Abstract
Multikinase inhibitors (MKIs) have been the only first-line treatment for advanced hepatocellular carcinoma (HCC) for more than a decade, until the approval of immune checkpoint inhibitors (ICIs). Moreover, the combination regimen of atezolizumab (anti-programmed cell death protein ligand 1 antibody) plus bevacizumab (anti-vascular endothelial growth factor monoclonal antibody) has recently been demonstrated to have superior efficacy when compared with sorafenib monotherapy. The remarkable efficacy has made this combination therapy the new standard treatment for advanced HCC. In addition to MKIs, many other molecularly targeted therapies are under investigation, some of which have shown promising results. Therefore, in the era of immuno-oncology, there is a significant rationale for testing the combinations of molecularly targeted therapies and ICIs. Indeed, numerous preclinical and clinical studies have shown the synergic antitumor efficacy of such combinations. In this review, we aim to summarize the current knowledge on the combination of molecularly targeted therapies and immune checkpoint therapies for HCC from both preclinical and clinical perspectives.
Keywords
Introduction
Liver cancer is one of the most common cancers and the fourth leading cause of cancer death globally, and its burden continues to rise steadily.1,2 Hepatocellular carcinoma (HCC) is the most common primary liver cancer, accounting for approximately 90% of all liver cancer cases. 3 Molecularly targeted therapies, rather than chemotherapy, play crucial roles in the systematic therapy of unresectable HCC. The theory behind molecularly targeted therapies is that some molecular alterations function importantly during cancer development, growth, and metastasis, and may be potential targets for cancer treatment. For HCC, the most important molecular target is the vascular endothelial growth factor (VEGF) and its receptor (VEGFR). Despite the extensive attempts at new drug development for HCC, sorafenib [multikinase inhibitor (MKI) of VEGFR] remained the sole, effective systemic therapy for nearly a decade (2007–2016) since its approval. 4 In 2017, the situation changed due to the approval of regorafenib (MKI of VEGFR) as a second-line treatment for HCC patients who progressed on sorafenib treatment. 5 Since then, numerous clinical trials on MKIs of VEGFR have yielded positive results, leading to the approval of lenvatinib 6 as first-line treatment and cabozantinib 7 and ramucirumab 8 as subsequent-line treatments. In addition, there are many promising molecularly targeted agents currently being studied in clinical trials, such as transforming growth factor beta (TGF-β) inhibitors,9,10 MET inhibitors, 11 fibroblast growth factor receptor 4 (FGFR4) inhibitors, 12 and other similar molecular inhibitors.
Over the past decade, immune checkpoint blockade therapies targeting programmed cell death protein 1 (PD-1), programmed cell death protein ligand 1 (PD-L1), and cytotoxic T lymphocyte antigen 4 (CTLA-4) have made tremendous breakthroughs that have revolutionized the treatment of various cancers. Anti-PD-1 monotherapy has been approved for the treatment of advanced HCC in both first- and second-line settings based on clinical study data.13–18 However, in HCC, the overall response rate (ORR) to anti-PD-1 monotherapy was only 15–20% and, more importantly, the overall survival (OS) did not significantly improve.13–18 Given this data, researchers are currently making great efforts in improving the therapeutic efficacy of immune checkpoint inhibitors (ICIs) for HCC, including combining ICIs with other molecularly targeted therapies. Recently, on the basis of promising results of the IMbrave150 study, the combination of atezolizumab (anti-PD-L1 antibody) and bevacizumab (anti-VEGF monoclonal antibody) has been approved as a novel first-line treatment for advanced HCC. 19 The final analysis of the clinical trial demonstrated that this combination resulted in significantly better OS and progression-free survival (PFS) than sorafenib monotherapy in patients with advanced HCC. 19 This exciting result also suggests the promising future of combining molecularly targeted therapies and immune checkpoint blockade therapies in the treatment of advanced HCC.
Currently, there are an increasing number of studies evaluating various combination strategies of molecularly targeted therapies and immune checkpoint therapies for HCC. Although most combination strategies lack robust evidence at present, some have shown promising efficacy in animal models or clinical studies. In this review, we focus on the current knowledge of combining molecularly targeted therapies and immune checkpoint therapies in HCC and provide an outlook on the future of such combination therapies. For molecularly targeted therapies, we will mainly focus on VEGF/VEGFR inhibitors, TGF-β inhibitors, MET inhibitors, FGFR4 inhibitors, and epigenetic drugs, which have showed promising results in clinical studies (Figure 1; Table 1).
Clinical trials testing various combinations of molecularly targeted therapies and ICIs in HCC.
DOR, duration of response; FGFR4, fibroblast growth factor receptor 4; GC, gastric cancer; GI cancers, gastrointestinal cancers; HCC, hepatocellular carcinoma; ICI, immune checkpoint inhibitor; ORR, objective response rate; OS, overall survival; PD-1, programmed cell death protein 1; PD-L1, programmed cell death protein ligand 1; PFS, progression-free survival; TGF-β, transforming growth factor beta; VEGF, vascular endothelial growth factor; VEGFR, vascular endothelial growth factor receptor.
Molecularly targeted therapies and immune checkpoint inhibitors for hepatocellular carcinoma and their targets.
Selection of clinical trials
To include clinical trials on the combination of molecularly targeted therapies and immune checkpoint therapies in HCC, we first searched the https://clinicaltrials.gov website by using the terms “hepatocellular carcinoma” and “drug.” A total of 1367 interventional trials were identified and further screened until February 2021. Trials were included only if they met the following criteria: (1) included patients with advanced or unresectable HCC; (2) included combination therapies of molecularly targeted therapies and immune checkpoint therapies; and (3) specific classes of molecularly targeted therapies, including VEGF/VEGFR inhibitors, TGF-β inhibitors, MET inhibitors, FGFR4 inhibitors, and epigenetic drugs. Studies were excluded if they met any of the following criteria: (1) combination therapies, such as neoadjuvant or adjuvant therapy; (2) combinations of molecularly targeted therapies and immune checkpoint therapies with other therapies; (3) have already been terminated or withdrawn; and (4) contained insufficient required information. The included clinical trials are summarized in Table 1.
VEGF/VEGFR inhibitors
HCC is a hypervascularized solid tumor in which angiogenesis plays a critical role in cancer development, proliferation, and metastasis. 20 VEGF/VEGFR inhibitors that modulate angiogenesis, including sorafenib, regorafenib, lenvatinib, cabozantinib, and ramucirumab, currently dominate the approved systemic therapies for advanced HCC. As for bevacizumab, a monoclonal anti-VEGF antibody with a long history, there is also some evidence, from clinical trials, indicating its potential efficacy in HCC.21,22 With the encouraging results of ICIs in HCC and the mounting evidence supporting the synergic effects of anti-angiogenesis and immunotherapy, the potential of combining anti-VEGF/VEGFR drugs and ICIs in the treatment of advanced HCC has been investigated by plenty of animal and clinical studies.
In addition to promoting angiogenesis and increasing vessel permeability, the VEGF signaling pathway has also been found to play a crucial role in cancer immunity.20,23–25 In general, several mechanisms have been postulated for the immunosuppressive activity of the activation of the VEGF signaling pathway. Pathological angiogenesis generates abnormal vessels that may restrict the infiltration of immune cells into tumors. 23 The upregulated level of VEGF could also result in the immunosuppressive tumor microenvironment (TME) by inhibiting the trafficking, proliferation, and effector function of cytotoxic T lymphocytes (CTLs) as well as the maturation and antigen presentation of dendritic cells (DCs). This, in turn, promotes immunosuppressive cell recruitment and proliferation, leading to a hypoxic and low-pH TME. 23 Thus, inhibition of the VEGF–VEGFR interaction could not only normalize vasculature but also enhance antitumor immunity. 20 The addition of ICIs could further prevent the immune evasion of tumors and has a potential synergic effect with VEGF/VEGFR inhibitors. 20 Recently, Shigeta et al. 26 tested dual anti-PD-1/VEGFR-2 therapy in orthotopic murine models of HCC. They found that an anti-PD-1 antibody and an anti-VEGFR-2 antibody synergized in vessel normalization and immune microenvironment reprogramming, overcoming resistance to any of the individual treatments and thereby enhancing tumor elimination. 26
Recently released data from the IMbrave150 study demonstrated that atezolizumab plus bevacizumab resulted in higher ORR and prolonged OS and PFS when compared with sorafenib monotherapy. 19 There are a number of clinical trials focusing on various combination regimens of VEGFR inhibitors and ICIs currently underway (Table 1). Here we will review these clinical studies and mainly focus on those with preliminary results.
Bevacizumab
After bevacizumab treatment, CD8+ T-cell proliferation is significantly increased while the presence of PD-L1+ tumor cells, PD-1+ cells, regulatory T (Treg) cells, and tumor-associated macrophages (TAMs) is significantly decreased in glioblastoma tissues. 27 Bevacizumab could also increase CD8+ T-cell populations in tumor, reduce Treg cell populations in the blood, and sustain the circulation of the effector T cells.28–30 In vitro and in vivo administration of bevacizumab was shown to inhibit VEGF-mediated increases in Treg cells. 31 Bevacizumab treatment resulted in a significant reduction of myeloid-derived suppressor cells (MDSCs), thereby restoring effective antitumor immunity.32–34 Previous studies also found that bevacizumab could abrogate the inhibition of monocytes differentiation into DCs.35–37 Bevacizumab was shown to enhance the maturation and number of DCs in the peripheral blood of patients with lung, breast, and colorectal cancers. 38 Moreover, atezolizumab in combination with bevacizumab enhanced antigen-specific T-cell migration in metastatic renal cell carcinoma (RCC). 39 Similarly, in patients with melanoma, a significant increases in circulating CD4+ and CD8+ T cells occurred with ipilimumab plus bevacizumab versus ipilimumab alone. 40 To summarize, bevacizumab has significant effects on a variety of immune cells and contributes to the restoration of an immunosupportive TME.
These findings provide important rationales for the combination of bevacizumab and ICIs in cancer treatment. As previously mentioned, bevacizumab plus atezolizumab has recently become a new first-line treatment for advanced HCC (NCT03434379). 19 Other clinical trials are also testing its efficacy (NCT02715531, NCT04180072, NCT04563338, NCT04487067, and NCT04732286). 41 Other combinations of bevacizumab and ICIs, for instance, bevacizumab plus nivolumab (NCT04393220), bevacizumab plus durvalumab (NCT02519348), bevacizumab plus toripalimab (NCT04605796, NCT04723004), and bevacizumab plus HX-008 (NCT04741165) are also under investigation in clinical trials.
Sorafenib
Sorafenib can decrease suppressive immune cell populations, including Treg cells and MDSCs, and enhance functions of tumor-specific effector T cells.42–44 Sorafenib can also restore classical macrophage polarization in the TME of HCC, which correlates with a reduction in tumor burden.45,46 Sorafenib may also inhibit macrophage-induced hepatocarcinoma growth by acting on various targets.47,48 Similar to bevacizumab, sorafenib could abrogate the differentiation of human monocytes into DCs. 35 In mouse models of various cancer types, sorafenib induced increased infiltration of CD4+ and CD8+ T cells into tumors.44,49–51 By contrast, in HCC mouse models or studies using peripheral blood mononuclear cells (PBMCs), the number and function of CD4+ and CD8+ T cells may be depressed directly by sorafenib or indirectly by the elevation of MDSCs and Treg cells following sorafenib treatment.52,53 For natural killer (NK) cells, low-dose sorafenib may enhance NK cell effector functions, such as interferon gamma (IFN-γ)/tumor necrosis factor alpha (TNF-α) production, degranulation, and lytic functions,46,54 whereas high-dose sorafenib generally inhibits NK cell proliferation and function.55–58 On experimental models of HCC, sorafenib in combination with nivolumab showed a stronger tumor growth inhibition as opposed to sorafenib or nivolumab monotherapy. 59
Therefore, it is reasonable to test the combination of sorafenib and ICIs in clinical studies. Clinical trials of combined efficacy of sorafenib plus pembrolizumab (NCT03211416), nivolumab (NCT03439891), PDR001 (NCT02988440), toripalimab (NCT04069949), and atezolizumab (NCT04770896) are now ongoing.
Lenvatinib
In murine models of diverse cancers, lenvatinib monotherapy was found to increase the population of CD8+ T cells in TME and this effect was further enhanced when combined with an anti-PD-1 antibody.60–62 Lenvatinib enhanced the antitumor activity of PD-1 blockade by upregulating memory T-cell population and enhanced T helper type 1 (Th1) immune response.63,64 It was also found that the antitumor activity of lenvatinib was associated with enhanced tumor immune infiltration and activation of NK cells in HCC models. 65 More importantly, lenvatinib may decrease TAMs in RCC, colon cancer, and thyroid cancer tumor models, and exhibited synergistic antitumor effects in combination with anti-PD1 therapy.64,66–68
In a phase Ib clinical trial (NCT03006926), lenvatinib plus pembrolizumab was assessed as a first-line treatment in advanced HCC. 69 The combination therapy showed promising antitumor activity with an ORR of 36.0% [95% confidence interval (CI): 26.6–46.2%] and a tolerable safety profile. 69 Another phase Ib clinical trial of lenvatinib plus nivolumab, in patients with unresectable HCC, reported an ORR as high as 76.7% (95% CI: 57.7–90.1%) and manageable adverse events (NCT03418922). 70 A phase II trial of lenvatinib plus nivolumab in HCC is currently underway (NCT03841201). 71 A phase III trial aiming to compare the efficacy of the combination of lenvatinib and pembrolizumab with lenvatinib monotherapy, as first-line treatment for HCC, is also now underway (NCT03713593). 72 Other clinical trials on the combination of lenvatinib and ICIs are listed in Table 1.
Regorafenib
Regorafenib was previously reported to decrease macrophage accumulation in several murine models of colorectal cancer.73,74 When administered in mice, regorafenib suppressed melanoma progression in a CD8+ T-cell-dependent manner. 75 Recently, regorafenib was found to promote tumor immunity by targeting the RET–Src axis to further inhibit JAK1/2–STAT1 and MAPK signaling as well as attenuate IFN-γ-induced PD-L1 and indoleamine 2,3-dioxygenase 1 (IDO1) expression. 76
For HCC, early stage clinical trials of regorafenib plus avelumab (NCT03475953), pembrolizumab (NCT03347292, NCT04696055), nivolumab (NCT04170556, NCT04310709), sintilimab (NCT04718909), and tislelizumab (NCT04183088) are ongoing at present.
Cabozantinib
Cabozantinib rendered tumor cells more sensitive to immune-mediated killing and contributed to a more permissive immune environment in experimental study. 77 Cabozantinib treatment also resulted in a significant reduction of MDSCs in mouse models. 34 In clinical trials, cancer patients treated with cabozantinib were found to have increased CD8+ T cells and decreased Treg cells and MDSCs.78,79 Furthermore, cabozantinib was reported to trigger a neutrophil-mediated anticancer innate immune response that resulted in tumor clearance. 80 Cabozantinib could downregulate M1 macrophages to prevent bone metastasis of prostate carcinoma 81 and reverse c-MET-induced immunosuppressive PD-L1 upregulation. 82
In one cohort from the CheckMate 040 study, the triple combination of nivolumab, ipilimumab, and cabozantinib led to an ORR of 26% with tolerable toxicity. 83 There are also ongoing clinical trials testing cabozantinib plus pembrolizumab (NCT04442581), nivolumab (NCT04514484), atezolizumab (NCT03170960, NCT03755791), and durvalumab (NCT03539822) for HCC patients.
Ramucirumab
Ramucirumab-containing therapies for gastric cancer (GC) patients displayed significantly increased CD8+ T-cell infiltration, decreased Treg cell infiltration, and reduced PD-1 expression by CD8+ T cells. 84
A phase Ib study (NCT02572687) showed that ramucirumab plus durvalumab yielded increased antitumor activity without unexpected toxicity in HCC patients, especially those with high PD-L1 expression. 85
Others
SHR-1210 (anti-PD-1 antibody) and apatinib combination therapy demonstrated an ORR of 50.0% (95% CI: 24.7–75.4%) in advanced HCC patients (NCT02942329). 86 A phase Ib trial (NCT03289533) evaluating avelumab plus axitinib in treatment-naive patients with HCC showed an ORR of 31.8% (95% CI: 13.9–54.9%). 87
As shown in Table 1, many other combinations of anti-VEGF/VEGFR antibodies and ICIs are now under investigation in clinical trials.
TGF-β inhibitors
The TGF-β pathway has miscellaneous functions in cancers, including regulating cell growth, differentiation, apoptosis, motility and invasion, extracellular matrix production, angiogenesis, and immune response.88,89 In HCC, TGF-β signaling has been demonstrated to be significant in a subset of patients based on gene expression profiling. 90 It is found that TGF-β1, an isoform of TGF-β, can stimulate α3-integrin expression in noninvasive HCC cells at the transcriptional level, transforming them into a motile and invasive phenotype. 91 In an experimental model of hepatocyte transmigration using hepatic sinusoidal endothelial cells and malignant hepatocytes to mimic vascular invasion, TGF-β was shown to be crucially involved in blood vessel invasion of HCC cells. This suggests its key role in the dissemination and metastasis of HCC. 92 In addition, TGF-β was found to contribute to HCC invasion and metastasis by inducing FGFR4 expression through the extracellular-signal-regulated kinase (ERK) pathway. 93 TGF-β signaling also functions as a master regulator for immune cell proliferation, differentiation, development, and survival.94,95 In the liver, TGF-β signaling plays a crucial role in regulating immune cells to maintain a balance between immune tolerance and activation96–98 (Figure 2). TGF-β activity has been identified as essential for the pathogenesis of HCC, including the activation of cancer-associated fibroblasts (CAFs).96–98
Interactions between TGF-β and immune cells in the tumor microenvironment of hepatocellular carcinoma.
The potent immunosuppressive function of Treg cells is a major hurdle in generating an effective antitumor response in HCC. 99 Tumor cells and all stromal cells, including MDSCs, TAMs, DCs, CAFs, hepatic stellate cells (HSCs), and liver sinusoidal endothelial cells (LSECs) modulate Treg cell activity in HCC via the TGF-β pathway. MDSCs play a critical role in tumor-induced liver immunosuppression by inhibiting NK cell functions through membrane-bound TGF-β and promoting Treg cell expansion.100,101 TGF-β, in the TME, promotes the expression of T-cell immunoglobulin mucin receptor 3 (TIM3) on macrophages and facilitates the alternative activation of macrophages via TIM3, thereby promoting tumor growth via the nuclear factor-κB (NF-κB)-interleukin 6 (IL-6) axis. 102 TGF-β promotes the differentiation of M2 macrophages, which repress CD8+ T-cell, NK cell, and DC functions and increase Treg cell functions. 103 A subset of DCs selectively promote the proliferation of Treg cells in a TGF-β dependent manner. 104 TGF-β expressed and activated by cancer cells, or other cells in the TME, promote cancer progression through its effects on CAFs. 105 CAFs are important components of the HCC TME, generating cyclooxygenase-2 (COX2), IL-8, and other cytokines that, together, stimulate TAMs to release TNF-α and platelet-derived growth factor (PDGF).96–98 HSCs are activated by TGF-β and induce immune tolerance via enhancing the expansion of MDSCs, attenuation of effector T-cell functions, and augmentation of Treg cells.106,107 LSECs are the most efficient liver cell type in TGF-β-dependent Treg cell induction and produce the chemokine C-X-C motif chemokine ligand 16 (CXCL16) to control the accumulation of C-X-C motif chemokine receptor 6 (CXCR6) + NKT cells.108,109
TGF-β has been demonstrated to induce forkhead box P3 (Foxp3), a transcriptional factor expressed predominantly on Treg cells, through a Smad2/3-dependent mechanism. 110 Foxp3+ Treg cells are highly enriched in tumors of HCC patients and correspond to poorer clinical outcomes. 111 Treg cells also secrete TGF-β, as well as adenosine and IL-10, to suppress effector T cells, such as CD8+ CTLs, 112 which play a critical role in the antitumor immune response by the production of various effector molecules including IFN-γ, IL-2, and TNF-α. 113 CD28 is a co-stimulatory molecule that interacts with the CD80/CD86 complex on antigen-presenting cells (APCs). The immune checkpoint molecule expressed on Treg cells, CTLA4, can compete with CD28 to downregulate the CD80/CD86 complex through CTLA-4-mediated trans-endocytosis, thus inhibiting the antitumor immunity.114,115 TGF-β has also been implicated in inducing Treg cell polarization. 116 Accordingly, inhibition of the TGF-β signaling decreased Treg cell infiltration in tumor tissues and induced HCC tumor regression. 116 Therefore, targeting Treg cells, such as by blocking the TGF-β signaling or the membrane-bound receptor CTLA-4, may be a particularly effective immune-based approach for improving the immune response against HCC.
IFN-γ is a cytokine that supports the proliferation and differentiation of myeloid cells and CD8+ T cells. In numerous immune cell types, TGF-β suppresses IFN-γ expression via the Smad pathway, thereby inhibiting the antitumor activity of CD8+ T cells.117,118 Th1 and Th2 lineages produce cytokines that support the growth and functions of CD8+ cytotoxic T cells, B cells, and macrophages. TGF-β restricts the specification of Th1 and Th2 cells by repressing T-bet expression and inducing MSC and Sox4 expression, respectively. 118 Additionally, TGF-β was demonstrated to promote a shift of Th1 toward Th2 cell differentiation, resulting in a less efficient antitumor immune response. 119 Th17 cells are a proinflammatory Th cell subtype that expresses IL-17 and contribute to non-alcoholic steatohepatitis (NASH) and HCC.120,121 Through the activation of Smad2/3, TGF-β cooperates with IL-6/IL-21 to positively regulate the generation of Th17 cells from naive CD4+ T cells.120,121 Intratumoral TGF-β suppresses NKT cells, which are responsible for recruiting effector immune cells to the tumor by producing large amounts of IFN-γ. 122 NKG2D is one of the NK cell activating receptors. Active NKG2D can induce the activation, cytokine production, degranulation, and cytotoxic potential in NK cells. 123 Downregulation of NKG2D or upregulation of the inhibitory receptor NKG2A is associated with increased generation of immunosuppressive cytokines, such as TGF-β and IL-10, contributing to NK cell dysfunction in HCC.124–129 Although TGF-β promotes chemotaxis of eosinophils and mast cells, it inhibits effector functions of NK cells and antigen presentation by DCs.95,130 During B-cell maturation, TGF-β stimulates class switch recombination, which converts IgM-expressing B cells into IgA-expressing (IgA+) cells with regulatory activity. 131 IgA+ cells are involved in nonalcoholic fatty liver disease (NAFLD)-associated HCC by co-expressing PD-L1 and IL-10 and by restraining cytotoxic CD8+ T lymphocytes. 132
TGF-β enhances antigen-induced PD-1 expression on tumor-infiltrating lymphocytes (TILs) through Smad3-dependent transcriptional activation. 133 And TGF-β induced epithelial–mesenchymal transition (EMT) increases PD-L1 expression in tumor cells. 134 The engagement of the co-inhibitory receptor PD-1 or its ligand PD-L1 blocks T-cell antigen receptor (TCR) signaling and inhibits T-cell proliferation and secretion of cytotoxic mediators, including granzyme A (GZMA), granzyme B (GZMB), TNF-α, and IFN-γ, which all collectively lead to T-cell exhaustion.135,136 Thus, increased TGF-β signaling may allow tumors to evade host immune responses by upregulating PD-1 expression, and inhibition of TGF-β signaling may directly enhance antitumor immunity. 133 Tumors, particularly immune-excluded tumors that are intrinsically resistant to anti-PD-1/PD-L1 therapy, display elevated TGF-β signaling.137–139 Indeed, anti-PD-1 treatment can induce a competing TGF-β-driven immunosuppressive program in a mouse model. 140 Together, these findings provide an important rationale for blocking the TGF-β signaling in order to circumvent the aggressiveness and resistance to therapies of HCC, especially in the era of immune-oncology.
Pharmacological agents targeting the TGF-β pathway, including receptor kinase inhibitors and neutralizing antibodies that inhibit the interactions of TGF-β ligands with their receptors, have already demonstrated promising antitumor efficacy in early clinical trials with an acceptable safety profile in a variety of cancers, including HCC. 141 Dual blockade of TGF-β signaling and PD-1/PD-L1 checkpoint successfully reduced TGF-β signaling in stromal cells, promoted T-cell infiltration into tumor centers, elicited vigorous antitumor immunity, and contributed to tumor regression in mouse models.137,140,142–144 In addition, M7824, a bifunctional fusion protein consisting of an α-PD-L1 antibody moiety, based on avelumab linked to the extracellular domain of TGF-β receptor II, was developed.145,146 This novel agent blocked signaling from the immune checkpoint PD-L1 surface protein and reduced TGF-β signaling within the TME by binding to all three TGF-β isoforms.145,146 Its dual anti-immunosuppressive function led to the activation of the innate and adaptive immune systems, upregulation of PD-L1 levels in tumor cells, and induction of tumor regression in mouse models.145,146 Another bifunctional fusion protein that blocks both TGF-β and CTLA-4 has also been developed, which was more effective in reducing tumor-infiltrating Tregs and inhibiting tumor progression compared with CTLA-4 antibody. 147 Early clinical trials of M7824 in patients with metastatic or locally advanced solid tumors, including HCC, have shown encouraging efficacy and a manageable safety profile (NCT02517398, NCT02699515) (Table 1).148,149 A clinical trial of galunisertib (a systemic TGF-β receptor 1 inhibitor) in combination with nivolumab for HCC is ongoing (NCT02423343) (Table 1). Other clinical trials testing BCA101 (EGFR/TGF-β fusion monoclonal antibody) plus pembrolizumab (NCT04429542) or NIS793 (TGF-β inhibitor) plus PDR001 (spartalizumab, PD-1 inhibitor) (NCT02947165) are also ongoing (Table 1).
MET inhibitors
c-MET (MET) is a tyrosine kinase receptor with a single known ligand, hepatocyte growth factor (HGF). In HCC, overexpression of c-MET and HGF was noted in 20% and 33% of human samples, respectively. 150 The MET/HGF axis is involved in HCC progression by promoting cellular proliferation, survival, and invasion.151,152 The MET/HGF signaling pathway is also associated with tyrosine kinase inhibitor resistance,153,154 as patients with plasma HGF concentrations above 3279.1 pg/ml derived no obvious benefit from sorafenib compared with placebo in the pivotal phase III SHARP trial. 155 Therefore, the MET/HGF pathway may be used as a potential therapeutic target for the treatment of HCC. Moreover, the MET/HGF pathway has complex immunomodulatory effects (Figure 3). It can act as an immunosuppressive stimulus by negatively affecting DCs and T lymphocytes, or as an immune-positive stimulus by promoting the recruitment of DCs, B cells, and T lymphocytes.156–161
Immunomodulatory effects of HGF/MET signaling.
To date, the potential interplay between the HGF/MET pathway and DCs has been extensively investigated. It is well known that DCs are involved in the presentation of tumor-associated antigens (TAAs) and the activation of CD4+ Treg cells that control CD8+ cytotoxic T-cell activity. Several studies have found that HGF is able to enhance this function and plays a positive role in the immune response.156,162–164 However, it has also been shown that HGF can act as a potent negative regulator of DC function and can induce an increase of Treg cells, a decrease of IL-17-producing T cells, 156 and an increase in IL-10 and TGF-β, which are known for their immunosuppressive effects. 157 The inhibition of DC functions, together with the decrease of CD8+ T cells and the increase of Treg cells, leads to a decline of anticancer immunity. 156 Moreover, HGF favors the differentiation of monocytes into IL-10+ regulatory DCs. This inhibitory effect has also been demonstrated in monocytes, particularly with an induction toward differentiation into IL-10+ regulatory DCs. 165
Besides APCs, the impact of HGF/MET signaling on granulocytes has also been highlighted in previous studies. MET is required for neutrophil chemoattraction and cytotoxicity, while its deletion in mouse neutrophils enhances tumor growth and metastasis. 166 This phenotype is associated with reduced neutrophil infiltration to both primary and metastatic tumors in clinical samples. 166 Mechanistically, tumor-derived TNF-α or other inflammatory stimuli can induce MET in neutrophils, which results in neutrophil transmigration across an activated endothelium and inducible nitric oxide synthase production, thereby promoting cancer cell killing and abating tumor progression. 166 This mechanism may partially explain the resistance to MET inhibitors in some patients, which is an important issue for clinicians in developing treatment strategies. 166 This example reveals the complex role of MET in tumors regarding its immunomodulatory effects, which is of great importance when testing its combination with immunotherapy. In HCC, tumor neutrophils actively enhanced the metastasis of malignant cells in vitro and in vivo via HGF/c-MET interaction. 167
c-MET inhibition promoted perforin expression in CD8+ T cells and contributes to the activation of the immune system. 168 c-MET inhibition impaired the reactive mobilization and recruitment of neutrophils into tumors and draining lymph nodes, potentiating T-cell antitumor immunity. 169 Furthermore, in immunocompetent mice, the addition of MET inhibitors to immunotherapy increased the number of active T cells and changed their phenotype by reducing the proportion of exhausted T cells. 169 These results were independent of MET expression in the tumor models used, suggesting that c-MET inhibitor co-treatment may improve the response to cancer immunotherapy in settings beyond c-MET-dependent tumors. 169
MET was identified as a broadly expressed TAA that can be recognized by CD8+ cytotoxic T cells, which then triggered activation of the immune system against cancer cells overexpressing MET. 170 Similarly, Kumai et al. 171 showed that MET expression itself behaves as a TAA, which was able to activate CD4+ T cells and to induce tumor cell killing in NK/T-cell lymphoma (NKTCL) cell lines. Particularly, in this model, MET elicited a specific antitumor immune response with three newly identified MET-induced T-cell epitopes. The activation of T cells was stronger in the presence of MET inhibitors since it caused a reduction of the synthesis of TGF-β from tumor cells. Additionally, the presentation of MET-derived peptides by major histocompatibility complex class II (MHCII) to CD4+ T cells was influenced by chaperon processing and autophagy, thus proposing an innovative potential role of autophagy inducers as immune activators. Finally, since HGF/MET stimulation increases the proliferation of NKTCL cells in vitro, MET inhibition again displayed a dual role: direct tumor killing for MET-dependent cell survival and antitumor immune activation. 171
PDL-1/2 expression was found to correlate significantly with c-MET expression in GC patients. 172 c-MET was shown to upregulate PD-L1 expression, contributing to immune escape of tumor cells in various cancers including GC, non small cell lung cancer (NSCLC), RCC, and HCC.82,173–179 However, there are also some preclinical studies showing the opposite effect of c-MET in terms of the regulation of PD-L1 expression. MET-mediated phosphorylation and activated glycogen synthase kinase 3β (GSK3B) can lead to decreased expression of PD-L1. 180 Exposure of liver cancer cell lines to MET inhibitors increased their expression of PD-L1 and inactivated co-cultured T cells. 180 In combination with a MET inhibitor, anti-PD-1 and anti-PD-L1 produced additive effects to slow the growth of HCCs in mice. 180 Similarly, exposure of NCSLC cell lines to the c-MET inhibitor, tivantinib, increased their PD-L1 expression, which in turn caused cells to become more resistant to T-cell killing. This provides a rationale for the use of the combination therapy of c-MET inhibitors and ICIs in NSCLC. 181 Moreover, a novel bispecific antibody targeting both MET and PD-1 was developed and studied in multiple cancer cell-type models.182,183 The antibody exhibited strong anti-proliferative and anti-metastatic effects in vitro and in vivo and reduced the production of inflammatory chemokines such as IL-6 and TNF-α, thus suggesting an important therapeutic potential, although it is still in the preclinical model stage.182,183 Besides, clinical trial of APL-101 (c-MET inhibitor) plus nivolumab or APL-501 (PD-1 inhibitor) (NCT03655613) and capmatanib (MET inhibitor) plus PDR001 (NCT02795429) are recruiting patients (Table 1).
FGFR4 inhibitors
FGFR4 is the predominant FGFR expressed in the liver and has been identified as a promising target in HCC. 184 Fibroblast growth factor 19 (FGF19) can bind to and activate FGFR4 to induce hepatocyte proliferation. 185 The activation of the FGF19/FGFR4 pathway suppresses E-cadherin expression and hence promotes EMT, contributing to the aggressiveness of HCC. 186 FGF19 was significantly overexpressed in HCC specimens and was independently associated with tumor progression and poor prognosis in HCC patients.187,188 Additionally, FGF19 was found to be strongly associated with the immune checkpoint signature (CD274, PDCD1, BTLA, CTLA4, HAVCR2, IDO1, and LAG3) in lymphoepithelioma-like HCC (LEL-HCC). 189 As previously described, FGFR4 expression was upregulated by TGF-β through the ERK pathway, in HCC cell lines, which contributed to the metastatic dissemination of HCC in vivo. 93 More importantly, FGF19/FGFR4 signaling plays an important role in the resistance of HCC to sorafenib,190,191 making it of particular interest given the current widespread use of MKIs in HCC patients. A clinical trial testing FGF401 (FGFR4 inhibitor) plus PDR001 (NCT02325739), and another clinical trial testing fisogatinib (FGFR4 inhibitor) plus CS1001 (PD-1 inhibitor) (NCT04194801), in HCC, are both currently ongoing (Table 1).
Epigenetic drugs
Epigenetic events play a crucial role in tumor development, progression, and metastasis, providing another potential target for cancer treatment.192–194 Owing to their effect on immune response and cancer immunity, 195 there are strong rationales for the potential combination of epigenetic drugs and ICIs in cancers. Epigenetic modifiers function importantly in priming and enhancing the therapeutic effect of the host immune system against cancer.196,197 PBMCs and T-cell DNA methylation in HCC suggested that a broad molecular signature was involved in tumor progression, which was highly enriched in immune function-related genes such as PD-1. 198 Another study also found that highly upregulated DNA methyltransferases (DNMTs) were positively correlated with PD-L1 overexpression in sorafenib-resistant HCC cells. 199 Knockdown of PD-L1 induced DNMT1-dependent DNA hypomethylation and restored the expression of methylation-silenced Cadherin 1, a metastasis suppressor in HCC. 199
Epigenetic drugs, such as DNMT inhibitors (DNMTi) and histone deacetylase inhibitors (HDACi), were found to upregulate the expression of inhibitory immune checkpoints in either immune or cancer cells, which enhance the response to immune checkpoint therapy.196,197,200–207 DNMTi enhanced the therapeutic efficacy of PD-L1 blockade and increased tumor-infiltrating CD8+ T cells and Th1-type chemokine expression in ovarian cancer in C57/BL6 mice. 106 In a mouse model of fibrosis-associated HCC, the combination of i-BET762 (BET inhibitor) and anti-PD-L1 therapy was found to be able to repress monocytic MDSCs, enhance TILs, and lead to tumor eradication and prolonged survival. 208 The therapeutic effect of a HDACi belinostat in combination with immune checkpoint blockades (ICBs) was highlighted in a murine model of HCC, which was associated with enhanced IFN-γ production by antitumor T cells and a decrease in Treg cells. 209 Hong et al. 210 demonstrated that epigenetic therapy targeting EZH2 and DNMT1 could be a potential strategy to augment immunotherapy for HCC by stimulating T-cell trafficking into the TME. Guadecitabine, a second-generation DNMTi, showed synergic antitumor effects with immune checkpoint therapy and provided a rationale for such combination treatment. 211 As a result, a clinical trial of durvalumab plus guadecitabine (NCT03257761) for HCC patients is currently ongoing (Table 1).
Others
A phase I study (NCT02476123) of the anti-CCR4 antibody mogamulizumab in combination with nivolumab in patients with advanced or metastatic solid tumors, including HCC, demonstrated acceptable antitumor activity and safety profile. 212
There are now also a lot of combination therapies of other molecular targeted therapies and immune checkpoint therapies for HCC under investigation, such as pembrolizumab plus bavituximab (NCT03519997), nivolumab plus SF1126 [phosphatidylinositol 3-kinase (PI3K) inhibitor] (NCT03059147), nivolumab plus copanlisib (PI3K inhibitor) (NCT03735628), nivolumab plus abemaciclib [cyclin-dependent kinase 4 (CDK4) inhibitor] (NCT03781960), nivolumab plus cabiralizumab [colony-stimulating factor 1 receptor (CSF1R) antagonist] (NCT04050462), nivolumab plus BMS-986253 (IL-8 inhibitor) (NCT04050462, NCT04123379), pembrolizumab/nivolumab/atezolizumab/avelumab plus ALT-803 [IL-15 receptor (IL-15R) agonist] (NCT03228667), nivolumab plus BMS-813160 (CCR2 antagonist) (NCT04123379), nivolumab plus CC-122 (NCT02859324), etc.
Conclusion
Plenty of novel agents are under development and investigation in the post-sorafenib era of advanced HCC treatment. Among them, molecular targeted drugs and ICIs are by far the most promising. Various combination strategies of molecular targeted therapies and immune checkpoint therapies are being tested and have generated some encouraging preliminary data. We are continuing to experiment with various therapeutic approaches to improve the clinical outcome of advanced HCC. One of the biggest challenges in order to optimize treatment outcomes is the development of predictive biomarkers for both monotherapies and combination therapies to accurately identify patients most likely to respond to particular treatments. In HCC, MKIs have been used for more than a decade and ICIs have been approved for several years, yet there are still no satisfactory biomarkers for these two therapies. For combination therapies, a better understanding of the mechanisms of synergistic therapeutic effects would aid in the design of more effective treatment regimens, such as dosing and sequencing strategies. Finally, the prevention and management of toxicities of combination therapies should be taken seriously in clinical practice.
Footnotes
Acknowledgements
We thank the patients and investigators who participated in clinical trials for providing data.
Author contributions
Conceptualization, ZLL, and JC; Methodology, ZLL, and JHL; Software, ZLL; Formal Analysis, ZLL and JHL; Investigation, ZLL; Writing – Original Draft Preparation, ZLL and JC; Writing – Review and Editing, JC and DS; Supervision, JC. All authors have read and agreed to the published version of the manuscript.
Conflict of interest statement
The authors declare that there is no conflict of interest.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by the Basic and Applied Basic Research Fund of Guangdong Province, China (grant number: 2020A1515110491), the Fundamental Research Funds for the Central Universities (grant number: K20200089), the National Natural Science Foundation of China (grant number: 81702981), the China Postdoctoral Science Foundation Funded Project (grant number: 2020T130584), Zhejiang Provincial Natural Science Foundation of China (grant number:LY20H160021) and the China Scholarship Council (grant number: 201806325017). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.