Percutaneous Ablation-Induced Immunomodulation in Hepatocellular Carcinoma

Hepatocellular carcinoma (HCC) is one of the most common causes of cancer-related deaths worldwide and its incidence is rising. Percutaneous locoregional therapies, such as radiofrequency ablation and microwave ablation, are widely used as curative treatment options for patients with small HCC, but their effectiveness remains restricted because of the associated high rate of recurrence, occurring in about 70% of patients at five years. These thermal ablation techniques have the particularity to induce immunomodulation by destroying tumours, although this is not sufficient to raise an effective antitumour immune response. Ablative therapies combined with immunotherapies could act synergistically to enhance antitumour immunity. This review aims to understand the different immune changes triggered by radiofrequency ablation and microwave ablation as well as the interest in using immunotherapies in combination with thermal ablation techniques as a tool for complementary immunomodulation.


Introduction
Hepatocellular carcinoma (HCC) is a primary liver cancer with increasing incidence and high lethality. It often develops against the background of chronic inflammation and cirrhosis caused by chronic liver diseases such as chronic viral hepatitis B or C, alcohol-related liver disease or steatohepatitis. The diagnosis is delayed due to the absence of symptoms and HCC is often diagnosed at an intermediate or advanced stage. Thus, accessible treatments are often palliative instead (~70%) of curative (~30%).
The most effective treatment for HCC remains liver transplantation as it treats both the HCC and the underlying liver disease, but due to strict eligibility criteria and the shortage of organs, this solution is not an option for the majority of HCC patients. Percutaneous thermal ablations such as radiofrequency ablation (RFA), microwave ablation (MWA) and cryoablation are locoregional therapies that constitute the main alternatives to surgical resection. Due to underlying cirrhosis and micrometastases, the rate of recurrence is quite high, occurring in 70% of patients at five years. These minimally invasive procedures are safe and have been demonstrated to induce immunogenic necrosis through mechanisms that will be detailed in this review.
Recently, immunotherapies, mainly immune checkpoint inhibitors of the programmed cell death protein 1 (PD-1)/programmed death-ligand 1 (PD-L1) pathway, have emerged as an encouraging Cell apoptosis, which occurs during cryoablation, is described as an active, programmed process of autonomous cellular death that induces immunosuppression rather than antitumour response stimulation.

Immune Changes Induced by RFA
Several studies have investigated the effects of RFA on the immune system as it is the most frequently used thermal ablation technique [7]. Heat shock proteins (HSP) are chaperone proteins implicated in multiple oncogenic pathways. They are highly expressed in cancer cells, either intracellularly or extracellularly. Following RFA, HSP-70 is abundantly released in the serum and seems to be an activator of the innate and adaptive immune response, chaperoning antigenic peptides and promoting danger signals. The release of HSP-70 can be responsible for local inflammation and the activation of antigen-presenting cells in the tumour area, leading to antitumour response stimulation [13]. Another study demonstrated that RFA enhances the cross presentation with antigen-presenting cells. Using a murine model, Dromi et al. showed that RFA results in increased infiltration of dendritic cells (DCs) in the tumour and in significantly enhanced CD4+ and CD8+ T cell responses [14]. As DCs present antigens to naïve T cells, the evaluation of the number of tumourassociated antigen (TAA)-specific T cells is done after RFA has been conducted. Consistent with the previous observation, Mizukoshi et al. determined that after RFA, TAA-specific T cells significantly increase, and they mostly comprise CD8+ T cells [15]. Based on the phenotypic characterisation of circulating lymphocytes, it has been demonstrated that the amount of central memory lymphocytes

Immune Changes Induced by RFA
Several studies have investigated the effects of RFA on the immune system as it is the most frequently used thermal ablation technique [7]. Heat shock proteins (HSP) are chaperone proteins implicated in multiple oncogenic pathways. They are highly expressed in cancer cells, either intracellularly or extracellularly. Following RFA, HSP-70 is abundantly released in the serum and seems to be an activator of the innate and adaptive immune response, chaperoning antigenic peptides and promoting danger signals. The release of HSP-70 can be responsible for local inflammation and the activation of antigen-presenting cells in the tumour area, leading to antitumour response stimulation [13]. Another study demonstrated that RFA enhances the cross presentation with antigen-presenting cells. Using a murine model, Dromi et al. showed that RFA results in increased infiltration of dendritic cells (DCs) in the tumour and in significantly enhanced CD4+ and CD8+ T cell responses [14]. As DCs present antigens to naïve T cells, the evaluation of the number of tumour-associated antigen (TAA)-specific T cells is done after RFA has been conducted. Consistent with the previous observation, Mizukoshi et al. determined that after RFA, TAA-specific T cells significantly increase, and they mostly comprise CD8+ T cells [15]. Based on the phenotypic characterisation of circulating lymphocytes, it has been demonstrated that the amount of central memory lymphocytes (CD45RA-/CCR7+) increases significantly after local treatment, with CD45RA being the molecule expressed by naïve T lymphocytes [15,16]. In another study, the number of infiltrating CD45RO+ memory T cells was enlarged following RF-assisted liver resection, with CD45RO being the marker for activated memory T cells. The higher number of these cells was also associated with a lower recurrence rate in patients [11]. Another study focused on glypican-3, a carcinoembryonic antigen that is highly expressed in HCC. Interestingly, the number of glypican-3-specific cytotoxic T lymphocytes (CTLs) was also increased in circulation after RFA [17].
Regulatory T cells are predominant within tumour tissues and they contribute the immunosuppressive character of the HCC microenvironment [18]. A high number of circulating or intratumoural regulatory T cells is associated with poor survival rates in HCC patients. Moreover, following RF-assisted liver resection, the frequency of circulating regulatory T cells decreased, leading to an increased CD8+/CD4+, further demonstrating the stimulation of antitumour immunity induced by the RF technique. In addition, the expression of immunosuppressive cytokines such as interleukin (IL)-10 and transforming growth factor (TGF)-β, classically secreted in the HCC tumour microenvironment, seems to decrease after RFA as well. On the contrary, levels of interferon (IFN)-γ, an important proinflammatory molecule stimulating antitumour immunity, were higher after RFA [11]. Similarly, inflammatory cytokines IL-1β, IL-6, IL-8 and tumour necrosis factor (TNF)-α are expressed more following RFA, inducing a stronger immune response [5,19]. A recent study reported that in HCC, the innate cells are, perhaps, the main actors of RFA immune dynamics and that the setup of the immunomodulation occurs very early. Indeed, early after RFA, DCs and natural killer (NK) cells are recruited at the tumour site to potentialize the antitumour response. NK cells can lyse target cells based on the balance between activating and inhibitory receptors. Shortly after RFA, a high number of NK cells expressing the activation receptor NKp30 is associated with a lower tumour recurrence rate. However, when the proportion of NKp30+ NK cells is still high one month after ablation, it participates in a chronic local inflammation and seems to increase the risk or tumour recurrence. This is in correlation to the level of DCs as they interact with NK cells. In this study, NKp30+ NK cells and plasmacytoid DCs decreased in the peripheral blood of all patients one day after RFA, and then, increased one month after treatment to a different extent for each of them. Thus, systemic changes induced by percutaneous RFA in innate immunity may constitute predictive markers of recurrence after RFA and should be further studied in randomised trials [16].

Immune Changes Induced by MWA
Fewer studies have been performed to investigate the immunomodulatory effects of MWA compared to RFA, most likely because MWA is a less frequent and more recent ablation technique. As in RFA, cellular debris are released following heat application, such as HSP-70. However, Ahmad et al. described lower serum levels of these proteins following liver MWA compared to RFA in the rat model [20]. Besides, the proinflammatory cytokines IL-1β and IL-6 were also increased in circulation after MWA, corresponding to the induction of the inflammatory response. Although no significant difference was found for the level of circulating IL-1β between RFA and MWA, the increase in IL-6 was substantially smaller in the blood of rats treated with MWA compared to RFA. These results suggest that MWA may trigger a lower inflammation than RFA in rats.
MWA also promotes the infiltration of immune cells into the liver tissue, according to a study in which liver biopsies were performed on patients before and after MWA. Among infiltrating intrahepatic immune cells, authors observed mainly NK cells and macrophages as well as a high abundance of T cells [21]. Similarly, the effects of MWA on various subsets of circulating immune cells and cytokines in patients with HCC were analysed. Indeed, MWA had no impact on the frequency of regulatory T cells and NK cells in circulation. Besides, both circulating CD3+ and CD4+ cells were increased, with a higher ratio of CD4+/CD3+ after MWA. Moreover, the shift in the differentiation of helper T cells into the two main subtypes, Th1 and Th2, has been observed. Yet, IFN-γ, TNF-α, IL-2 and IL-12 are cytokines secreted mainly by Th1 cells that can promote the cytotoxic effects of CTLs. The changes to the IFN-γ, TNF-α and IL-2 levels were not statistically significant, but there was a great increase in circulating IL-12 after MWA. In addition, IL-4 and IL-10 decreased as well, following MWA. To conclude, the broad analysis of circulating cytokines demonstrated that the production of IL-12, a Th1 cytokine, is enhanced after MWA whereas the secretion of Th2 cytokines IL-4 and IL-10 is inhibited, resulting in a positive antitumour response [22].

Immune Changes Induced by Cryoablation
Contrary to heat-based ablative modalities, cryoablation is associated with both necrosis and apoptosis, which induces simultaneously an immunostimulatory and an immunosuppressive effect, as recently reviewed elsewhere [23][24][25]. As a result, cryoablation might be a less effective approach than RFA and MWA to enhance systemic antitumour immunity. Manipulating the highly variable cryoablation-induced immune response to promote a more cytotoxic immune response would be extremely beneficial for anticancer therapy. At the moment, very few studies, focused on immune changes induced by cryoablation in HCC patients, exist.
One study demonstrated that the percutaneous thermal ablations of solid tumours result in increased plasma levels of IL-6 and IL-10 in the first two days after procedure and interestingly, the cryoablation stimulated a greater change in IL-6 levels than other ablation techniques [26]. A similar study focused on the modulations of serum cytokine levels after hepatic cryoablation of metastatic tumours [27]. Authors divided a small cohort of patients into an immune reaction group, in which antitumour effect was identified not only in the treated area but also away from the area, and to a local effect group. C-reactive protein and IL-6 levels were increased in both groups after cryoablation. TNF-alpha and Th1/Th2 ratio was increased more in the immune reaction group compared to that in the local effect group. Interestingly, IL-10 levels before treatment were significantly higher in patients with the local effect compared to the immune reaction group. Zhou et al. demonstrated that the frequency of circulating regulatory T cells decreases in patients with HCC regression and dramatically increased in patients with HCC recurrence or progression following cryoablation [28]. Moreover, the numbers of CD8+, CD4+, and FoxP3+ cells infiltrating the tumours around the cryotherapeutic zones were decreased after cryoablation.

Immune Changes Induced by Other Ablation Techniques
Another ablation technique recently introduced in the HCC field is irreversible electroporation. Irreversible electroporation does not rely on thermal modulation, but uses high-current electrical pulses to create nanopores in cell membrane, leading to cell death, mainly by apoptosis [3]. This technique is relatively new and data about possible immunomodulation in HCC patients are scarce. A recent study presented the first evidence of irreversible electroporation-based immunomodulatory in patients with locally advanced pancreatic cancer. Authors observed transitory modifications of circulating CD4+ T cell, CD8+ T cell, NK cell, regulatory T cells, IL-6, IL-2, and IL10 after ablation. Moreover, the alteration of CD8+ T cell between D3 and D7 was identified as a prognostic factor for both progression-free survival and overall survival. Interestingly, even though irreversible electroporation is mainly associated with apoptosis, this ablation procedure is still capable of inducing an intense inflammatory cell response, characterized by robust infiltration of macrophages and T cells as observed in a rodent model of pancreatic cancer [29].

Immunotherapy in HCC
Immune checkpoint inhibitors constitute the most promising treatment for HCC in the future, especially PD-1 and PD-L1 inhibitors [1,2]. The goal of this immune checkpoint is to prevent the overstimulation of immune response. PD-1 receptor is expressed on many immune cells such as T cells and especially CD8+ T cells, while its ligand can be found mainly on cancer cells' and antigen-presenting cells' surfaces. When PD-1 is engaged with its ligand, effector T cell-mediated immune responses are impaired. Antibodies blocking PD-1 have been developed including nivolumab and pembrolizumab, while durvalumab, avelumab and atezolizumab target PD-L1. Cytotoxic T lymphocyte-associated protein 4 (CTLA-4) is another immune checkpoint expressed on regulatory T cells and activated T cells. This receptor inhibits T cell activation by recognising the ligand B7 on DCs with a higher affinity than the co-stimulatory molecule receptor CD28. Tremelimumab and ipilimumab are antibodies against CTLA-4. T cell immunoglobulin and mucin-domain containing-3 (TIM-3), lymphocyte activation gene 3 (LAG-3), T cell immunoreceptor with Ig and ITIM domains (TIGIT) or V-domain Ig Suppressor of T cell Activation (VISTA) are other immune checkpoints, with ongoing clinical trials to demonstrate their clinical efficacy and outcomes in patients with HCC [2,30]. Co-expression of multiple immune checkpoints has been associated with a severely exhausted T cell state, typical for CD8+ T cells in the tumour microenvironment. Blocking of one immune checkpoint pathway-for example, PD-1/PD-L1-often results in compensatory upregulation of additional immune checkpoint molecules, which limits the efficacy of monotherapy approaches [31,32]. In order to enhance antitumour efficacy and clinical success, future therapeutic strategies require combinations between immune checkpoint inhibitors or between checkpoint inhibitors and other therapeutic strategies. The best example today is the combination of PD-L1 and vascular endothelial growth factor (VEGF) antibodies, particularly the combination of atezolizumab and bevacizumab, which represents a significant breakthrough in the first-line treatment of unresectable HCC [2,33,34].
Another way of improving antitumour immunity is to inject cytokine-induced killer cells that have been cultivated ex vivo to increase cytotoxic effects against tumour cells and inhibit tumour progression. Transfer of cultured DCs isolated from the peripheral blood to enhance the immune response have also shown beneficial outcomes [14].
Finally, other agents such as oncolytic viruses can be injected at the tumour site to infect and kill cancer cells with interesting preliminary results on survival, even though studies of this approach in HCC have been limited so far [30].

Combination of Ablation Techniques with Immunotherapy in HCC
All the observations mentioned above show that thermal ablation techniques can induce immunomodulation, as summarised in Table 1. However, this effect is insufficient to prevent tumour recurrence in patients. These studies point out the advantages of combining RFA or MWA with immunotherapy to boost the antitumour immune response and thereby prevent recurrence following ablation and improve beneficial outcomes [35]. The combination of thermal ablation therapies with cellular immunotherapy has raised interest due to its potential to improve the strength of tumour-specific immunity. Cui et al. used the combination of RFA and cellular immunotherapy based on immune cells, comprising cytotoxic T cells, NKT, NK and γδT cells that were harvested in the form of peripheral blood mononuclear cells (PBMCs) from patients with HCC before RFA, expanded and then, injected back intravenously after RFA. Authors demonstrated that the risk of tumour recurrence was significantly reduced compared to patients treated with RFA alone [36]. Recently, the therapeutic efficacy of irreversible electroporation in combination with the intratumoural injection of the immunogenic adjuvant was successfully tested in an animal model of HCC [37].
The combination of MWA and cellular immunotherapy was assessed in a phase I clinical trial, with immune cells collected from the peripheral blood of HCC patients. Immature DCs and effector cells were injected into the marginal area of ablated tumours under contrast-enhanced sonographic guidance. The first observations were encouraging, as a depletion of regulatory T cells occurred and was associated with an increase in effector CD8+ CD28-T cells one month after the RFA treatment, even though this difference was no longer significant after six months [38]. Similarly, adoptive transfer of co-cultured DCs and effector cells was tested in combination with cryoablation in patients with metastatic HCC. Patients who received cryoablation with cellular immunotherapy exhibited improved survival outcomes in comparison to patients who received only single treatment [39].
A more recent promising approach for the treatment of HCC was the combination of heat-based ablation therapies with immune checkpoint inhibitors. The use of a CTLA-4 inhibitor, tremelimumab, with RFA was proved as safe and without dose-limiting toxicity. Tumour biopsies revealed that PD-1 expression was increased on the T cell surfaces and a higher number of tumour infiltrating T cells was observed following the combination treatment, especially CD8+ T cells [40]. On the contrary, incomplete percutaneous ablation may worsen the prognosis and induce immunotherapy resistance, as reported in a mouse model of metastatic colorectal cancer [41].
Nonetheless, many questions remain unanswered as to which ablation technique is the most efficient in inducing immunomodulation, what is the best time to start immunotherapy and which type of immunotherapy is the most promising for this strategy. In addition, local ablation can be used as a treatment during the waiting time prior to liver transplantation. In such cases, the combination with immunotherapy may not be recommended due to fear of possible future liver rejection. Moreover, it should be noted that some patients may experience serious, immunotherapy-related adverse events and a careful selection of patients for combination of ablation techniques with immunotherapy will be therefore required.
Today, several ongoing clinical studies are evaluating the combination of ablation techniques and immunotherapy in HCC patients, or the use of, as summarised in Table 2. The efficacy and safety results are eagerly awaited.

Discussion
It has been shown that thermal ablation therapies enable the establishment of a specific antitumour immune response by inducing the mobilisation of antigen-presenting cells and immune effector cells.
It is important to mention that most of the studies are based on the immunomodulatory effects of thermal ablation techniques on the circulating level and not directly on the liver environment. Information about intrahepatic modulations are mostly based on animal models, and the immune system as well as the microenvironment of induced tumours in rodents may not be comparable to the human HCC [42].
Regarding the combination of ablation and immune-based therapies, studies are still lacking. Some features remain unclear and have to be further studied, such as the optimal timing to administer immunotherapy with regards to ablation or the selection of patients who are the best candidates for the combination treatment. Additionally, to date, no studies have focused directly on the effects of thermal ablation therapies on different immune checkpoint pathways. Therefore, at this moment, it is difficult to select the most promising immune checkpoint pathway to target in the combination strategy.
Nevertheless, it seems that the combination of ablation techniques with immunotherapy can boost the induced immune response following each of these modalities, which can lead to the decrease in recurrence rate and improve survival rates in HCC patients. Indeed, two therapeutic strategies should be explored in the future: i) ablation combined with immunotherapy for focal lesion to prevent the recurrence of HCC and ii) ablation of several lesions in advanced HCC combined with immunotherapy to boost the efficiency of immunotherapy.
To conclude, this type of combination strategy is complex and necessarily brings together experts from hepatology, cancer immunology and interventional radiology with the ultimate goal of improving the outcome of patients with HCC.
Funding: This research was funded by Dotation Agir pour les Maladies Chroniques 2019, France.

Conflicts of Interest:
The authors declare no conflict of interest.