Following the approval of atezolizumab (anti-PD-L1) plus bevacizumab, an anti-vascular endothelial growth factor A (anti-VEGF-A), as first-line therapy for unresectable HCC in 2020, treatment with ICIs has generated growing interest for the treatment of HCC. The combination of atezolizumab plus bevacizumab in IMbrave150, a global, multicenter, open-label, phase 3 randomized trial, resulted in superior outcomes in terms of overall survival (OS) and progression-free survival (PFS) over the tyrosine kinase inhibitor sorafenib, which had previously been the standard of systemic therapy for patients with advanced HCC (OS at 12 months: 67.2% vs. 54.6%; median PFS: 6.8 months vs. 4.3 months; p < 0.001) [11]. Another phase 2/3 randomized trial investigating the combination of sintilimab (anti-PD-1) plus IBI305 (a bevacizumab biosimilar) as first-line therapy revealed increased OS and PFS over sorafenib (median OS: NA vs. 10.4 months; median PFS: 4.6 months vs. 2.8 months; p < 0.0001) in patients with unresectable HBV-related HCC [26]. In the more recent randomized, open-label, sponsor-blind, multicenter, global, phase 3 HIMALAYA trial, durvalumab (anti-PD-L1) monotherapy was found to be non-inferior to sorafenib (16.56 months vs. 13.77 months, respectively; p = 0.0035) and the combination of durvalumab plus tremelimumab (anti-CTLA-4) was shown to improve median OS over sorafenib (16.43 months vs. 13.77 months, respectively; p = 0.0035) in patients with unresectable HCC [12]. Updated findings were recently reported by the European Society for Medical Oncology (ESMO), in which the combination of durvalumab plus tremelimumab was found to yield a higher 4-year OS rate compared to sorafenib (25.2% vs. 15.1%, respectively) [27]. Currently, nivolumab (anti-PD-1) monotherapy, nivolumab plus ipilimumab (anti-CTLA-4), pembrolizumab (anti-PD-1), ramucirumab (anti-VEGF-R2), cabozantinib (mTKI), and regorafenib (mTKI) monotherapy are approved by the FDA as second-line therapy for unresectable HCC following the results of several clinical trials [28,29,30,31,32,33]. Results from a phase II study evaluating treatment with tislelizumab (anti-PD-1) as second-line therapy in patients with unresectable HCC have demonstrated an acceptable safety and tolerability profile with an overall response rate (ORR) of 12.4%, a median PFS of 2.7 months, and median OS of 12.4 months [34]. Tislelizumab is currently under study in comparison with sorafenib as first-line therapy in a phase 3 trial in patients with unresectable HCC (NCT03412773). Similarly, camrelizumab (anti-PD-1) was investigated as second-line treatment in a phase 2 trial for patients with advanced HCC and demonstrated anti-tumor activity with an ORR of 14.7%, median PFS of 2.1 months, and a median OS of 13.8 months [35]. A phase 3 trial looking to compare camrelizumab plus apatinib (a VEGF-R2 inhibitor) versus sorafenib as first-line therapy for patients with advanced HCC reported significantly improved outcomes in the combination group over sorafenib (median OS: 22.1 vs. 15.2 months; median PFS: 5.6 vs. 3.7 months, respectively; p < 0.0001) [36]. Moreover, toripalimab (anti-PD-1) plus anlotinib (a VEGF-R2 inhibitor) showed promising results as first-line therapy for patients with unresectable HCC with an ORR of 29%, median PFS of 11 months, and median OS of 18.2 months [37].

Immune checkpoint molecules correspond to ligand–receptor pairs expressed on immune cells, APCs, and tumor cells possessing inhibitory or stimulatory functions, which serve to mediate the innate and adaptive immune responses. The presence of these molecules on tumor cells does not only facilitate escape from the immune system but is also involved in maintaining tumor activities, such as epithelial–mesenchymal transition, self-renewal, metastasis, resistance to anti-tumor drugs, angiogenesis, and anti-apoptosis, among others [38]. Examples of inhibitory immune checkpoint receptors include CTLA-4, PD-1, T cell immunoglobulin and mucin domain 3 (TIM-3), lymphocyte activation gene 3 (LAG-3), and killer immunoglobulin-like receptors (KIRs) [39].

The activation or inhibition of the immune response begins with the binding of the MHC receptors on T cells to the tumor-associated antigens (TAMs) on APCs. This process also requires additional costimulatory or coinhibitory binding associations, which constitute the basis of the immune checkpoint. For instance, the costimulatory marker CD28 on T cells binds to CD80 (B7-1) and CD86 (B7-2) on APCs. The result of both signals is the activation of the immune response as T cells proliferate and cytokines are released [40]. CTLA-4, which is expressed on active CD4⁺, CD8⁺, and regulatory T cells, is upregulated following T cell activation and acts as a competitive inhibitor of CD28 and binds to its receptors with higher affinity, thus diminishing T cell costimulation and curtailing the T cell response [40,41,42]. While its role under healthy conditions is to modulate T cell activity by inhibiting excessive T cell activation, CTLA-4, in the context of cancer, prevents the proliferation and activation of tumor-specific T cells [43]. Anti-CTLA-4 ICIs, such as ipilimumab and tremelimumab, bind to the CTLA-4 receptor in order to lift the inhibitory signal and favor the activation and proliferation of T cells and, therefore, the enhancement the immune response [40,44]. Similarly, anti-PD-1 and anti-PD-L1 ICIs serve to block binding at the PD-1/PD-L1 or PD-1/PD-L2 checkpoints and prevent the interaction between T cells and other immune cells (e.g., B cells, myeloid cells) expressing PD-1 and tumor cells, APCs, and other immune cells expressing PD-1 ligands [45]. In the absence of these ICIs, the binding between PD-1 and PD-L1 allows tumor cells to evade immune detection by the inactivation of T cells through the dephosphorylation of T cell-activating kinases [46]. In addition, the prevention of interleukin-2 (IL-2), interferon-𝛾 (IFN-𝛾), and tumor necrosis factor-𝛼 (TNF-𝛼) release and the decrease in T cell survival are characteristic of both CTLA-4 and PD-1 signaling. However, these pathways differ in the timing and location of their inhibitory functions as CTLA-4 acts in the lymph nodes early in the immune response during T cell priming while PD-1 generally acts later in the immune response during the T cell effector phase in peripheral tissues [45].

Table 1 summarizes the timing, location, and mechanisms of action of several ICIs that have been administered to patients with HCC.

Combinations of ICIs, such as atezolizumab plus bevacizumab and tremelimumab plus durvalumab, have shown promising additive anti-tumor activity for patients with HCC [11,12]. In the HIMALAYA trial, a single dose of tremelimumab (anti-CTLA-4) was given as a priming dose and was followed by durvalumab (anti-PD-L1), which was given every 4 weeks to patients with unresectable HCC [12]. Similarly, the CheckMate 040 trial aimed to assess different dosing plans of nivolumab (anti-PD-1) plus ipilimumab (anti-CTLA-4) in HCC patients that had received sorafenib treatment and reported the highest ORR in the treatment arm that received nivolumab plus a higher dose of ipilimumab (4 doses/every 3 weeks) followed by nivolumab every 2 weeks [29]. The rationale behind administering a priming dose of an CTLA-4 inhibitor followed by a PD-L1 inhibitor is that anti-CTLA-4 drugs can restore cytotoxic T cell activation in lymphoid tissues, which would consequently result in increased CD8⁺ T cell infiltration at the tumor site. In the absence of CTLA-4 blockers, the inhibition of the PD-1/PD-L1 checkpoint would not lead to anti-tumor activity as T cells would have been inactivated in the priming phase, and since they have not yet reached the effector phase they would be absent from tumor tissues [43]. In addition, the combination of VEGF blockers (e.g., bevacizumab or ramucirumab) with PD-1/PD-L1 ICIs has displayed the ability to remodel the immunosuppressive microenvironment characteristic of tumor growth. In the case of solid tumors, such as HCC, VEGF is secreted by hypoxic tumor cells and vascular endothelial cells, and leads to the recruitment of regulatory T cells, TAMs, and MDSCs, which in turn release additional VEGF and immunosuppressive cytokines (e.g., IL-10, TGF-β). VEGF also affects both the priming and effector stages by suppressing the maturation of dendritic cells and the presentation of antigens and by inhibiting the infiltration of activated T cells into cancer tissues, respectively. Consequently, anti-VEGF drugs facilitate the restoration of an immunostimulatory microenvironment. This is carried out through (1) a decrease in the production and release of regulatory T cells, TAMs, MDSCs, IL-10, and TGF-β, (2) the enhancement of CD8⁺ T cell activation in the priming stage by promoting dendritic cell antigen presentation, and (3) the normalization of tumor vasculature to allow the migration of activated T cells to the cancer. The addition of PD-1/PD-L1 ICIs serves to promote the anti-tumor response by activated T cells [47]. Currently, the TRIPLET-HCC phase 2/3 trial (NCT05665348) is planned to assess the efficacy and safety of combining an anti-CTLA-4 with an anti-PD-L1 and an anti-VEGF agent in patients with HCC by comparing treatment with ipilimumab plus atezolizumab and bevacizumab to the atezolizumab plus bevacizumab combination [48].