Immune Checkpoint Inhibitors for Unresectable Hepatocellular Carcinoma

Despite the advances in screening protocols and treatment options, hepatocellular carcinoma (HCC) is still considered to be the most lethal malignancy in patients with liver cirrhosis. Moreover, the survival outcomes after failure of first-line therapy for unresectable HCC is still poor with limited therapeutic options. One of these options is immune checkpoint inhibitors. The aim of this study is to comprehensively review the efficacy and safety of immune checkpoint inhibitors for patients with HCC.


Introduction
Hepatocellular carcinoma (HCC) is still the most common and most lethal malignancy in patients with liver cirrhosis, despite the advances in screening programs, chemoprophylaxis for high-risk patients and treatment options [1,2]. With the rapid increase in prevalence of metabolic disorders, nonalcoholic fatty liver disease became one of the leading risk factors of HCC after hepatitis B and C [3,4]. Overall, HCC is considered an inflammatory prototypic cancer. The high mortality rate from HCC is related to late diagnosis and the concomitant liver dysfunction. In that case, usually, curative resection or liver transplantation is not feasible [5].
The tumor microenvironment of the HCC is infiltrated with different types of immune cells, mainly T-cells (CD8+, CD4+, Treg), natural killer cells and myeloid cells (myeloid-derived suppressor cells and tumor-associated macrophages). Due to the chronic inflammation and cirrhosis present in most HCC patients, the tumor ecosystem gets complicated affecting the behavior of the tumor and response to treatment. These changes are due to complex interactions between immune cells and tumor cells in the tumor microenvironment conveyed through cytokines and signaling pathways leading to exhaustion of pro-inflammatory immune cells and the dominance of the regulatory leukocytes hindering the anti-tumor response. A study by Yu et al. [19] concluded that improved overall survival was associated with high immune infiltration. The study further identified different immune clusters based on their prognostic value showing that better outcomes were associated with clusters with high levels of T-cells (mainly CD8+) and low levels of macrophages. A subset of tumor-associated macrophages (M1) was shown to be associated with improved outcomes. Poor prognosis is associated with the accumulation of myeloid-derived suppressor cells, tumor-associated macrophages, CD4+/CD25+/FOXP3+ immune-suppressive T-cells(T-reg), exhausted Th1 CD4+, CD8+ T-cells, dysfunctional NK cells and the expansion of Th2 CD4+ T-cells. Immune checkpoint molecules including programmed cell death (PD-1), CD274, cytotoxic T lymphocyte antigen -4 (CTLA-4), lymphocyte activated gene -3 (LAG-3) and IFNG were identified in clusters that had high levels of CD8+ T-cells. However, these clusters were associated with poor prognosis which leads to the assumption that these molecules are implicated in the HCC immune-exhaustion [20]. Therefore, it is assumed that the administration of immune checkpoint inhibitors would be beneficial for these HCC patients. In the United States, accelerated approval has been granted by the Food and Drug Administration (FDA) to two anti-programmed cell death monoclonal antibodies (nivolumab and pembrolizumab) and a combination of nivolumab plus ipilimumab, a monoclonal antibody against CTLA-4, for patients who progressed on sorafenib based on the results of several phase III trials [21][22][23]. However, data from phase III trial did not show superior efficacy of nivolumab as first-line therapy over sorafenib [24]. Moreover, the results of KEYNOTE 240 which assessed pembrolizumab as second-line therapy compared to placebo did not meet its predetermined level of statistical significance [25]. Therefore, we aimed to review the current evidence in the literature regarding the use of immune checkpoint inhibitors for the treatment of HCC.

PD-1/PD-L1 Inhibition
Avoiding immune destruction is one of the hallmarks of cancer. The PD-1/PD-L1 pathway plays a pivotal role in this escape mechanism [26]. Studies have shown that PD-L1 is overexpressed in tumor cells in different types of cancers including HCC, which leads to an increase in binding between PD-L1 and PD-1 on T cells within the tumor microenvironment resulting in immune anergy and apoptosis [27,28]. As a result, with overexpression of PD-L1, the tumor continues to grow unchecked which leads to worse prognosis in patients with HCC [28,29]. Interfering with this binding can result in enhancing immune reaction toward the cancer cells. (Figure 1) Therefore, the introduction of monoclonal antibody in the landscape of treatment of HCC has gained accelerated approval for patients who previously progressed on sorafenib based on the results CheckMate 040 trail [21]. However, for first-line therapy, the CheckMate 459 trial compared nivolumab to sorafenib in patients with Child-Pugh A (non-severe liver cirrhosis). Although the objective response rate was higher in the nivolumab group than the sorafenib group, the overall survival and progression-free survival were not significantly different between both groups [24] (Table 1)    One of the well-known mechanisms of resistance to anti-PD-1 therapy is Fc assumption that these molecules are implicated in the HCC immune-exhaustion [20]. Therefore, it is assumed that the administration of immune checkpoint inhibitors would be beneficial for these HCC patients. In the United States, accelerated approval has been granted by the Food and Drug Administration (FDA) to two anti-programmed cell death monoclonal antibodies (nivolumab and pembrolizumab) and a combination of nivolumab plus ipilimumab, a monoclonal antibody against CTLA-4, for patients who progressed on sorafenib based on the results of several phase III trials [21][22][23]. However, data from phase III trial did not show superior efficacy of nivolumab as first-line therapy over sorafenib [24]. Moreover, the results of KEYNOTE 240 which assessed pembrolizumab as second-line therapy compared to placebo did not meet its predetermined level of statistical significance [25]. Therefore, we aimed to review the current evidence in the literature regarding the use of immune checkpoint inhibitors for the treatment of HCC.

PD-1/PD-L1 Inhibition
Avoiding immune destruction is one of the hallmarks of cancer. The PD-1/PD-L1 pathway plays a pivotal role in this escape mechanism [26]. Studies have shown that PD-L1 is overexpressed in tumor cells in different types of cancers including HCC, which leads to an increase in binding between PD-L1 and PD-1 on T cells within the tumor microenvironment resulting in immune anergy and apoptosis [27,28]. As a result, with overexpression of PD-L1, the tumor continues to grow unchecked which leads to worse prognosis in patients with HCC [28,29]. Interfering with this binding can result in enhancing immune reaction toward the cancer cells. (Figure 1) Therefore, the introduction of monoclonal antibody in the landscape of treatment of HCC has gained accelerated approval for patients who previously progressed on sorafenib based on the results CheckMate 040 trail [21]. However, for first-line therapy, the CheckMate 459 trial compared nivolumab to sorafenib in patients with Child-Pugh A (non-severe liver cirrhosis). Although the objective response rate was higher in the nivolumab group than the sorafenib group, the overall survival and progression-free survival were not significantly different between both groups [24]. (Table 1) One of the well-known mechanisms of resistance to anti-PD-1 therapy is Fc ɤ ΥR1 mediated macrophage antibody-dependent phagocytosis [30]. (Figure 1) Therefore, another monoclonal antibody has been developed to evade the Fc R1 mediated resistance that is, tislelizumab [31]. Clinical data from the RATIONAL 301 trial, which is comparing tislelizumab against sorafenib, supporting this mechanism are still pending [31].
R1 mediated macrophage antibody-dependent phagocytosis [30] (Figure 1). Therefore, another monoclonal antibody has been developed to evade the Fc 3 (LAG-3) and IFNG were identified in clusters that had high levels of se clusters were associated with poor prognosis which leads to the ules are implicated in the HCC immune-exhaustion [20]. Therefore, it is ion of immune checkpoint inhibitors would be beneficial for these HCC tes, accelerated approval has been granted by the Food and Drug o anti-programmed cell death monoclonal antibodies (nivolumab and ination of nivolumab plus ipilimumab, a monoclonal antibody against gressed on sorafenib based on the results of several phase III trials [21-ase III trial did not show superior efficacy of nivolumab as first-line oreover, the results of KEYNOTE 240 which assessed pembrolizumab pared to placebo did not meet its predetermined level of statistical e aimed to review the current evidence in the literature regarding the hibitors for the treatment of HCC. bitors as a First-Line Therapy uction is one of the hallmarks of cancer. The PD-1/PD-L1 pathway plays mechanism [26]. Studies have shown that PD-L1 is overexpressed in es of cancers including HCC, which leads to an increase in binding T cells within the tumor microenvironment resulting in immune anergy result, with overexpression of PD-L1, the tumor continues to grow rse prognosis in patients with HCC [28,29]. Interfering with this binding mune reaction toward the cancer cells. (Figure 1) Therefore, the ntibody in the landscape of treatment of HCC has gained accelerated eviously progressed on sorafenib based on the results CheckMate 040 ine therapy, the CheckMate 459 trial compared nivolumab to sorafenib A (non-severe liver cirrhosis). Although the objective response rate was up than the sorafenib group, the overall survival and progression-free ly different between both groups [24]. (Table 1) mechanisms of resistance to anti-PD-1 therapy is Fc ɤ ΥR1 mediated dent phagocytosis [30]. (Figure 1) Therefore, another monoclonal d to evade the Fc R1 mediated resistance that is, tislelizumab [31]. IONAL 301 trial, which is comparing tislelizumab against sorafenib, re still pending [31].
R1 mediated resistance that is, tislelizumab [31]. Clinical data from the RATIONAL 301 trial, which is comparing tislelizumab against sorafenib, supporting this mechanism are still pending [31].

Dual Immune Checkpoint Blockade
CTLA-4 is expressed on T regulatory cells regulating the early immune response after the primary stimulation by antigens mainly in lymphoid organs whereas PD-1 is expressed mainly on activated T cells in the tumor microenvironment regulating late immune response. Moreover, inhibition of the CTLA-4/B7 signal in lymph nodes increases activated CD8+ cells which will subsequently infiltrate the tumor and be part of the microenvironment [32,33]. Based on this, several studies have shown promising results with dual immunotherapy [34,35]. The success achieved in these trials especially in for patients with melanoma [36] has inspired the application of dual immune blockage for other types of cancers including HCC. Therefore, after the success achieved by the phase I/II trial investigating the efficacy and safety of dual immune therapy for patients progressed on sorafenib [37], a comparative randomized controlled trial, HIMALAYA study, was designed to compare Duravalumab versus the combination of Duravalumab plus Tremelimumab versus sorafenib. Its results are still bending (Table 1).

Combination with Biological Therapy
Vascular endothelial growth factor (VEGF) has been linked with the development and progression of HCC [38,39]. Moreover, it has a role in immune suppression as it has been found that it creates an immunosuppressive microenvironment through the recruitment of several inhibitory cells such as T regulatory cells, tumor-associated macrophages and myeloid-derived suppressor cells. Those cells release cytokines such as IL-10 and TGF-β that inhibit natural killer cell and T cell activation and impedes dendritic cell maturation as shown in Figure 2 [33,40,41].
The landmark IMbrave 150 trial comparing atezolizumab (PD-L1 monoclonal antibody) plus bevacizumab (a monoclonal antibody against vascular endothelial growth factor) versus sorafenib found a better objective response rate and survival for patients treated with the combination therapy [42]. Moreover, the combination of atezolizumab plus bevacizumab showed a better progression-free survival when compared to atezolizumab alone [43]. As the main concern for patients with liver cirrhosis treated with bevacizumab is upper gastrointestinal bleeding, it occurred in 7% of the patients who received the combination therapy which is comparable to earlier reports evaluating bevacizumab alone in patients with HCC [42,44,45]. However, proteinuria and hypertension, as main side effects of bevacizumab, still among the top side effects of combination therapy. However, further evaluation of combination therapy versus sorafenib or lenvatinib as first-line therapy for HCC is still under investigation. For example, the combination of nivolumab plus ipilimumab versus sorafenib/lenvatinib as first-line therapy for HCC is still under investigation by the CheckMate 9DW trial (NCT04039607), the combination of cabozantinib plus atezolizumab versus sorafenib is under investigation by COSMIC 312 trial (NCT03755791) and the combination between pembrolizumab plus lenvatinib versus lenvatinib alone is under investigation by the LEAP 002 trial (NCT03713593). Nevertheless, the success achieved by the landmark IMbrave 150 trial and G030140 trial has a great implication for the practice regarding the upfront therapy for patients with unresectable HCC. Nevertheless, these trials included only patients with early liver disease and the efficacy and safety of the combination therapy in patients with advanced liver disease is still unelucidated. Furthermore, no data available about subsequent therapy after the failure of immune checkpoint. More details are provided in Table 1.

Dual Immune Checkpoint Blockade
CTLA-4 is expressed on T regulatory cells regulating the early immune response after the primary stimulation by antigens mainly in lymphoid organs whereas PD-1 is expressed mainly on activated T cells in the tumor microenvironment regulating late immune response. Moreover, inhibition of the CTLA-4/B7 signal in lymph nodes increases activated CD8+ cells which will subsequently infiltrate the tumor and be part of the microenvironment [32,33]. Based on this, several studies have shown promising results with dual immunotherapy [34,35]. The success achieved in these trials especially in for patients with melanoma [36] has inspired the application of dual immune blockage for other types of cancers including HCC. Therefore, after the success achieved by the phase I/II trial investigating the efficacy and safety of dual immune therapy for patients progressed on sorafenib [37], a comparative randomized controlled trial, HIMALAYA study, was designed to compare Duravalumab versus the combination of Duravalumab plus Tremelimumab versus sorafenib. Its results are still bending. (Table 1)

Combination with Biological Therapy
Vascular endothelial growth factor (VEGF) has been linked with the development and progression of HCC [38,39]. Moreover, it has a role in immune suppression as it has been found that it creates an immunosuppressive microenvironment through the recruitment of several inhibitory cells such as t regulatory cells, tumor-associated macrophages and myeloid-derived suppressor cells. Those cells release cytokines such as IL-10 and TGF-β that inhibit natural killer cell and T cell activation and impedes dendritic cell maturation as shown in Figure 2 [33,40,41].

CTLA-4 Inhibition
Sangro et al. recruited 21 patients with hepatitis C virus who had progressed on previous lines of treatment for HCC. The treatment was tremelimumab at a dose of 15 mg/kg IV every 90 days. The drug showed a safe profile with a partial response rate of 17.6% [46]. Interestingly, the viral load for HCV decreased. Denoting the antiviral effect with the enhanced immunity. Moreover, the addition of ablation therapy to the anti-CTLA-4 showed a higher response rate with a similar safety profile [47] ( Table 2).

PD-1/PD-L1 Inhibition
The progression-free survival, overall survival and response rates were found to be better for patients treated with anti-PD-1/PD-L1 compared to placebo [25]. However, the data from the retrospective analysis did not show differences between anti-PD-1/PD-L1 when compared to regorafenib [48,49]. Interestingly, a combination of anti-PD-1/PD-L1 with radiation therapy showed better progression-free survival and overall survival when compared to anti-PD-1/PDL-1 alone [50] ( Table 2).

Dual Immune Checkpoint Blockade
Initial results of a single-arm study examining a combination between durvalumab and tremelimumab in patients with or without hepatitis infection. Out of 40 patients treated, 6 (15%) had an objective response rate with an acceptable safety profile [37] (Table 2).

Combination with Biological Therapy
Two studies evaluated the combination of biologic therapy with immunotherapy. The first evaluated ramucirumab plus duravalumab revealing an objective response rate of 11% and progression-free survival of 4.4 months [51]. The other one evaluated camrelizumab plus apatinib revealing an objective response rate of 44.4% and progression-free survival of 5.8 months [52] (Table 2).

Predictors of Response Using PD-L1 Expression
Immunohistochemical detection of PD-L1 has been studied in clinical trials as a predictor of response. It has been found that the expression of PD-L1 is associated with better overall response and survival outcomes [21,23]. A high tumor mutation burden (TMB), the number of somatic non-synchronous mutations in the genome of cancer cells, is a known predictive factor for response in different solid tumors. However, HCC has a low TMB compared to other solid malignancies which limited the predictive ability of this marker for HCC [53][54][55].

Use of Immune Checkpoint Inhibitors in Patients Autoimmune Diseases
One of the main concerns while treating patients with immune checkpoint inhibitors is immune-related adverse events which can be irreversible and even fatal [56][57][58]. Therefore, patients with a pre-existing auto-immune disease usually excluded from clinical trials [42], and, as a consequence, data about safety profiles in these populations is not available. However, liver cirrhosis can develop due to autoimmune diseases such as primary sclerosing cholangitis, autoimmune hepatitis, primary biliary cholangitis and so forth [59,60]. And, patients with HCC may suffer from another non-hepatobiliary autoimmune disease. Thus, understanding the underlying pathophysiological mechanisms and its interaction with the immune checkpoints' pathways is crucial in order to provide these patients with the therapeutic advantages without devastating side effects.
Several retrospective studies and case reports evaluated the safety profile of immune checkpoint inhibitors for patients with cancer and concomitant autoimmune disease [61][62][63][64][65][66][67][68]. Abdel-Wahab et al. conducted a systematic review evaluating the safety of immune checkpoint in patients with preexisting autoimmune disease and they found that; although some events may be severe and even fatal, most immune flares and immune-related side effects are managed without permanent drug discontinuation. However, for patients with neurological diseases such as myasthenia graves and multiple sclerosis, almost all patients developed exacerbation or immune-related side effects. Therefore, careful evaluation should be considered before prescribing immune checkpoints inhibitors for patients with neurological autoimmune diseases [61]. In a more recent large scale study, patients with a preexisting autoimmune disease treated with immune checkpoints had a higher risk of immune-related side effects than the control group. Furthermore, active disease and female gender were found to be independent predictors for the development of immune-related side-effects [62].
In summary, the immune-related side effects seem to be higher in patients with pre-existing autoimmune disease. Active disease and female gender are independent risk factors for immune-related side effects. Although immune-related side effects in patients treated with immune checkpoint inhibitors with a pre-existing autoimmune disease can be fatal, most cases are managed successfully without permanent discontinuation of the immune checkpoints. Nevertheless, these observations are derived from case reports and small retrospective studies and a well-designed large scale trial still represents an unmet need. Moreover, data about patients with HCC carcinoma specifically and hepatobiliary autoimmune diseases is still sparse.

Use of Immune Checkpoint Inhibitors in Patients with Inflammatory Bowel Disease
Patients suffering from inflammatory bowel disease (IBD) usually suffer from other hepatobiliary diseases such as drug-induced liver injury (about 30% of patients with IBD), primary sclerosing cholangitis (1.4% to 7.5% of patients with IBD), autoimmune hepatitis, primary biliary cirrhosis and nonalcoholic steatohepatitis [69]. These factors, along with the other traditional risk factors, can lead to HCC either directly or indirectly through liver cirrhosis [59]. Therefore, some patients who will suffer from HCC will have a concomitant IBD in which, as discussed before, immune checkpoint inhibitors with or without biologic therapy may be an option. However, the safety of immune checkpoint in this particular population is an ongoing and unanswered question. As known, the CTLA-4 and PD-1/PD-L1 signaling are crucial for gut homeostasis [70,71]. Interestingly, defects in the CTLA-4 gene or overexpression of PD-1/PD-L1 on intestinal epithelium were found to be higher in patients with IBD [72][73][74]. Figure 3 Therefore, in murine models, it was not surprising that the blockade of these pathways led to CD8 autoimmune enteritis [75]. And, it is not uncommon for a patient treated with immune checkpoint inhibitors to suffer from diarrhea [76]. Thus, IBD exacerbation during treatment with immune checkpoint inhibitors is a theoretical risk. Indeed, evidence about this question started to emanate from high volume centers. For example, in a recently published case series from Mayo Clinic, thirteen patients with a pre-diagnosed IBD were treated with immune checkpoint inhibitors and of them flare occurred in 4 patients (31%) [77]. This observation was also noted in a previous cohort in which 36% of patients with IBD treated with immune checkpoint inhibitors permanently discontinued them for IBD flare [68]. From a larger sample size study, data from a multicenter retrospective analysis included 102 patients with IBD treated with immune checkpoint inhibitors. Overall gastrointestinal side effects occurred in 42 patients (41%) after a median 62 days compared to 11% without IBD. Moreover, about 21% of patients with IBD treated with immune checkpoint inhibitors suffered from grade 3-4 diarrhea and 4 patients (3.9%) had intestinal perforation two of them had surgery [78]. Of note, the rate of intestinal perforation in patients receiving immune checkpoint inhibitors without concomitant IBD has been reported to be about 2.2% [79]. Importantly, most (~90%) of the included patients, in the aforementioned study evaluating the safety of immune checkpoints in patients with IBD, received a monotherapy of immune checkpoint inhibitors, only 10 patients (~10%) received a combination of two immune checkpoint inhibitors and none of the included patients received biologic therapy [78].
The combination of biologic therapy, especially bevacizumab, with immune checkpoint inhibitors in patients with IBD is of particular importance. Indeed, in patients not suffering from IBD, the intestinal perforation rate after using bevacizumab is about (1.5 to 2.5%) and severe bleeding is about 3% [80,81]. Importantly, the mortality rate for patients who develop intestinal perforation due to bevacizumab is high (up to 16%) [80,81]. Thus, even in non-gastrointestinal malignancies, the treatment with bevacizumab was found to be independently associated with a high risk of gastrointestinal perforation [82,83]. Indeed, the evidence about the safety of bevacizumab in patients with IBD is still lacking. Importantly, in patients with HCC, liver cirrhosis is common and gastrointestinal bleeding especially esophageal varices is a major concern, especially when selecting bevacizumab for treatment [84]. Overall, the use of atezolizumab with bevacizumab in patients with IBD carries a risk for intestinal perforation, gastrointestinal bleeding and the safety profile is still lacking in the literature. Binding of CD28 to B7 was associated with induction of immune response through upregulation of production of IL2. On the other side, CTLA-4 B7 binding regulates the late immune response by decreasing IL2 [71]. Therefore, inhibition of CTLA-4 by antiCTLA-4 antibodies was found to be associated with an exaggerated immune response which might lead to colitis [85,86]. PD-1/PD-L1 binding leads to immune response regulation through PI3K and AKT pathways. Therefore, inhibition of this binding might lead to immune response dysregulation which might lead to colitis and autoimmune exacerbation [87].

Novel Immunotherapies
With the recent advances in the immunotherapeutic mechanisms, novel immunotherapies have gained popularity. Different therapeutic targets have been evaluated such as lymphocyte activation gene 3 (LAG-3). LAG-3 is first described by Triebel et al. and thereafter it was found to be overexpressed on the activated T cytotoxic and T regulatory cells with a negative impact on T helper cells. Therefore, during tumorigenesis, cancer cells use this pathway to escape from the immune system. Therefore, immunoglobin against LAG-3 has been investigated in different clinical trials [88]. Several other novel targets including T cell immunoglobulin and ITIM domain (TIGIT), T cell immunoglobulin and mucin domain-containing -3 (TIM-3) and B and T lymphocyte attenuator (BTLA) have been evaluated in clinical trials [88]. The current ongoing phase II trial [NCT03680508] is designed to evaluate the efficacy of TIM-3 in combination with PD-1 antibody for patients with HCC with no results published yet.
One other therapeutic target is the killer immunoglobulin-like receptor (KIR) which has an inhibitory effect on the NK cells. Therefore, Lirilumab, a KIR antibody, is under investigation in combination with immune checkpoint inhibitors in clinical trials [89]. Binding of CD28 to B7 is associated with induction of immune response through upregulation of production of IL2. On the other side, CTLA-4 B7 binding regulates the late immune response by decreasing IL2 [71]. Therefore, inhibition of CTLA-4 by antiCTLA-4 antibodies was found to be associated with an exaggerated immune response which might lead to colitis [85,86]. PD-1/PD-L1 binding leads to immune response regulation through PI3K and AKT pathways. Therefore, inhibition of this binding might lead to immune response dysregulation which might lead to colitis and autoimmune exacerbation [87].
Overall, the use of atezolizumab with bevacizumab in patients with IBD carries a risk for intestinal perforation, gastrointestinal bleeding and the safety profile is still lacking in the literature.

Novel Immunotherapies
With the recent advances in the immunotherapeutic mechanisms, novel immunotherapies have gained popularity. Different therapeutic targets have been evaluated such as lymphocyte activation gene 3 (LAG-3). LAG-3 is first described by Triebel et al. and thereafter it was found to be overexpressed on the activated T cytotoxic and T regulatory cells with a negative impact on T helper cells. Therefore, during tumorigenesis, cancer cells use this pathway to escape from the immune system. Therefore, immunoglobin against LAG-3 has been investigated in different clinical trials [88]. Several other novel targets including T cell immunoglobulin and ITIM domain (TIGIT), T cell immunoglobulin and mucin domain-containing -3 (TIM-3) and B and T lymphocyte attenuator (BTLA) have been evaluated in clinical trials [88]. The current ongoing phase II trial [NCT03680508] is designed to evaluate the efficacy of TIM-3 in combination with PD-1 antibody for patients with HCC with no results published yet.
One other therapeutic target is the killer immunoglobulin-like receptor (KIR) which has an inhibitory effect on the NK cells. Therefore, Lirilumab, a KIR antibody, is under investigation in combination with immune checkpoint inhibitors in clinical trials [89].
Overall, the novel immunotherapies' investigation in HCC is still restricted to being a part of evaluation of their role in solid tumors in general. Therefore, a better understanding of these pathways and their contribution to the HCC microenvironment is needed.

Conclusions
Hepatocellular carcinoma treatment represents a real challenge in patients with cirrhosis and several pharmacological [72][73][74][75] and loco-regional [76][77][78][79][80][81] therapies have been tested with mixed results. A combination of immune checkpoint inhibitors with biologic therapy seems to be promising for a new therapeutic standard of care for patients with unresectable HCC. However, for the subset of patients such as patients with preexisting autoimmune disease, inflammatory bowel disease or nonalcoholic steatohepatitis, the safety and efficacy are still not well established and further studies are needed to address all these open unanswered questions.