The efficacy of new therapeutic regimens has imposed a re-evaluation of the role of systemic therapy among the therapeutic options of Hepatocellular carcinoma (HCC) and opened new scenarios, such as the opportunity of sequential treatments ^[1][25]. In fact, the use of systemic therapy in combination with other techniques could optimize the outcomes and lower the currently high recurrence rates. In patients with HCC scheduled for LT, systemic treatment can be used as bridging therapy with the aim of reducing the risk of dropout due to disease progression, currently estimated at 25% at 1 year ^[2][44]. In patients not eligible for LT, on the other hand, adjuvant or neoadjuvant systemic therapies could reduce recurrence rates after loco-regional therapy, potentially increasing OS, similar to what has already been validated for non-hepatic neoplasms. Indeed, pre- or post-operative antiproliferative and antiangiogenic therapies could be able to decrease the occult intra- and extrahepatic dissemination frequently associated with the risk of early recurrence. Previous studies could not demonstrate a clear efficacy of neoadjuvant or adjuvant treatments with sorafenib, although some evidence suggested that adjuvant therapy could increase recurrence-free survival (RFS) ^[3][45]. The availability of new therapeutic regimens that have proven to be more effective than sorafenib and potentially complementary to it requires a re-evaluation of the effectiveness and role of systemic therapies before or after loco-regional treatment. In this regard, growing evidence is currently available, and several trials are underway.

Similar to what has been validated for other non-hepatic neoplasms, in patients with high tumor burden (i.e., single large HCC, tumors adjacent to major vascular structures requiring narrow-margin hepatectomy, unilobar multifocal disease), neoadjuvant therapy could allow the downstaging of the tumor, favoring safer and more radical LR. Furthermore, neoadjuvant systemic therapy schemes could reduce the risk of early recurrence after loco-regional treatment and provide useful information on tumor susceptibility to anticancer drugs ^[4][28]. Before the use of ICIs in the treatment of HCC, the low efficacy and poor tolerability profile of the available drugs did not favor the development of neoadjuvant therapy schemes. However, the availability of new molecules with greater efficacy and a good tolerability profile has re-opened the possibility of pre-operative systemic therapies also in the HCC setting.

The current evidence on the role of systemic neoadjuvant therapy before loco-regional therapy for HCC are illustrated in Table 1. Some phase I ^[5][46] and II ^{[6][7][8][9][10]}[47,48,49,50,51] trials have confirmed the feasibility and safety of neoadjuvant regimens in the treatment of the HCC and opened the way for their clinical study and validation. Preliminary data provided by Ho et al. ^[5][46] showed that neoadjuvant systemic therapy with cabozantinib + nivolumab was feasible and could result in margin-negative resections. This single arm phase Ib study performed on 15 patients with locally advanced HCC showed in particular that 12 patients (80%) had undergone successful negative margin resection after neoadjuvant therapy, and 5 patients had major pathologic responses. Similarly, Marron et al. ^[6][47] preliminarily evaluated the efficacy of neoadjuvant cemiplimab (an anti-PD-1) in patients with resectable HCC. A total of 21 patients with stage Ib, II and IIIb HCC received two cycles of cemiplimab (350 mg intravenously) every 3 weeks before LR and an additional eight cycles after. Among the 21 patients who underwent successful resection, 20% had achieved >70% tumor necrosis, 15% had a partial response, and all other patients showed stable disease. Kaseb et al. ^[7][48] evaluated safety and tolerability of perioperative nivolumab with or without ipilimumab in 27 patients with resectable HCC. The authors highlighted that both therapeutic schemes were safe and feasible and that adverse events (more frequent for the nivolumab + ipilimumab combination therapy) did not delay the LR. Preliminary data also showed that estimated median progression-free survival was 9.4 months with nivolumab and 19.53 months with nivolumab plus ipilimumab. Finally, Zhu et al. ^[9][50] analyzed the safety and impact of the neoadjuvant TACE plus PD-1 inhibitor (camrelizumab or sintilimab) in patients with intermediate-stage HCC. Patients receiving this regimen demonstrated 1-year and 2-year OS rates of 100.0% and 76.4%, whereas the 1-year RFS rate was 86.6%. Successful downstaging has been reported in 70% of patients undergoing neoadjuvant therapy and is associated with more prolonged RFS than patients with failed downstaging.

However, case-control trials are needed to validate these treatment protocols and confirm their efficacy. In this regard, Wu et al. ^[11][52] recently evaluated retrospectively the outcomes of patients treated with hepatectomy alone versus patients treated with triple neoadjuvant therapy using the combination of lenvatinib, anti-PD-1 monoclonal antibodies and TACE before LR. OS and RFS were significantly longer in the triple therapy group than in the LR-alone group. After 12 and 24 months of follow-up from the LR, the neoadjuvant therapy group showed an OS rate of 100% and 85.7% and an RFS rate of 66.95% and 48.8%, respectively, whereas the LR-alone group showed a rate of OS of 73.7% and 48.7% and an RFS rate of 28.34% and 22.99%. Furthermore, triple therapy reduces the rate of major vascular invasion and increases the probability of margin-negative resections. In addition to the benefit in patients with resectable HCC, this triple neoadjuvant therapy has been shown to favor the downstaging of unresectable HCC and allow resection in more than half cases with high objective response rates ^[12][53].

In this setting, triple combination therapy proved to be superior to dual combination therapy (Lenvatinib plus TACE) ^[13][54]. Wu et al. ^[11][12][52,53] hypothesize that lenvatinib may suppress HCC angiogenesis and antitumor immunity, potentially increasing the antiproliferative effect of PD-1 antibodies (not able alone to obtain adequate levels of anticancer immunity by PD-L1/PD-1 axis blockade) and counteracting the undesired neoangiogenic action of TACE (per se capable of increasing the efficacy of PD-1 antibodies through the release of tumor-specific antigens). The combination therapy would therefore act on different immune and antiproliferative patterns with a complementary and synergistic action of the individual treatments, resulting in high tumor response rates and a significant improvement in OS. Although the study has several limitations (retrospective design, significant imbalance between groups, abstract availability only), Xia et al. ^[14][55] provide additional data to support the potential efficacy of neoadjuvant therapy in HCC. In particular, neoadiuvant ICI (camrelizumab) combined with anti-angiogenic targeted drug (apatinib) for resectable HCC can reduce the 1-year recurrence rate and improve the 1-year survival rate, especially for those with solitary tumor.

Currently there are several ongoing trials in this setting. Pinato et al. ^[15][56], for example, are evaluating safety and bioactivity of the neoadiuvant nivolumab/ipilimumab combination before LR in early-stage HCC (trial registration number: NCT03682276). Li et al. ^[16][57] are, instead, analyzing the safety and efficacy of the combination scheme lenvatinib plus sintilimab plus radiotherapy as neoadjuvant treatment regimen in patients with HCC and portal vein tumor thrombus (NCT05225116). Finally, there is an ongoing evaluation by Zhang et al. ^[17][58] of the efficacy of neoadjuvant tislelizumab with stereotactic body radiotherapy in patients with resectable HCC (NCT05185531).

If the main goal of neoadjuvant treatment schemes is to improve the resectability of tumor, adjuvant therapy after loco-regional treatment for HCC aims to reduce the recurrence rate and increase overall and disease-free survival. The two types of treatment (loco-regional and systemic) could have complementary effects: on the one hand, loco-regional treatments reduce tumor burden and induce the release of tumor antigens and proinflammatory cytokines; on the other, VEGF and TKIs boost antitumor immunity.

Some retrospective studies ^{[18][19][20][21][22][23][24][25]}[59,60,61,62,63,64,65,66] or small clinical trials ^[26][27][67,68] have found that adjuvant therapy with sorafenib can reduce the postoperative recurrence rate and improve OS. In particular, Zhang et al. ^[18][59] retrospectively analyzed the efficacy of adjuvant treatment with sorafenib in 113 patients who underwent R0 LR for HCC with microvascular invasion (MVI) comparing them to data from 113 matched patients who undergone LR alone. In this study adjuvant treatment with sorafenib was associated with significantly higher OS and RFS than the LR alone group, with similar results for both early or very early HCC (BCLC 0-A) and intermediate HCC (BCLC B).

Conversely, one of the few randomized clinical trials (RCTs) in this setting showed opposite results. In the STORM trial, Bruix et al. ^[28][69] prospectively evaluated the efficacy of sorafenib in 1114 patients with HCC after a complete radiological response after LR (n = 900) or RFA (n = 214). No differences in median RFS between the two groups (33.3 months in the sorafenib group vs. 33.7 months in the placebo group, p = 0.26) were noted. Furthermore, sorafenib was associated with poor toxicity profile (28% of patients experiencing grade 3 or 4 adverse event in the sorafenib group vs. <1% of patients in the placebo group), with four treatment-related deaths. However, a recent meta-analysis conducted on 2655 patients from 13 studies seems to contradict what was previously found in the STORM trial ^[29][70]. Indeed, the combined results of the studies would indicate that adjuvant therapy with sorafenib after LR could significantly increase OS (hazard ratio, HR = 0.71, 95%: confidence interval, CI = 0.59–0.86) and RFS (HR = 0.68, 95% CI = 0.54–0.86) and reduce recurrence rates. However, the authors underline that further evidence is needed before reaching definitive conclusions.

Recently, Lin et al. ^[30][71] evaluated the benefit of the combination of TACE + TKIs (sorafenib, lenvatinib or apatinib) after curative LR in HCC patients at high risk of early recurrence. The combination of adjuvant TACE + TKIs after LR has been shown to significantly increase the 1-year and 2-year RFS rate (45.5% and 34.9%, respectively) compared to post-operative TACE alone (26.8% and 18.3%, respectively). Patients with HCC diameter ≥ 5 cm, number of lesions < 3, absence of vascular and/or biliary infiltration, capsule integrity and stage IIIB (according to the American Joint Committee on Cancer 8th staging system) could benefit more from adjuvant TACE plus TKI treatment. A phase IIb clinical trial is currently underway aiming to evaluate the efficacy of adjuvant donafenib combined with TACE in patients with HCC at a high risk of recurrence after LR ^[31][72].

Overall, data available on adjuvant therapy for HCC with TKIs seem to indicate that treatments targeting only angiogenesis could be ineffective and could benefit from the association with molecules having complementary mechanism of action. In fact, in addition to the well-known risk factors of relapse (e.g., high tumor burden, microvascular invasion), numerous immune mechanisms have been shown to play a role in HCC recurrence. Interestingly, the presence of low-grade tumor infiltration by immune effector cells (e.g., CD4+ T cells, CD8+ T cells and NK cells) is associated with a greater risk of recurrence compared to HCC with higher levels of immune infiltration ^[32][33][84,85]. For these reasons, immunotherapy was evaluated in the adjuvant therapy setting of HCC.

In this regard, some studies have analyzed the role of cancer vaccination (Table 2). Through the activation of effector T-cells and the development of an immunological memory, this vaccine should alter the tumor microenvironment and induce tumor-specific immunological effects, potentially favoring an adjuvant action after HCC treatment ^[34][27]. The first data provided by Kuang et al. ^[35][73] through a small phase II randomized trial demonstrated the safety, feasibility and efficacy of adjuvant autologous formalin-fixed tumor vaccine for preventing postsurgical recurrence of HCC. Compared to the group of patients not receiving adjuvant therapy, the risk of recurrence in vaccinated patients was reduced by 81%, with a significant prolongation of time to first recurrence and improvement in both RFS and OS. More recently, Lee et al. ^[36][75] evaluated the role of adjuvant infusion of autologous dendritic cells (DCs) pulsed with tumor-associated antigens after loco-regional treatment for HCC. DC vaccine significantly reduced the risk of recurrence of non-RFA group patients, whereas it unexpectedly increased this risk in RFA group. Finally, Shimizu et al. ^[37][74] analyzed the role of adjuvant combination therapy of both DC vaccines and CD3-activated T-cell transfer (ATVAC) after LR for HCC. Adjuvant combination significantly improved RFS and OS (24.5 and 97.7 months in the adjuvant ATVAC group and 12.6 and 41.0 months in the LR alone group, respectively).

Some clinical trials evaluated the efficacy of adjuvant immunotherapy using CIK or LAK cells (adoptive immunotherapy, AIT) after loco-regional treatment for HCC, with not univocal but overall favorable results (Table 2). Before others, Takayama et al. ^[38][76] evaluated the safety and efficacy of infusion of autologous lymphocytes activated with recombinant interleukin-2 and antibody to CD3 in patients with HCC undergoing curative LR. They showed that the adjuvant AIT decreased the recurrence rate by 18% and significantly prolonged the time to first recurrence compared with LR alone. However, no impact was demonstrated on the OS. Similar results (recurrence rate reduction, RFS prolongation, no impact on OS) were later confirmed also with adjuvant immunotherapy with CIK cells by Hui ^[39][77] and Xu ^[40][78] et al. In contrast, the RCT by Lee et al. ^[41][79] and the subsequent 5-year follow-up ^[42][83] showed the efficacy of adjuvant therapy with autologous CIK cells in determining a significant increase in both RFS and OS. In particular, RFS rate was 44.8% in the immunotherapy group and 33.1% in the control group after 5 years of follow-up, with significantly reduced risk of recurrence (HR: 0.67; 95% CI: 0.48–0.94) and death from all causes (HR: 0.33; 95% CI: 0.15–0.76) for both. Overall, all the meta-analyses available to date ^[43][44][45][80,81,82] agree in confirming the efficacy of adjuvant AIT in increasing both RFS and OS significantly. However, this benefit appears to be time-dependent as it lasts only the first 3 ^[43][45][80,82] or 5 years ^[44][81] after treatment. AIT would, in fact, be able to effectively counteract the occult intra- or extrahepatic neoplastic dissemination responsible for early recurrences but would not affect the carcinogenic substrate determined by cirrhosis, which is the cause of de novo tumorigenesis and late recurrence, instead ^[34][27].

In the context of immunotherapeutic drugs, ICIs could radically change the scenario of adjuvant therapy in the treatment of HCC. In addition to lowering the risk of early recurrence by improving systemic clearance of occult residual disease, unlike TKIs and AIT, adjuvant therapy with ICIs could also have a chemo-preventive effect as it reduces the incidence of de novo HCC by activating immune surveillance processes ^[46][86]. This could also decrease the late recurrence rate, overcoming the limitations showed by previous systemic therapies. Although the rationale for the use of ICIs in adjuvant therapy protocols is extremely promising, no data on this issue are available yet. However, there are several ongoing clinical trials aiming to evaluate the efficacy of ICIs in the setting of adjuvant therapy after loco-regional treatment for HCC. IMbrave 050 (NCT04102098) is one of the most interesting, randomized, open-label phase III trial, launched at the beginning of 2020 with the purpose to evaluate the effect of dual PD-L1/VEGF blockade using atezolizumab plus bevacizumab in high-risk HCC after curative resection or ablation ^[47][87]. The first results are expected by the end of 2023. Similarly, phase III double-blinded, two-arm KEYNOTE-937 (NCT03867084) ^[48][88] and CheckMate 9DX (NCT03383458) ^[49][89] trials are two ongoing studies with the aim to assess the safety and efficacy, respectively, of pembrolizumab (the former) and nivolumab (the latter) versus placebo as adjuvant therapy in HCC patients with complete radiological response after LR or RFA. Finally, EMERALD-2 (NCT03847428) is a phase III, randomized, double-blind, placebo-controlled, multi-center trial aiming to assess the efficacy and safety of durvalumab in combination with bevacizumab or durvalumab monotherapy or placebo as adjuvant therapy in HCC patients at high risk of recurrence ^[50][90]. It is likely that patients at high risk of recurrence (high tumor burden, large or multifocal neoplasm, poor degree of differentiation, microvascular invasion) could enjoy the greatest benefit from adjuvant therapy with ICIs ^[51][91]. The results of these and other trials could provide useful data for the validation of adjuvant immunotherapy schemes after loco-regional treatment and define the predictors of treatment response, opening the way for a new era in the management of HCC patients together with selecting those who could benefit more from these therapeutic protocols.

Although it cannot be considered an antineoplastic therapy, the antiviral treatment for chronic HBV infection in patients with HCC deserves special mention. In fact, the presence of viral replication correlates with an increased risk of both early and late recurrence. Patients undergoing LR for HCC with high preoperative HBV-DNA levels show lower median OS and RFS compared to those with low viral load ^[52][53][5,92]. If the association between active HBV infection and increased risk of late recurrence appears intuitive in relation to the persistence of the cause of liver damage and hepatocarcinogenesis, the higher risk of early recurrence seems to be related to the greater probability (about 40% more) of microvascular invasion in viraemic versus non-viraemic patients ^[54][93]. In the presence of both high ^[54][55][93,94] and low ^[56][95] pre-operative viral load, starting antiviral therapy after loco-regional treatment (LR or RFA) allows to reduce both early and late HCC recurrence rates ^[57][96] and improve OS ^[58][97]. Despite comparable efficacy on virological outcomes, tenofovir disoproxil fumarate (TDF) therapy seems associated with a significantly lower risk of recurrence after LR compared to entecavir (ETV) ^[59][98]. Indeed, 5-year HCC recurrence rate after LR is 33.6% in patients treated with TDF and 44.5% in those treated with ETV ^[59][98]. Unlike ETV, TDF induces the synthesis of high serum levels of interferon-lambda 3 (IFN-λ3) ^[60][99], which is able to exert a strong antitumor activity ^[61][62][100,101]. The antineoplastic action of TDF-induced IFN-λ3 seems to be additive to the ability to stop viral replication and turn off liver necroinflammatory activity, which is common to both ETV and TDF. No difference in recurrence rates between the two drugs was noted after RFA, instead ^[63][102].

The same concepts also seem applicable for treatment with direct-acting antiviral agents (DAAs) during chronic HCV infection. After an initial phase in which some data had suggested potential increased risk of early recurrence after HCV eradication using DAAs ^[64][65][103,104], numerous studies confirm that recurrence rates after antiviral therapy did not differ between patients who received IFN-based or IFN-free therapy ^[66][67][105,106]. On the contrary, the most recent data suggest lower risk of early and late HCC recurrence after viral eradication achieved by DAA ^[68][69][107,108], as well as an increase in OS ^[69][70][108,109] and improvement in hepatic ^[71][110] and extrahepatic ^[72][73][111,112] outcomes. Some predictors, including the baseline liver stiffness, could allow to stratify the residual risk of HCC occurrence or recurrence ^[74][113].

Other non-chemotherapy drugs were finally evaluated for their impact on HCC recurrence rates after loco-regional treatment. In this regard, some retrospective data would indicate that the use of angiotensin II receptor 1 blockers (sartans) after RFA is associated with longer OS and delayed time to recurrence ^[75][114]. However, specific clinical trials appear necessary in order to give preferential indication for the use of this class of drugs in the treatment of comorbidities (e.g., arterial hypertension) ^[76][115].