine whether inhibition of PD-L1 reduces lung fibrosis. The in vitro experiments demonstrated that blocking the PD-1/PD-L1 pathway can inhibit STAT3 expression in CD4 + T cells, leading to low expression of IL-17A and TGF-β [67]. A previous report showed that CD274 knockdown in IPF fibroblasts by FAK inhibition or CD274-neutralizing antibodies, reduced invasion and alleviated fibrosis, whereas PD-L1 expression on invasive fibroblasts promoted lung fibrosis [21]. Inhibition of PD-1 appears to be less effective than inhibition of PD-L1 in preclinical IPF models, which may be related to the genetic ablation of PD-1 [67]. Given the elevated levels of immune checkpoint expression in patients with IPF, novel ICIs may provide new immunotherapeutic options.

The above preclinical studies have shown that inhibition of immune checkpoint pairs can attenuate the progression of pulmonary fibrosis, and that inhibition of PD-L1, in particular, can produce an antifibrotic effect. However, PD-1/PD-L1 inhibitors cannot be assumed to directly lead to results for the clinical treatment of IPF. While the fibrotic process is a complex process in which antifibrotic mediators interact with pro-fibrotic mediators in a microenvironment composed of different cell types [76], immune tolerance imbalances and the occurrence of immune-related adverse events (irAEs) require attention. Therefore, although preclinical studies have shown that targeting PD-1/PD-L1 immune checkpoints attenuates pulmonary fibrosis, more evidence is needed to support how these preclinical studies can be transferred to practical clinical applications.

ICIs, while enhancing the normal immune response, may enhance the anti-tumor effects of cellular immunity, leading to an immune tolerance imbalance and immune-related adverse events (irAEs). IrAEs can involve all organs throughout the body and are common, with data from one study showing that over 80% of patients using ICIs developed irAEs in some systems [101, 102]. In the targeted treatment of ICIs for LC, immunosuppressant-related pneumonia requires consideration. Studies have found that the use of ICIs in patients with LC with ILD is associated with a higher risk of developing immune checkpoint inhibitors related pneumonitis (CIP) than that in patients without ILD, but these studies suggest that CIP is not associated with increased mortality [103‐105]. The response rate, progression-free survival, and overall survival of NSCLC patients with concomitant ILD treated with ICIs were similar to those of patients without ILD, suggesting that patients with NSCLS treated with ILD could also benefit from ICI treatment [103]. However, since the presence of interstitial pneumonia is currently regarded as the exclusion criterion for ICI clinical trials [106, 107], direct evidence is lacking to prove the effect of ICIs on pulmonary fibrosis symptoms. Currently, no studies have directly validated the utility of PD-1/PD-L1 inhibitors on pulmonary fibrosis, and conclusive experimental evidence to support their therapeutic value is scarce. Studies have shown that ICIs in patients with IPF combined with squamous cell carcinoma, the addition of nidanib may prevent drug-induced pneumonia or acute exacerbation of IPF [108]. Another study showed that pirfenidone combined with PD-L1 inhibitors in the treatment of tumor pulmonary fibrosis in mice significantly delayed tumor growth, reduced pulmonary fibrosis, and improved mouse survival. This suggests that pirfenidone could be used as an adjunct to immunotherapy in cancer treatment [99]. Therefore, the study of the potential mechanism of irAEs not only contributes to the immunotherapy of tumors but also plays an important role in the treatment of IPF. Whether the combination of ICIs and antifibrotic drugs (such as nidanib and pirfenidone) can delay the pathogenesis of IPF can be considered as a research direction. In addition to investigating whether ICIs can be used to improve disease progression in patients with IPF, future studies should also clarify its role in prognosis.