Fibrinogen-like protein 1 (FGL1): the next immune checkpoint target

Immune checkpoint therapy has achieved significant efficacy by blocking inhibitory pathways to release the function of T lymphocytes. In the clinic, anti-programmed cell death protein 1/programmed cell death ligand 1 (PD-1/PD-L1) monoclonal antibodies (mAbs) have progressed to first-line monotherapies in certain tumor types. However, the efficacy of anti-PD-1/PD-L1 mAbs is still limited due to toxic side effects and de novo or adaptive resistance. Moreover, other immune checkpoint target and biomarkers for therapeutic response prediction are still lacking; as a biomarker, the PD-L1 (CD274, B7-H1) expression level is not as accurate as required. Hence, it is necessary to seek more representative predictive molecules and potential target molecules for immune checkpoint therapy. Fibrinogen-like protein 1 (FGL1) is a proliferation- and metabolism-related protein secreted by the liver. Multiple studies have confirmed that FGL1 is a newly emerging checkpoint ligand of lymphocyte activation gene 3 (LAG3), emphasizing the potential of targeting FGL1/LAG3 as the next generation of immune checkpoint therapy. In this review, we summarize the substantial regulation mechanisms of FGL1 in physiological and pathological conditions, especially tumor epithelial to mesenchymal transition, immune escape and immune checkpoint blockade resistance, to provide insights for targeting FGL1 in cancer treatment.

Recent studies have focused on one of the most promising immune checkpoints. LAG3 (CD223) is an inhibitory receptor expressed mainly on the surface of T lymphocytes [26,42,43]. The binding of LAG3 with its ligands delivers negative signals to activated T cells, preventing immune-mediated tissue damage [44]. Similar to PD-1/ PD-L1 [5,45,46], LAG3 has an inhibitory function based on interactions with its ligands, which include major histocompatibility complex II (MHC II) [47,48], galectin-3 [49], liver sinusoidal endothelial cell lectin (LSECtin) and FGL1 [12,50]. FGL1, also known as liver fibrinogenrelated gene-1 (LFIRE-1)/Hepassocin (HPS) or hepatocyte-derived fibrinogen-related protein-1 (HFREP-1) [51][52][53], is a proliferation-and metabolism-related factor secreted by the liver [54][55][56]. It has recently emerged as a novel ligand of LAG3 beyond the classic ligand MHC II and can bind with LAG3 to form a new immune checkpoint pathway independent of PD-1/PD-L1, which results in T cell depletion and subsequent dysfunction, as well as tumor cell escape from immune surveillance [50,57]. Apart from its relatively high expression in the liver and pancreas, FGL1 is upregulated in tumor tissues (including lung, prostate, melanoma, colorectal, breast and brain tumors) based on several datasets [50]. Hence, FGL1 has potential as another immune checkpoint target in clinical practice, especially in targeted therapy for non-small cell lung cancer (NSCLC) [58][59][60][61][62][63][64][65]. In this review, we summarize the function and molecular mechanism of FGL1 in the regulation of cancer development and metastasis and provide promising applications in therapeutic strategies for malignant tumor treatment.

Structure and distribution of FGL1
FGL1 is located on human chromosome 8 (8p22-21.3) and is a 68-KD protein comprised of a disulfide bondlinked homodimer [66,67]. Its carboxyl terminus contains the β and γ subunits, which are highly homologous to fibrinogen but irrelevant to coagulation-related binding sites [50]. Under normal physiological conditions, FGL1 is secreted mainly by hepatocytes in the liver (some of which may also exist in the pancreas) [50]. It is now clear that FGL1 is the product of hepatocyte regeneration and participates in hepatocyte mitosis and liver energy utilization (including lipid metabolism and blood glucose regulation) [54,68,69]. Apart from the above functions, FGL1 can also be detected in the plasma as an acute reactant [70], implying that FGL1 secreted by the liver acts not only on hepatocytes (autocrine) but also on other tissues, such as muscle and brown adipose tissues (telecrine) (Fig. 1a) [56,68]. Under stimulation by metabolic factors (hyperglycemia, hyperlipidemia, hormones, etc.), the liver secretes FGL1 and participates in the blood circulation [52,69,71]. This function of FGL1 acts on brown adipose tissue [54,56], regulates body productivity and maintains body temperature. Moreover, FGL1 also acts on muscle tissue and affects the sensitivity of myoblasts to insulin [52,68].

Expression, regulation and function of FGL1 in benign disease
Given that FGL1 is a physiological secretory factor of the liver, its expression can be regarded as a self-protective mechanism acting against exogenous injury or stimulation. FGL1 expression is regulated by radiation-induced lung, liver and stomach injury [71][72][73]. In addition, factors released by hepatectomy and postoperative inflammation (interleukin (IL)-6 and transforming growth factor (TGF)-β) can also affect the expression of FGL1 [70,[73][74][75][76][77]. Under such exogenous induction, normal hepatocytes regulate FGL1 transcription via signal transducer and activator of transcription 3 (STAT3) and hepatocyte nuclear factor-1α (HNF-1α) (Fig. 1b) [78][79][80][81][82][83]. FGL1 expression can also be regulated in chronic medical diseases, such as hyperlipidemia, insulin resistance, and hyperglycemia crisis. Hitherto, there have been studies on the mechanism of the FGL1-mediated regulation of substance metabolism in chronic diseases (Table 1). In clinical practice, changes in the FGL1 expression level can be specifically observed in type 2 diabetes patients with non-alcoholic liver disease [52]. Recently, plasma tests in these patients have found that FGL1 is upregulated when insulin resistance occurs. The high concentration of FGL1 in the plasma acts not only on itself but also on both adipocytes and myoblasts (Fig. 1a). In a hyperlipidemic model, palmitate-treated primary hepatocytes regulated the expression of FGL1 through CCAAT/ enhancer-binding protein β (C/EBPβ)-mediated transcription, and the expression level of FGL1 in the liver of mice fed a high-fat diet (HFD) was also increased [68]. This high level of FGL1 could act on C2C12 cells (myoblasts) and generate insulin resistance through the phosphorylated JNK pathway [68]. In addition, in patients with type 2 diabetes, FGL1 induces insulin resistance in HepG2 cells via the hepatocyte-dependent extracellular regulated protein kinase (ERK)1/2 pathway [52]. Furthermore, in adipocytes, extracellular FGL1 induces adipogenesis through the canonical ERK1/2-C/ EBPβ-peroxisome proliferator-activated receptor (PPAR) γ pathway [56].
Apart from protecting hepatocytes (promoting regeneration and apoptosis) and regulating metabolism, FGL1 expression can also be used as a marker of some benign diseases (autoimmunity, infectious diseases, acute inflammation, etc.). First, the value of FGL1 can predict the activity of rheumatoid arthritis (RA; moderate/high: 91.46%; remission/low: 80.77%) and the severity of dengue fever [84,85]. Second, the fluctuation in FGL1 expression in vivo can also affect the progression of obesity and malaria parasite infection [56,86]. These above studies reveal that FGL1 plays important roles and acts as a potential biomarker in several common benign diseases. The mechanism by which hepatocytes express and secrete FGL1 under physiological and pathological conditions. a FGL1 has two origins, with the liver being the classic source. FGL1 secreted by the liver can be distributed in adipose tissue and circulate in the blood. Tumor tissues, such as cancerous lung tissues (e.g., NSCLC), were recently identified as a new source of FGL1; FGL1 can also be detected on the surface of CTCs in NSCLC patients. FGL1 acts on both adipocytes and myoblasts via exerting telecrine effects. b The mechanism of normal hepatocyte expression and secretion of FGL1 under physiological conditions: In normal hepatocytes, the JAK2-STAT3 pathway is activated by either the inflammatory factor IL-6 or radiation. HNF1α forms a complex by binding to the HMGB1 and CREB proteins in the cytoplasm. Then, it enters the nucleus and binds to the HPS promoter through pSTAT3, thereby upregulating FGL1 transcription. The overexpressed FGL1 protein is distributed via an autocrine process. c The mechanism of low FGL1 expression in HCC: The IL-6-JAK2-STAT3 pathway, which induces FGL1 expression, can be blocked by oxysophocarpine (OSP). Deletion of HNF1α and inhibition of the AKT-mTOR pathway downregulate FGL1 via an endocrine mechanism Qian et al. J Hematol Oncol (2021) 14:147

FGL1 is upregulated in solid tumors
At present, studies on the regulation of FGL1 expression are not limited to the exploration of its physiological and pathological functions [87]. The expression of FGL1 in human solid tumors is different from that in paracancerous tissues. According to the bioGPS tissue microarray database and a proteomic analysis, both the mRNA and protein expression levels of FGL1 are mainly confined to the normal liver and pancreas in humans. A meta-analysis of Oncomine datasets showed that FGL1 expression was upregulated in lung, prostate, melanoma, colorectal, breast cancer and brain tumors but downregulated in pancreatic, breast, liver and head and neck cancers. Among these changes, the trends observed in lung, prostate and liver tumors were consistent with those in The Cancer Genome Atlas (TCGA) datasets, while FGL1 was significantly upregulated in the lung adenocarcinoma cancer map (FGL1 ranked 38th among the top 200 highly expressed genes in lung cancer). In addition, multiple quantitative immunofluorescence (QIF) staining was performed on a tissue microarray of 275 NSCLC samples to detect FGL1 protein expression in cells and tissues ( Table 2). The data showed that the FGL1 protein was expressed in local pan-keratinpositive tumor cells but the stromal compartment exhibited almost no expression. These above results indicate that the expression of FGL1 is upregulated in human tumors, as represented by NSCLC [50]. Moreover, FGL1 is downregulated in hepatocellular carcinoma (HCC) in TCGA datasets. This result was also confirmed in a recent study showing that the expression of FGL1 could be downregulated due to the deletion of HNF1α in HCC [51,67,78] (Fig. 1c). Hence, FGL1 is released at lower levels by HCC tumor tissues but at higher levels in other solid tumors.

FGL1 mediates EMT process in tumors
The process of tumor progression is accompanied by changes in angiogenesis, invasion and migration [88]. Tumor epithelial to mesenchymal transition (EMT) is not only reflected in changes in cell morphology but also closely related to the behaviors of tumor cell invasion and migration [89][90][91][92]. Previous preclinical studies have confirmed the correlation between FGL1 and the tumor EMT process in lung and gastric cancers [93,94]. First, following FGL1 silencing in liver kinase B1 (LKB1)-mutant lung adenocarcinoma cells (A549 and H157 cells), E-cadherin was shown to be downregulated, while N-cadherin and vimentin were upregulated [93,95,96]. However, another opposite conclusion was also reached: E-cadherin was upregulated and the expression of N-cadherin and vimentin was suppressed when FGL1 was knocked out in SGC-7901 gastric cancer cells [94,[97][98][99][100][101]. Hence, the relationship between FGL1 and EMT progression in tumor cells still needs to be further investigated. In addition, a correlation between FGL1 and the EMT process has also been reported in a pulmonary fibrosis model, which is distinct from tumor models. Knockout of FGL1 impeded the processes of radiation-or TGF-β-induced EMT [73,89]. After silencing the FGL1 gene in L132 cells, EMT markers (increased: Snail [102], Twist [103], MMP12 and fibronectin [104][105][106]; decreased: ZO-1 [107,108]) changed at both the mRNA and protein levels. These above studies demonstrate that there is a correlation between FGL1 and the EMT process in tumors but the conclusions are not consistent. In the presence of FGL1, interstitial markers are downregulated, confirming that FGL1 modulates EMT under radiation or inflammatory induction [109]. This effect may be closely related to the protective effect of FGL1 in response to injury stimulation. As tumor EMT is tightly associated with prognosis and drug resistance [110][111][112][113], the specific mechanism by which FGL1 regulates tumor EMT needs to be clarified (Fig. 2).

FGL1 regulates other tumor characteristics
Apart from mediating the EMT process, FGL1 is also involved in tumor proliferation, apoptosis, radiation and drug sensitivity [71,72,[114][115][116]. Tumor growth is closely correlated with cell proliferation and apoptosis. The clonal proliferation ability of PC9/GR cells was evaluated in a study on targeted drug resistance in NSCLC, and the data showed that the proliferation of PC9/GR cells decreased significantly after FGL1 silencing [114]. Knockout of FGL1 could significantly dampen the proliferation of SGC-7901 cells [94]. It has been reported that the activation of mTOR is tightly related to the proliferation of HCC cells [117]. As an inhibitor of HCC, FGL1 regulates the occurrence of HCC mainly by inhibiting AKT and downstream pathways, indicating that AKT-mTOR is an important pathway downstream of FGL1 involved in regulating cell proliferation [67,118]. On the other hand, in the mechanism of FGL1-mediated gefitinib resistance, FGL1 was found to regulate apoptosis through the PARP1/caspase-3 pathway [114], which also confirmed that FGL1 is pertinent to the sensitivity of tumor cells to targeted EGFR-tyrosine kinase inhibitors (TKIs) (Fig. 2). A recent study on FGL1-mediated targeted therapy resistance have also been reported in HCC [116]. In addition, although FGL1 is a mediator of radiation injury [71][72][73], the correlation between FGL1 and radiation sensitivity needs to be confirmed by additional studies.

FGL1 is a major immune inhibitory ligand of LAG3
FGL1 was first identified as a new high-affinity ligand for the inhibitory receptor LAG3 by Dr. Chen's team in 2019 [50]. LAG3 is a type I transmembrane protein that exists mainly on the surface of activated T cells (including CD4 + and CD8 + T cells and Tregs) [146][147][148], NK cells [149], plasmacytoid dendritic cells (pDCs), and so on. The extracellular domain of LAG3 is 20% similar to that of CD4 [12], which determines the high affinity between LAG3 and MHC II. LAG3 interacts with MHC II and generates a negative signal (blocking the TCR activation signal) to restrict T helper 1 (Th1) cell activation, proliferation and secretion [150,151]. Conversely, activated T cells expressing LAG3 stimulate the production of cytokines (tumor necrosis factor (TNF)-α and IL-12) in antigen-presenting cells (APCs) (DCs and monocytes) and induce their maturation and activation [152]. Notably, researchers found that junction peptides between the D4 transmembrane region and transmembrane region could be cleaved to generate soluble LAG3 (s-LAG3) [12,153,154], which could activate thymic epithelial cells and increase CD4 + T cell levels through the transduction of MHC II signals (Fig. 3).
In some subsequent studies, it was confirmed that MHC II is not the only ligand of LAG3 [43]. Most recently, evidence has demonstrated that FGL1 is another type of high-affinity immunosuppressive ligand that binds to LAG3. A number of basic research studies have Fig. 2 Characterization of tumor cells regulated by FGL1. FGL1, which is expressed in tumor cells, regulates cell proliferation, apoptosis, EMT, drug resistance and radiation sensitivity. FGL1 regulates downstream signaling pathways related to cell proliferation by modulating the AKT-mTOR pathway. FGL1 mediates EMT by regulating epithelial markers (E-cadherin and ZO-1) and mesenchymal markers (N-cadherin, vimentin, Snail, Twist, MMP12 and fibronectin). In addition, FGL1 mediates PARP1-caspase apoptotic pathways by affecting EGFR phosphorylation. Radiation sensitivity may be pertinent to TGFβ signaling. The processes of EMT and tumor cell apoptosis are closely related to drug resistance revealed that there is a high-affinity interaction between LAG3 and FGL1 [13]. In the Genome-Scale Receptor Array (GSRA) system, FGL1 was the main binding protein of LAG3-Ig, and the interaction between them was stable, specific and conserved across species. A further domain deletion study showed that the fibrinogen-like domain (FD) in FGL1 and the D1-D2 domain in LAG3 are involved in the MHC II-independent interaction between FGL1 and LAG3. Based on in vivo experiments, loss of the FGL1/LAG3 interaction could be achieved by gene knockout or antibody blockade, and antitumor immunity could be promoted by stimulating tumor-infiltrating lymphocyte (TIL) activation and expansion in the TME [50].

Synergistic inhibition of T cells with PD-1 and FGL1 as a LAG3 ligand
Although the specific mechanism by which FGL1/LAG3 modulates T cell functions remains unclear, the synergistic inhibitory effect of FGL1 blockade in conjunction with anti-PD-1 therapy has been confirmed in animal models [50]. It has been indicated that blocking FGL1 can cooperate with anti-PD-1/PD-L1 therapy to suppress the effect of PD-1/PD-L1 signaling, which is based on previous basic research studies on the synergistic inhibitory activities [146,153,[155][156][157]. The synergistic inhibitory effect of LAG3 and PD-1 on T cells has been confirmed in a variety of tumors, including NSCLC [158], melanoma [159], renal cell carcinoma (RCC) [160,161], head and neck cancer [6], and breast cancer [162,163]. Among these diseases, melanoma shows a close relationship between the coexpression of the immune checkpoint molecules LAG3 and PD-1 and the expression of CD163 and density of TAMs [129,164]. These preclinical results have paved the way for combinatorial blockade of PD-1 and LAG3 in clinical trials [165]. Most importantly, FGL1 will be a potential biomarker for predicting the outcome of PD-1/PD-L1 blockade therapy, since high plasma FGL1 levels were reported to be significantly correlated with a worse therapeutic response to anti-PD-1/ PD-L1 therapy in NSCLC and melanoma patients [50,166]. Therefore, FGL1 can be identified as a next-generation cancer immunotherapy target capable of a functional interaction with the LAG3 pathway and synergistic inhibition of T cells with PD-1.

FGL1/LAG3 is an immunosuppressive pathway independent of PD-1/PD-L1
A series of immunological animal experiments have confirmed the synergistic effect of the inhibitory receptors LAG3 and PD-1 on T cells [146,157]. However, the FGL1/LAG3 pathway also plays an immunosuppressive role independent of the PD-1/PD-L1 pathway. It is well known that the PD-1/PD-L1 inhibitory pathway produces exhausted T cells (Fig. 4a). Anti-PD-1/PD-L1 antibodies can block this immune brake and release the antitumor activity of T cells (Fig. 4b). However, due to the existence of another inhibitory receptor, LAG3, on the T cell surface, another newly characterized immune checkpoint is generated when LAG3 interacts with its ligand FGL1 (Fig. 4c). It has been reported that blocking LAG-3 is more significant in T cell activation, proliferation and IFN-γ secretion than blocking the PD-1 pathway [155]. Under tolerance conditions, signal transduction through the PD-1/PD-L1 and FGL1/LAG-3 pathways have different functional consequences for CD8 + T cell subsets; that is, external signaling initiation can affect the tolerance of CD8 + T cell subsets through the LAG3 and PD-1 pathways [164]. The most direct manifestation upstream of these consequences is the production of three CD8 + T cell subsets with inconsistent expression and localization of LAG3 and PD-1. Further cytolytic functional analysis showed that the T cell subsets with high expression of LAG3 produced higher levels of cytokines, especially IFN-γ, TNF-α and CD107, than those with high expression of PD-1 but were not associated with the expression of the costimulatory molecules inducible T cell co-stimulator (ICOS) and 4-1BB (CD137) [164,[167][168][169]. The differences in cytokine production patterns mentioned above indicate the independence of the two pathways. In addition, in animal experiments, more than 30% of MC38 tumor-bearing mice were found to have no tumor formation within 150 days of treatment with anti-FGL1/LAG3 as a monotherapy or in combination with anti-PD-1/PD-L1 therapy [50]. Because of the high affinity between FGL1 and LAG3, FGL1/LAG3 and PD-1/PD-L1 can regulate T cells independently, and blocking both these checkpoints can produce synergistic antitumor effects. However, the upregulation of LAG-3 expression in NSCLC is associated with insensitivity to PD-1 axis blockade and a poor prognosis [170], which suggests the independence of immune escape pathways and the potential of the synergistic action of anti-PD-1 and anti-LAG3 antibodies in clinical trials.

Anti-FGL1 may overcome immune checkpoint blockade resistance via a tumor-intrinsic mechanism
Based on the above evidence that the FGL1/LAG3 pathway plays an important role in immune evasion during cancer development, anti-FGL1 may help overcome cancer immunotherapeutic resistance as a promising novel checkpoint. Characteristics that are necessary for the immune checkpoint inhibitor (ICI) response include IFN-γ signaling, antigen-presenting signaling, immuneevasive oncogenic signaling, and a mutational signaling landscape [171,172]. It is well known that the T cell receptor (TCR) on the tumor-reactive T cell surface recognizes tumor neoantigens and subsequently releases IFN-γ, which binds to the IFN-γ receptor (IFN-γR1/R2) on the tumor cell surface, activates downstream JAK-STAT signaling pathways and further initiates PD-L1 transcription in tumor cells [171,173,174]. The plasma level of IFN-γ was reported to be obviously increased after the injection of CD8 + OT-1T cells together with the FGL1 antibody 177R4 into syngeneic mice (Table 3), which further proves that the blockade of FGL1 by a monoclonal antibody stimulates T cell immunity in a manner similar to anti-LAG3 therapy [50]. Moreover, FGL1 is highly secreted from tumor cells, and higher plasma levels of FGL1 are associated with resistance to ICIs and poor prognosis in cancer patients [50]. Hence, using FGL1 agents can theoretically enhance the antitumor effect of T cells and thus affect the acquired resistance of ICIs in patients with differential levels of PD-L1 expression. Furthermore, antigen presentation blockade is another intrinsic factor underlying ICI resistance, which is mainly caused by MHC deficiency and loss of tumor antigenicity [175,176]. MHC II is the first identified ligand of LAG3 [177]. However, their binding does not affect antitumor activity, suggesting that MHC II is not solely responsible for the function of LAG3. More valuable ligands presumably restart the "immune brake" by binding with LAG3. Therefore, as a newly discovered high-affinity ligand of LAG3, FGL1 has the potential to induce ICI resistance in a receptor-ligand interdependent manner [50]. Moreover, cancers with a high tumor mutation burden (TMB) and high neoantigen expression, such as melanoma and NSCLC, are generally more sensitive to immune checkpoint blockade [60,173,178]. However, these patients were observed to have higher plasma levels of FGL1 expression after the acquisition of ICI resistance [50]. Thus, the FGL1 secretion level in plasma is a potential biomarker for identifying patients who will not benefit from ICI treatment. Finally, Wnt/β-catenin is an essential signaling pathway in cancer development [179,180]. Emerging evidence indicates that the Wnt5α-β-catenin-PPARγ signaling pathway drives a metabolic program that triggers dendritic cell (DC) tolerance and immunotherapy resistance [181,182]. PPARγ is the ultimate target of FGL1-induced lipid synthesis in adipocytes, suggesting that FGL1 has the potential to induce immunosuppression by regulating PPARγ synthesis in metabolic programs [56].

Anti-FGL1 may overcome immune checkpoint blockade resistance via exogenous factors
In addition to intrinsic factors, alterations in the functions and numbers of immune effector cells (mainly CD8 + T cells) in the tumor microenvironment (TME) are challenging for immunotherapies [183]. Various immune cells (DCs, Tregs, myeloid-derived suppressor cells (MDSCs)) and their related secretory products and metabolites in the TME play essential roles in tumor development, metastasis and immunotherapeutic response [184]. Persistent antigen exposure in the TME clearly results in sustained LAG3 expression, thereby having a substantial effect on the immunosuppression status and cytokine production. As a major LAG3 functional ligand independent of MHC-II, FGL1 is positively correlated with MDSC and Treg populations and negatively correlated with CD8 + T cells as determined by analysis of the TIMER database (http:// timer. cistr ome. org/). The regulatory effect of FGL1 on the TME was further demonstrated in a breast cancer model constructed using biomimetic nanomaterials designed to deliver a short interfering RNA targeting FGL1 [185]. Moreover, FGL1 was shown to be closely associated with tumor cell invasion and metastasis via acquisition of the EMT phenotype, which was underlined by bidirectional crosstalk between tumor cells and the surrounding TME [93,94]. This evidence indicates preferential suppression of T cell immunity in the TME upon FGL1-LAG3 signaling. Indeed, the current insights into the mechanism by which FGL1 induces immunotherapy resistance via immunosuppression of the TME are relatively limited, and more in-depth investigations are still needed (Fig. 5).

Potential toxicity patterns of anti-FGL1 therapy
Based on clinical trials and medical practices, ICI therapies not only initiate the activation of an anticancer immune response but also cause a wide range of immune-related adverse events (irAEs), and the costs of immunotherapies continue to increase due to the diversity of T cell expansion and cell infiltration [186]. Different from the toxic effects of traditional chemotherapy, the most frequently affected organs are the skin, digestive system (gastrointestinal, liver) and endocrine system, with symptoms of nausea, diarrhea, pruritus, rashes and thyroid problems, and fatal irAEs include myocarditis, pneumonia, hepatitis and meningitis [187]. The frequency of irAEs and potential  toxicity patterns are dependent on the immunotherapy strategy, exposure time, medicinal dose administered and individual intrinsic risk factors of the patient [186]. Some novel emerging immune checkpoints are approved for clinical trials in combination with CTLA-4 or PD-1(L1) blockade, such as anti-LAG-3, which increases the probability and new descriptions of irAEs (Table 4). Multiple new agents have been identified as targeting LAG3 across various tumors [188]. The most common agent-related AEs, fatigue and nausea, are well tolerated. Nonetheless, the discovery of more novel checkpoints is still urgently needed in the paradigm of cancer treatment. Among the potential candidates, FGL1 has promise despite that it remains in the preclinical stage, and researchers are awaiting more advanced data. Theoretically, as another ligand of LAG3 in addition to the canonical ligand MHC-II, FGL1 may have more advantages than anti-LAG3 in regards to irAEs. Future studies should also optimize the selection of new ICIs by screening potential predictive biomarkers for the irAE risks [189].

Future directions of anti-FGL1 in cancer therapy
Despite the early stage of tumor research, FGL1 is a new immune checkpoint molecule that is thought to have a promising future in clinical applications, especially in NSCLC immunotherapy [190][191][192][193], which is attributed to its overexpression in NSCLC cells and close correlations with immune regulation, tumor neovascularization [194], EMT progression [195], resistance and metastasis [116,196,197]. The novel discovery of a high-affinity interaction between FGL1 and LAG3 in immunology marks a major breakthrough in research on immune checkpoint blockade therapy.
This strategy holds great potential as a third-generation immune checkpoint blockade after targeting CTLA-4 and PD-1/PD-L1 [198]. Currently, multiple clinical trials evaluating the synergistic effects of anti-LAG3 and anti-PD-1 antibodies are ongoing, and significant efficacy has been achieved in completed trials on melanoma (stage III/IV), metastatic breast cancer and RCC (stage IV) [163,165]. Although the number of ongoing clinical trials targeting FGL1 are not sufficient, a significant positive correlation between FGL1 expression and long-term prognosis has been observed for patients with multiple types of metastatic cancer who are treated with anti-PD-1/PD-L1 therapy ( Table 2). Based on the above results, patients with advancedstage tumors reliant on ICIs are likely to benefit from anti-FGL1 therapy. More prospective clinical trials targeting FGL1 alone or in combination with other ICI agents are needed to facilitate translocation from bench to bedside.

Conclusions
As an acute inflammatory factor secreted by the liver, FGL1 is upregulated in various solid tumors and associated with the EMT, proliferation, apoptosis and drug resistance of tumors as well as with poor prognosis. Most importantly, an immunosuppressive pathway distinct from that of PD-1/PD-L1 forms between FGL1 and the inhibitory receptor LAG3. Targeting FGL1 acts on both tumor cells themselves and immune cells, which exerts a two-way synergistic effect on advanced cancers through a targeted therapeutic strategy. Therefore, FGL1 demonstrates significant advantages as a biomarker for predicting the efficacy of anti-PD-1/ PD-L1 therapy in patients with advanced tumors, and