Immunological effect of irreversible electroporation on solid tumors

This study intends to investigate the immunological effects of tumor ablation with irreversible electroporation (IRE). Methods We evaluated the systemic immune response in patients with hepatocellular carcinoma (HCC) after IRE treatment. Furthermore, we analyzed the tumor inltrating T lymphocytes and the level of serum cytokines in IRE and control groups of tumor-bearing mice. of a until post-IRE, a increase of activated T cells; 1 Treg cells appear be recovering, increased signicantly. Above indicated the immunomodulatory effect of IRE with decreased immunosuppressive Treg cells and expansion of effective T cells.


Abstract Background
This study intends to investigate the immunological effects of tumor ablation with irreversible electroporation (IRE).

Methods
We evaluated the systemic immune response in patients with hepatocellular carcinoma (HCC) after IRE treatment. Furthermore, we analyzed the tumor in ltrating T lymphocytes and the level of serum cytokines in IRE and control groups of tumor-bearing mice.

Results
We observed that IRE induced an increase in WBC, neutrophil and monocyte counts and a decrease in lymphocyte count 1 day post-IRE and returned to baseline values within 7 days in the patients.
Meanwhile, circulating CD4 + T cell subsets, but not CD8 + , decreased 1 day post-IRE. The activated T cells and natural killer (NK) cells increased, and regulatory T (Treg) cells decreased. Furthermore, a signi cant increase in cytotoxic CD8 + T cells in ltration was observed on ablative tumors in mice. The level of serum IFN-γ also signi cantly increased in the IRE group.

Conclusions
Our study demonstrated that IRE not only induced immediate innate immune response dominated by the increase of neutrophils, monocytes and NK cells, but also upregulated activated T cells and downregulated Treg. Meanwhile, the results from the animal model indicated that IRE could induce antitumor adaptive immunity dominated by cytotoxic CD8 + T cells.
Background Image-guided tumor ablation therapies such as radiofrequency ablation, microwave ablation and cryoablation are now in widespread clinical use to treat a broad range of benign or malignant solid tumors 1,2 . Most of these therapies rely on thermal energy to destroy the tumor tissues by inducing coagulation necrosis. Irreversible electroporation (IRE) is a new non-thermal tumor ablation technique that involves the application of high voltage electrical pulses across the target tissue causing the formation of permanent nanopores in the cell membranes 3 . Unlike conventional thermal ablation modalities, IRE induces cell death, sparing the connective tissue scaffolds, therefore, vital structures such as blood vessels or bile ducts are preserved. Additionally, its e cacy is not impaired by heat sink effects in the treatment of tumors located close to large blood vessels 4,5 . These advantages make IRE suitable to target tumors that cannot be treated by thermal ablations.
The accumulating evidence from the literature suggest that the host immune response is involved in cancer development and progression; activating antitumor immune response plays a crucial role in cancer control and therapy [6][7][8] . Since increased permeability of the cell membrane is thought to be the primary mechanism of cell death caused by IRE 9 , the substantial native tumor antigens may be exposed, allowing for them to act as in situ vaccines to generate antitumor immune reaction 10 . Meanwhile, intact and persistent microvessels within the IRE ablation zone may facilitate the in ltration of immune cells 11 . While a few studies had reported that IRE did not induce any change in immune cell in ltration, other contradictory reports suggest that IRE provides bene cial immunological effects [12][13][14][15][16] . Although the literature on immune cell recruitment after IRE treatment remain con icting, there is strong evidence of local and systemic antitumor immune response resulting from IRE.
Based on the contradictory results in these preclinical animal studies, we aimed to further investigate the immunological response to tumor ablation with IRE in patients with solid tumors and mouse models bearing tumors.

Patients
The Ethics Committee of Ruijin Hospital A liated to Shanghai Jiao Tong University School of Medicine approved this retrospective study (reference number: AF-0406). Written informed consent was obtained from each patient. From October 2016 to August 2019, 61 patients in our center underwent IRE treatment (including 26 liver tumors, 17 pancreatic tumors, 4 renal tumors, 5 adrenal gland tumors, 6 retroperitoneal tumors; one patient with both liver and pancreatic tumors, and two patients with both liver and retroperitoneal tumors). Due to the immunity of these types of patients were relative to the types of malignancies that were treated were highly variable, we selected a single tumor type---hepatocellular carcinoma (HCC) for immunological analysis. A total of 11 patients with HCC (8 men, 3 women, mean age, 60.8 ± 9.3 years) were enrolled in the study. All patients had a history of chronic HBV infection and cirrhosis. No chemotherapy or interferon therapy was received previously. After the diagnosis of HCC, all patients were treated with oral nucleoside antiviral drugs. During the follow up, all patients received no adjunctive therapies.

IRE procedure in patients and sample collection
Two experienced interventional radiologists performed all the procedures. All patients were administered with muscle relaxants and general anesthesia. IRE was performed using a NanoKnife system (AngioDynamics, Latham, NY, USA) with an electrocardiogram (ECG) synchronization device under the guidance of CT. Nineteen gauge monopolar needles were placed in parallel around the tumors percutaneously at the intervals of 1.2-2.2 cm. Tip exposure of the needles was 1.0-2.0 cm. The number of needles was decided according to the tumor size. The parameters of IRE ablation were set as follows: average electric eld intensity, 1500 V/cm; pulse length, 70-90 μs; 90 pulses. The ablation range covered the whole tumor with an ablation margin of at least 5 mm. Peripheral blood samples were collected 1 day before IRE therapy and used as baseline values. Additional blood samples were collected 1 day, 3 days, 7 days, 2 weeks, and 4 weeks after IRE treatment. Blood tests included blood cells analysis and immune cells analysis. A routine clinical ow cytometry test protocol was followed for analyzing immune cells in the peripheral blood.
Cell culture and animal models All animal experiments were conducted by the Guidelines for the Care and Use of Laboratory Animals of Shanghai Jiao Tong University School of Medicine. The mouse hepatic carcinoma cell lines, H22 were purchased from China Center for Type Culture Collection (Wuhan, China). H22 cells were cultured in RPMI containing 10% FBS and 1% penicillin-streptomycin, in an incubator with a humidi ed atmosphere of 5% CO 2 at a temperature of 37°C. H22 cells (5 × 10 6 ) were suspended in 200 μL phosphate buffered saline and injected subcutaneously into the right ank of 5-to 6-week-old male BALB/c mice (commercially obtained from LINGCHANG BIOTECH, Shanghai, China). Two weeks after the injection, the diameter of the tumors reached nearly 1 cm.

IRE procedure in mice and sample collection
After two weeks of modeling, a total of 30 mice were randomly divided into two groups: the control group (n = 15) and the IRE group (n = 15). For the IRE group, the mice were anesthetized by injecting sodium pentobarbital (10 mg/mL, 50 mg/kg body weight) intraperitoneally. Then, each mouse was xed on an insulating plate, and the IRE procedure was performed using an ECM 830 Square Wave Electroporation system (BTX Harvard Apparatus, Holliston, MA, USA) with a pair of genetrodes (BTX item #45-0161, BTX Harvard Apparatus, Holliston, MA, USA). The genetrodes with a 10 mm gap were inserted into the tumors to deliver electric pulses with the following parameters: voltage, 1200 V; pulse length, 90 μs; 90 pulses.
This protocol was selected to produce a complete ablation for the tumors. The mice in the control group received sham procedures with the genetrodes inserted into the tumors but no electric pulses were given.
The blood samples and tumor samples were collected at 3, 7, and 14 days post-IRE procedure from ve mice separately. Samples from the control group were collected at the same time. At the end of the experiments, the mice were killed by standard CO 2 asphyxiation.

Lactate dehydrogenase (LDH) cytotoxicity assay
Single-cell suspensions from mice tumors were prepared. T cells were isolated from the single-cell suspensions by negative selection using the Pan T Cell Isolation Kit II (Miltenyi Biotec, Bergisch Gladbach, Germany). The isolated T cells were stimulated with recombinant murine IL-2 (PeproTech, Rocky Hill, NJ, USA) for 3 days. Then, the T cells were added to H22 cells with an effector to target cell ratio of 20:1, and the cells were co-cultured in 96-well plates for 24 h. Evaluation of T cell cytotoxicity activity was performed using an LDH-Cytox TM Assay kit (BioLegend, San Diego, CA, USA), according to the manufacturer's protocol. The cytotoxicity percentage was calculated as follows: (LDH experimental − LDH spontaneous) / (LDH maximum − LDH spontaneous) × 100%. LDH experimental represents the LDH release activity from the T cells and tumor cells co-culture. Spontaneous LDH release activity was obtained from tumor cells cultured separately. The maximal LDH release activity was obtained following lysis of the tumor cells.

Immunohistochemistry analysis
The tumor tissue removed from each mouse was xed in 4% paraformaldehyde and embedded in para n for 5 μm-thick sections. After being depara nized and rehydrated, the sections were treated with sodium citrate buffer (pH = 6), and the microwave was used for antigen retrieval. The activity of endogenous peroxidase was blocked with 3% H 2 O 2 in methanol. The sections were then incubated with anti-CD3 monoclonal antibody (Abcam, Cambridge, UK), anti-CD4 monoclonal antibody (Abcam, Cambridge, UK), and anti-CD8 monoclonal antibody (Abcam, Cambridge, UK) at 4°C overnight, respectively. Afterwards, the sections were stained with HPR-conjugated secondary antibody, and the positive reactions were visualized with diaminobenzidine (DAB). Finally, the sections were counterstained with Mayer's hematoxylin. Digital images of the stained sections were obtained in ve randomly selected elds both at the interior regions and the margin of tumors using a uorescence microscope. The positive cell numbers were counted and the results from the ve areas were averaged and used in the statistical analysis.
Cytometric bead array (CBA) analysis Blood serum was separated from the blood sample obtained from the tumor-bearing mouse by centrifugation at 3000 g for 20 min and then stored at −80°C, until later analysis. Serum cytokine analysis was performed using the CBA Flex Set (BD Biosciences, Franklin Lakes, NJ, USA), containing mouse IFNγ, IL-1β, IL-2, IL-10, and TNF-α. Mouse Soluble Protein Flex Set Standards and samples were prepared according to the manufacturer's instruction. The samples were acquired on the ow cytometer (BD LSRFortessaTM X-20, Franklin Lakes, NJ, USA). The data were analyzed using FCAP Array software (BD Biosciences, Franklin Lakes, NJ, USA).

RNA sequencing analysis
Tumors from IRE and control groups 7 days postoperative were separated for RNA sequencing analysis. Total RNA was extracted using Trizol reagent (Roche, Basel, Switzerland). Total RNA quality was evaluated on an Agilent 2100 Bioanalyzer (Agilent, Santa Clara, CA, USA). Library preparation was performed from the pooled RNA using an Illumina TruSeq RNA Sample Preparation Kit v2 (Illumina, San Diego, CA, USA) and sequenced on the Illumina HiSeq 4000 platform (Illumina, San Diego, CA, USA). The sequenced reads were aligned to the mouse genome mm10 by HISAT2 17 . FeatureCounts was used to quantitate the transcriptome, using the GTF annotation les 18 . Differential analyses were performed to the count les using DESeq2 packages, following standard normalization procedures 19 . The differentially expressed genes (DEGs) were identi ed with p values < 0.05 and absolute log2 fold change > 1. Gene Ontology (GO) enrichment analysis was performed using Metascape (http://metascape.org).

Statistical analysis
Statistical analysis was performed using SPSS statistical software, version 23 (IBM, Armonk, NY, USA). The immunohistochemistry results were analyzed with the Mann-Whitney test. For other analysis, Student's t-test was used. All data were expressed as mean ± SEM (standard error of the mean) of n independent measurements. GraphPad Prism 7 software (GraphPad, San Diego, CA, USA) was used to plot graphs. P < 0.05 was considered statistically signi cant.

Results
An immediate innate immune response was observed after IRE Peripheral blood samples obtained from 11 patients with HCC at six different time points, pre-and post-IRE ablation, were tested for systemic immune reaction. However, complete follow-up information was not obtained from some patients. We applied scatter plots to show the detailed information collected. As shown in Fig. 1, WBC, neutrophil and monocyte counts, as well as neutrophil to lymphocyte ratio (NLR) were found to be elevated signi cantly 1 day post-IRE; however, they returned to baseline values gradually within 7 days after IRE. On the contrary, lymphocyte count declined 1 day post-IRE, which then increased gradually. Meanwhile, IRE led to a slight and non-signi cant increase on the percentage of natural killer (NK) cells (CD56 + CD16 + ) on day 1 (11.4 ± 8.5 vs 15.9 ± 9.7%), followed by a decrease on day 3, then a signi cant increase was observed from day 3 to day 14 (10.4 ± 6.4 vs 15.3 ± 8.2%, p < 0.05) (Fig. 1F).

Cytotoxic CD8 + T cells increased but Treg decreased in mice tumor after IRE
To probe whether the IRE treatment resulted in signi cant up-regulation of pathways associated with the adaptive immune process, RNA-seq expression was performed in IRE treated mice and controls. The results of GO analysis indicated that several adaptive immune process related pathways were upregulated in the IRE treatment group, including antigen processing and presentation of exogenous peptide antigen, antigen processing-cross presentation, and adaptive immune response (Fig. 3A, B). Additionally, several pathways, such as signaling by TGF-beta Receptor Complex were down-regulated in IRE treatment group (Fig. 3C, D).
Gene expression pro ling also revealed a signi cant increase in T cell mediated cytotoxicity associated genes (Gzmb) in the tumor post-IRE treatment (Fig. 5A), which was consistent with the results of ow cytometry analysis (Fig. 4F, G). To further con rm whether IRE treatment induced speci c antitumor immunity, we analyzed cytotoxic activity of T lymphocytes. Antitumor cytotoxicity activity of T cells treated with IRE was higher than that of the control group (97.8 ± 1.2 vs 66.0 ± 3.9%, p < 0.001) on day 3 (Fig. 5B). All above results con rmed speci c antitumor immunity induced by IRE.

CD8 + T cells in ltrated predominantly at margins but CD4 + T cells also in ltrated into the mice tumor after IRE
The immunohistochemistry results revealed that while the in ltration of CD3 + (Fig. 6A, B) and CD4 + T lymphocytes (Fig. 6C, D) was intense at the margin, they were also found in the center of tumors. In the IRE group, the number of CD3 + T lymphocytes signi cantly higher than the control group on day 3 (3.1 ± 2.5 vs 2.1 ± 1.0, p < 0.01) and day 14 (6.9 ± 6.0 vs 3.0 ± 1.8, p < 0.001) in the center of tumors, and CD4 + T lymphocytes on day 7 (1.9 ± 1.0 vs 0.9 ± 0.4, p < 0.001) and day 14 (1.8 ± 1.2 vs 1.0 ± 0.6, p < 0.05).
However, we only found in ltration of CD8 + T lymphocytes (Fig. 6E, F) at the margins, but scarcely in the center. Although it did not reach a statistically signi cant level, we did nd that in IRE group, there were higher numbers of in ltrating CD8 + T lymphocytes in the margin of tumors at different time points compared to the control group (10.3 ± 5.7 vs 4.9 ± 3.9; 7.6 ± 6.0 vs 6.8 ± 6.3; 10.0 ± 7.0 vs 8.0 ± 6.4, respectively).
IFN-γ increased in the serum of tumor bearing mice after IRE Furthermore, we observed a signi cant increase in the expression pro les associated with cytokine production, predominantly IFN-γ-mediated signaling pathway. Tumor-bearing mice in the IRE group also showed signi cantly higher serum level of IFN-γ (Fig. 7A) on both day 7 and 14, in comparison with the control group (2.0 ± 0.7 vs 0.8 ± 0.7, p < 0.05; 1.1 ± 0.6 vs 0.1 ± 0.1, p < 0.05, respectively). Although not statistically signi cant, we found that the serum levels of IL-2, TNF-α, and IL-1β in the mice treated with IRE were higher than the levels in the control group ( Fig. 7B-D). The elevated levels of these would help increase anti-tumor immunity in the body. In addition, the IL-10 level in the IRE group decreased at all time points after IRE treatment (Fig. 7E).

Discussion
IRE is a novel, non-thermal ablation modality, which was used to treat tumors. Unlike with surgical resection, the ablated tumor tissues are not removed, apoptotic or necrotic cells release damageassociated molecular patterns, such as HMGB1, ATP, ROS and calreticulin, which may serve to achieve in situ tumor vaccination 20 . As a result, antitumor immune response may be induced that in turn can effectuate regression of distant and metastatic lesions, increase the e cacy of IRE. In this study, we investigated the immunological response post-ablation of IRE in patients with HCC, as well as in mice bearing tumors, to provide more evidence for clinical tumor treatment.
First, we tested whether IRE induced a systemic immunological response in patients with HCC. Shortly after IRE treatment, an increase in WBC, neutrophil and monocyte counts were observed in the peripheral blood. Neutrophils are the predominant circulating leukocyte population and constitute an important part of the innate immunity. Several studies have demonstrated various antitumor effects of neutrophils, including direct cytotoxicity and antibody-dependent cell-mediated cytotoxicity (ADCC) 21,22 . Recent reports also suggest neutrophils display antimetastatic properties, which were mostly attributed to circulating neutrophils 23,24 . Moreover, neutrophils could secrete cytokines and chemokines activating other immune cells 25 . Bulvik et al. compared the effects of IRE with those of RFA and found persistent patency of vascular within the ablation area from IRE increased the in ltration of immune cells and induced more robust systemic effects 11 . In a more recent report Sugimoto et al. found a signi cant early increase in macrophage migration inhibitory factor, followed by rapidly mobilizing of monocytes from the peripheral blood to the ablation zone after IRE, but not RFA, which may facilitate the early reparative process and result in shrinkage of ablation zone 26 . As another key effector of the rst-line defense against tumors, NK cells were also found to be elevated post-IRE. Similar to neutrophils, NK cells kill tumor cells through the cytotoxic activity without any speci c antigen stimulation; they can also regulate innate and adaptive immunity by secreting cytokines and chemokines 27 . The major cytokine produced by NK cells is IFN-γ, which mediates the induction of T helper 1 (Th1) cells, which are associated with a good prognosis of patients with cancer 28 . Indeed, we did found that IFN-γ increased in the serum of mice bearing tumors after IRE. Our results showed that IRE indeed could induce an immediate innate immune response characterized by the increase of neutrophils, monocytes and NK in patients, which may help the patients to reconstruct anti-tumor immunity.
Increasing evidence shows that elevated NLR is a prognostic indicator of mortality 29,30 . In our patient study, although NLR was found to be elevated signi cantly 1 day post-IRE, it recovered to the base level within 1 week. This transient elevation may represent a relative lymphocytopenia, capacity of NLR to decline for a short period, re ecting host immunomodulatory activity. Indeed, we observed that IRE caused short-term depletion of circulating CD4 + T lymphocytes (activated and memory subsets), but not CD8 + T cells. Lymphopenia has been explored in patients with cancer undergoing chemoradiotherapy 31,32 . Some reports have shown that immune reconstruction following lymphopenia shifts T subsets toward a predominance of activated T cells, enhancing antitumor immunity 33  treatment, and a signi cant decrease was also found 37 . With a longer follow-up time in our study, similar ndings showed a transitory increase of Tregs by 3 days followed by a decrease until two weeks post-IRE, accompanied by a remarkable increase of activated T cells; after 1 month Treg cells appear to be recovering, which increased signi cantly. Above results indicated the immunomodulatory effect of IRE with decreased immunosuppressive Treg cells and expansion of effective T cells.
Apart from systemic antitumor immunity, tumor-in ltrating lymphocytes are associated with favorable prognostic effect 38,39 . Especially, the introduction of immunoscore has gained signi cance for the classi cation of cancers and aid in predicting the outcomes of treatments 40,41 . Our tumor model study revealed that IRE treatment induced adaptive antitumor immunity within the ablative tumors, dominated by increase in cytotoxic CD8 + T cells and reduced Treg cells. It is well known that cytotoxic CD8 + T lymphocytes are crucial components of tumor-speci c cellular adaptive immunity, and they produce perforin, granzyme, or TNF, IFN-γ to kill tumor cells or induce apoptosis 42,43 . Consistently, serum cytokines, predominantly IFN-γ, were found to be increased post-IRE. However, CD8 + T cells in ltrated merely the margin of tumors in both IRE and control groups, which is because immune cells already inside the tumor are probably destroyed by IRE pulses, whereas, CD4 + T cells in ltrated both in the center and margin of the same tumor. This difference could be contributed by regional heterogeneity of tumor architecture and tumor antigens. The combined data of our systemic and local antitumor immunity induced by IRE proposed an ideal treatment window for immunotherapy by increasing the effective T cells and decreasing Tregs, resulting in control recurrence and metastasis of ablation therapy. Recent literature reported that IRE combined with immune checkpoint blockade enhanced antitumor immune response, and help overcome the immunosuppressive tumor microenvironment of pancreatic cancer 44,45 . Therefore, IRE may be an effective modality to overcome the immunosuppressive "cold" tumor microenvironment. More studies are needed in the future to evaluate this role in other tumors such as HCC.

Conclusions
In conclusion, in the current study, we preliminarily investigated the immune activity caused by tumor ablation with IRE in both patients with HCC and tumor-bearing mice. The results demonstrated that IRE not only induced immediate innate immune response dominated by the increase of neutrophils, monocytes and NK but also upregulated activated T cells and downregulated Treg, which are a bene t for enhancing the sustained anti-tumor activity of patients. Meanwhile, the results from the animal model indicated that IRE could rapidly inhibit local tumor growth by inducing the in ltration of CD4 + and CD8 + T cells, predominantly cytotoxic CD8 + T cells. However, further investigations are needed to ascertain the long-term antitumor immunity and to evaluate how these immune responses impact the prognosis of patients with various cancer phenotypes. Moreover, the immune mechanism induced by IRE needs more in-depth studies. Furthermore, an optimal combination therapy requires validation in animal and patients, such as IRE combined with immune checkpoint inhibitors, which may improve the prognosis of patients.

Consent for publication
Not applicable.

Availability of data and materials
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.

Funding
This work was funded by Shanghai Municipal Key Clinical Specialty (No. shslczdzk06002). The funding body had no role in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript.
Author's contributions X.Y.D. and Z.Y.W. designed the study, X.X.G., F.D. and Y.G. performed animal experiments, Q.L. and Q.B.W. planned data collection from patients, W.H., Z.M.W., X.Y.D. and Z.Y.W. performed IRE on patients, X.X.G. drafted the paper, and Z.Y.W. revised the paper.