Plasma-activated medium triggers cell death and the presentation of immune activating danger signals in melanoma and pancreatic cancer cells

Over the past decade, cold atmospheric plasmas have shown promising application in cancer therapy. The therapeutic use of plasma-activated media is a topic addressed in an emerging field known as plasma pharmacy. In oncology, plasma-activated media are used to harness the therapeutic effects of oxidant species when they come in contact with cancer cells. Among several factors that contribute to the anticancer effect of plasma-activated liquid media (PALM), H2O2 and NO derivatives likely play a key role in the apoptotic pathway. Despite the significant amount of literature produced in recent years, a full understanding of the mechanisms by which PALM exert their activity against cancer cells is limited. In this paper, a sealed dielectric-barrier discharge was used to disentangle the effect of reactive nitrogen species (RNS) from that of reactive oxygen species (ROS) on cancer cells. Two cancers characterized by poor prognosis have been investigated: metastatic melanoma and pancreatic cancer. Both tumour models exposed to PALM rich in H2O2 showed a reduction in proliferation and an increase in calreticulin exposure and ATP release, suggesting the potential use of activated media as an inducer of immunogenic cell death via activation of the innate immune system.

damage-associated molecular patterns (DAMPs), which ultimately increases the visibility of the killed cells to the immune system 32,42,43 . After direct CAP exposure, as well as in response to chemotherapeutic agents such as anthracyclines, tumour cells expose complexes with the DAMP calreticulin (CRT) on the extracellular side of their plasma membrane at a pre-apoptotic stage 44 . This feature, which is strictly correlated with immunogenicity, allows the use of dying tumour cells with CRT exposed on the surface as a vaccine therapy against cancer. Another DAMP is the release of ATP from dying cells, which in turn activates macrophages 45 . Miller's research has demonstrated that plasma-delivered ROS and charged species produced in the plasma increase intracellular ROS and induce the exposure of DAMPs 10 . Due to the complex phenomena present at the interface of the plasma and the biological system, it is important to note that all possible in vivo and in vitro applications of plasma medicine, including the triggering of ICD, are mediated by the presence of biological liquids (e. g., blood, wound exudate, cell culture media). Despite the expanse of literature on PALM-assisted treatment of cancer cells for testing proliferation, apoptosis, motility and colony formation, only a few studies have investigated the induction of cellular DAMP emission stimulated by PALM. A recent paper showed an increased levels of extracellular-facing CRT after exposure to PALM-treated medium, which was formed by treating Dulbecco's modified Eagle's medium (DMEM)/F12 medium with a plasma jet (kINPen MED) fed with Ar, in an in vivo model of peritoneal spread of pancreatic cancer 46 . In the approach considered in this study, a DBD is ignited on top of the liquid to promote the diffusion of the active plasma species in the liquid contained in Petri dishes ~57 mm in diameter; due to the containment of the plasma source in the Petri dish, tight control of the active species produced is possible without contamination from external air. With this approach, depending on the gas feed used, the plasma can enrich the liquid with selected RONS derived from the cascade reactions of the plasma species diffused in the liquid phase. The chemistry that occurs in liquid media interfaced with CAP as well as the chemical composition of the gas feed depends on several factors 31,47 : the input power of the plasma; the composition of the substrate solutions; the temporal parameters (e.g., plasma exposure time and lifetime of solution post-preparation); the distance between plasma and liquid surface; the surface area of plasma and the volume of liquid used. Compared to jets, the DBD configuration guarantees homogeneous treatments of larger surface areas (i.e., several centimetres) and greater liquid volumes 48,49 . With this approach, it is possible to properly design plasma treatments and, in principle, to select the RONS to be delivered to the liquid for interfering with cell cycle progression as well as triggering DNA damage, apoptosis, and cell membrane damage 50 .
In this paper, after a preliminary characterization of the antiproliferative activity of PALM, its capability to activate ICD will be investigated in metastatic melanoma (MM) and pancreatic cancer (PDAC) models. Moreover, by using a sealed plasma reactor and by feeding the plasma with O 2 in principle, it should be possible to enrich the liquid with ROS instead of reactive nitrogen species (RNS) and study the effects of the PALM on tumour cells, which will contribute to new insights into plasma-assisted ICD. The reason why the role of specific reactive species has been addressed in this paper is that, as shown before, the chemical composition of the PALM plays a pivotal role in its effects. H 2 O 2 has been identified as a critical player in apoptosis via a caspase-independent apoptotic pathway 51 , and nitric oxide (NO) and RNS species are ubiquitous molecules that play a key role in various physiological and pathological processes, including genotoxic mechanisms, antiapoptotic effects, promotion or inhibition of angiogenesis, limitation of host immune response against tumours, and promotion of metastasis. NO is a short-lived molecule that, when in contact with a complex biological milieu such as a cell culture medium, is rapidly metabolized into nitrite and nitrate after its rapid diffusion through liquid media. This effect limits the biological half-life of NO in vitro to less than a second, whereas the concentrations of NO relevant for cellular signalling can persist in phosphate-buffered saline for an hour 52 . Thus, the role of NO-related compounds in biochemistry should not be underestimated. Quantification of such components can indirectly give an estimation of the amount of NO produced in a DBD. It has been known for at least a decade that the nitrite anion (NO 2 − ) can act therapeutically, most likely as an intracellular precursor source of NO 53 . The present paper investigates the effect of H 2 O 2 and NO 2 − produced in the plasma on the survival of MM and PDAC cells and the role of these molecules in stimulating DAMP secretion from tested tumour cells. By demonstrating the efficacy of PALM in stimulating DAMP emission, the applicability of plasma pharmacy for cancer treatment can be dramatically extended even to internal tumours that are difficult to reach when using topical approaches.

Materials and Methods
experimental set up. Figure 1 shows a schematic diagram of the PetriPlas + source utilized in this research.
This particular DBD setup was designed at the Leibniz Institute for Plasma Science and Technology (INP) in collaboration with co-authors of this paper (E.S., NANOTEC, Bari, ITA). The source consists of aluminium housing with a ground stainless steel grid electrode. The high voltage electrode, 4 mm from the grid, is made of a copper disc (30 mm dia.) covered with quartz. A picture of the discharge ignited between the copper electrode and the grid is shown at the bottom of Fig. 1.
The PetriPlas + source is designed for operation with commercial Petri dishes 57 mm in diameter. A properly designed plexiglass chamber (flow unit) provides gas connectors and sockets and allows the uniform evacuation of the plasma region and the gap between the source and the liquid. The feed gas is ported in the discharge through an orifice and pumped through a second orifice positioned at the opposite side (longitudinal injection). The discharge unit is placed on the Petri dish as shown in Fig. 1, with the dish fitting exactly the small volume between the discharge and the flow units. A silicone gasket is used to seal the flow unit and the upper rim of the Petri dish. In our experiments, 3 ml of culture medium were added to the Petri dish with a liquid/gas interface 2 mm from the ground grid. A dry diaphragm pump (Pfeiffer) is used to keep the working pressure constant, measured by an MKS Baratron. Experiments were carried out at 750 mTorr and room temperature, with a flow rate of 0.5 l·min-1 of either synthetic air (Air Liquide, 99.999%) or O 2 (Air Liquide, 99.999%). The gas flow rates were controlled with MKS mass flow controllers. Before any treatment, the flow unit was purged for 1 min with the feed gas to reduce the presence of impurities in the DBD region. The AC (6 kHz) power supply used in this www.nature.com/scientificreports www.nature.com/scientificreports/ study was connected to a TG1010A programmable 10 MHz DDS TTi function generator, which allows variation in the sinusoidal voltage, amplitude and frequency of the input power for on/off discharge pulsing. Electric parameters were monitored with a Tektronix TDS 2014 C digital oscilloscope and a Tektronix P6915A HV probe. The transported charges (Q) in the discharge space are determined indirectly through a 100 nF capacitor. When the voltage is high enough to cause a breakdown of the gas gap, the voltage across the measuring capacitor is proportional to the transported charges in the discharge space. Therefore, a charge-voltage (Q-U) plot resembling a closed parallelogram, i.e., a Lissajous figure, is obtained. The area of the parallelogram is equal to the energy deposited in the cycle discharge 54 . The plasma was pulsed with a 50% duty cycle, with a time on (ton) of 50 ms over a period (ton + toff) of 100 ms. The dissipated energy is calculated by multiplying the mean energy per cycle by the ton. The 13 kVpp value of the applied voltage was selected to achieve stable operation with the liquid film and uniformly distribute the microdischarges in the interelectrode gap. This value, along with the frequency (6 kHz) of the applied field, was kept constant in all the experiments. The dose energy was calculated by dividing the dissipated discharge energy by the area of the copper electrode. PALM were produced with either synthetic air or O 2 at 1 or 3 min of treatment time. Table 1 lists the different experimental conditions investigated with and without the liquid in the Petri dish. H 2 o 2 detection. A Spectroquant ® Hydrogen Peroxide Test was used for the detection of H 2 O 2 in the cell culture medium. In the presence of a phenanthroline derivative, hydrogen peroxide reduces copper (II) ions to copper (I) ions, which results in the formation of an orange-coloured complex that can be measured photometrically at 445 nm. The samples were analysed within 10 min after plasma processing. nitrite in acid solution reacts with sulphanilic acid to form a diazonium salt, which in turn reacts with N-(1-naphtyl)-ethylenediamine dihydrochloride to form a red-violet azo-dye that can be measured at 525 nm.
Cell culture. The human PDAC cell line PANC-1 was purchased from ATCC ® , Hmel1 MM cells (from human sporadic melanoma biopsy specimens, with BRAF mutations) were established as described by Zanna et al. 55 , and HBL MM cells (BRAF wild type) were a gift from Prof. G. Ghanem, University of Bruxelles, Belgium. All cell lines were grown in high-glucose DMEM supplemented with 10% (v/v) FBS, 1% (v/v) l-glutamine, and 1% (v/v) penicillin/streptomycin in a humidified atmosphere at 37 °C and containing 5% CO 2 . All materials for cell culture were purchased from EuroClone, Italy.  www.nature.com/scientificreports www.nature.com/scientificreports/ Cell imaging. After 5000 cells/well were seeded in 96-well plates and reached 80% confluence, they were incubated with PALM (in O 2 1 min, O 2 3 min or air 1 min conditions) and viewed on an inverted light microscope at 48 h after treatment. Images were acquired with an inverted microscope Olympus CKX41 (10 X magnification).

Cell proliferation assay.
A total of 5000 cells/well were seeded in 96-well plates and cultured to 80% con- statistical analysis. Statistical significance was calculated using a two-way analysis of variance (ANOVA) followed by Bonferroni post hoc tests (GraphPad Prism ver. 5). Data are indicated as follows: *p < 0.05, **p < 0.01, and ***p < 0.001.

Results
electrical characterization of plasma. Typical waveforms of discharge voltage and discharge current for air and O 2 carrier gases for discharge operating with and without liquid are shown in Fig. 2. Microdischarge activity is indicated in Fig. 2A by short peaks on top of the capacitive current, I(A), and observed only when the voltage in the gap reaches the average discharge voltage. Due to the high amplitude of the applied voltage, the formation of microdischarges is favoured 57 . The plasma source was 2 mm from the surface of the liquid, and the bottom of Fig. 1 shows the discharge switched on in synthetic air. Along with the current-voltage waveforms, Fig. 2B,C also shows Lissajous figures (Q-U plots). These plots were used to determine the energy dissipated into the plasma after one cycle of high voltage. The area of the Lissajous figure represents the energy of the electrical discharge in one AC cycle.
Under the investigated experimental conditions, the energy dissipated in the discharge was the same with and without the liquid underneath, as summarized in Table 2. In all conditions, only the treatment time influenced the energy dissipated in the system.  (Fig. 4). This was most evident in Hmel1 cells compared to HBL and PANC-1 cells. The other treatment conditions (O 2 1 min and air 1 min) had a lower impact on morphology and viability in all three cell models. For Hmel1 cells, exposing these cells to PALM O 2 1 min resulted in a slightly lower number of adherent cells with respect to the control cells (untreated cell culture medium) and cells treated with PALM air 1 min. This lead to the hypothesis that ROS exert a greater effect than RNS, and, in order to harness this activity and improve the efficacy of PALM, we decided to conduct all the further in vitro experiments with PALM O 2 3 min.
The MTT assay results reported in the histogram in Fig. 5A as % cell viability compared to control cells confirmed a stronger reduction in cell viability in MM cells than in PDAC cells after a 48-h incubation with PALM O 2 3 min. In all samples, the reduction in cell viability was statistically significant (p < 0.001). These results seem to be correlated with the ROS levels found in the cell lines under investigation before and after treatment with PALM (Fig. 5B). PANC-1 cells showed a higher basal level of ROS than Hmel1 and HBL cells (data not shown). When PALM treatment increased intracellular ROS production compared to that in control cells, the result was a greater cytotoxic effect. In fact, the calculated ratio ROS afterPALM /ROS basal reported in Fig. 5B is greater in Hmel1 cells than in HBL and PANC-1 cells. The difference in the ratio between Hmel1 and PANC-1 cells was significantly different (p < 0.05).   www.nature.com/scientificreports www.nature.com/scientificreports/ To characterize the types of cell death responsible for the anticancer activity of PALM, all cell lines were treated with PALM for 24 h and 48 h and then stained with Annexin V/PI. Flow cyotmetry analysis showed that PANC-1 and Hmel1 cells exposed to PALM for 48 h died because of apoptosis. This phenomenon was markedly evident in the PDAC model, while HBL cells did not undergo this mechanism of cell death. The results are reported in the histogram in Fig. 6A as fold change of apoptosis (early and late) in the treated group vs control group for every cell line after 24 h and 48 h of treatment with PALM.
We also evaluated whether autophagy was involved in the observed cell death. Protein extracts from cells treated with PALM for 72 h were screened for LC3 by western blotting. The detection of LC3 conversion from  www.nature.com/scientificreports www.nature.com/scientificreports/ LC3-I to LC3-II, when correlated with the number of autophagosomes, is an indication of autophagy 58 . The results, shown in Fig. 6B as the ratio of LC3-II vs LC3-I, demonstrated that the activation of autophagy represented by the increase in the LC3-II/LC3-I ratio in the histogram and by the strong LC3-II band in the western blot occurred only in HBL and Hmel1 cells incubated with PALM for 72 h; in particular, LC3-II was 4 and 1.3 times higher than LC3-I in HBL and Hmel1 cells, respectively, after PALM treatment. The optical images of Hmel1 cells in Fig. 6B show autophagic cells only after PALM treatment. In contrast, in PANC-1 cells, autophagy was not observed in either control cells or cells treated with PALM.  The protein bands are reported, and the stain-free method was used to normalize the values. Data from 3 sample replicates are shown (mean ± SD, ***p < 0.001, **p < 0.01, *p < 0.05). The full-length blots are presented in Supplementary Fig. 1. The samples were obtained from the same experiment, and the blots were processed in parallel. Images of Hmel1 cells before and after PALM treatment are shown.
www.nature.com/scientificreports www.nature.com/scientificreports/ These data show that PALM induced a reduction in cell viability through two different types of cell death, apoptosis and autophagy, which could occur simultaneously.

Activation of immunological cell death (ICD) after PALM treatment.
To investigate the activation of ICD in all tumour cell lines treated by PALM, the exposure of the chaperone CRT to the extracellular environment and the release of ATP from dying cells were evaluated. Hmel1 and PANC-1 cells showed much more CRT exposure than HBL cells (Fig. 7A,B). A comment is necessary for PANC-1 cells: after 72 h, even though a strong increase in CRT was shown by FCM (Fig. 7B), the histogram in Fig. 7A indicated that a portion of the cell population seemed to have less CRT, maybe for the presence of a resistance mechanism. To investigate this hypothesis, the release of ATP was evaluated, and it was revealed that in Hmel1 cells, its release into the medium markedly increased as a function of time (Fig. 7C), while in PANC-1 cells, there was release only at 72 h after PALM treatment. HBL cells seemed to be the most resistant to the activation of ICD; in fact, neither the displacement of CRT on the membrane nor the release of ATP were evident at short time points after treatment, and only a 72 h of incubation with PALM did these cells show slightly increases of CRT on the membrane and the beginnings of ATP release.

Discussion
It is known that MM and PDAC are the cancer pathologies with the worst prognosis. The low survival of patients affected by these two diseases depends on several factors, such as diagnosis of cancer at advanced stages, presentation of metastatic lesions, inability to treat with surgery, and the low response to chemotherapeutic or radiotherapy treatments. In particular, the establishment of either innate or acquired resistance, which are due to a tumour microenvironment that reduces the penetration of drugs or ionizing radiations (i.e., the dense desmoplastic stroma of PDAC) 59 , is among the most accredited causes of patients exhibiting no response to conventional treatments such as chemotherapy, radiotherapy and new biological approaches 60,61 . When exploring new therapeutic strategies, particular attention should be paid to possible indirect activities such as the induction of ICD, along with the study of direct anticancer activity. When cancer cells are dying, they can induce the production and release of DAMPs that are responsible for the increase in visibility of the tumour by the immune system's cells that, in turn, initiate a response against the tumour.
The hypothesis of this work is to use a physical tool, cold plasma, to activate a medium (e.g., PALM) that might impact cell proliferation, as other authors have already demonstrated 27,62 , by inducing cell death via ICD. In the two cancer models, PDAC and MM, after preliminarily establishing the optimal conditions for preparing PALM, we characterized the antiproliferative effectiveness of PALM together with the mechanisms of cell death induced and the possibility of activating ICD. www.nature.com/scientificreports www.nature.com/scientificreports/ In particular, we observed that when air was used as feed, almost no effect occurred on cell viability, most likely due to the minimal presence of H 2 O 2 found in the cell culture medium and the high concentration of nitrites, the latter of which were completely absent in PALM prepared with O 2 as the gas feed. Nitrites, that can almost be considered products of the reaction of NO with other components present in the PALM, were not as effective as H 2 O 2 against the cancer lines investigated. Among the three different conditions investigated, the best was "O 2 3 min", which was used in all subsequent experiments. Discharges operated by increasing the treatment time up to 3 min dissipated more energy; therefore, an increase in H 2 O 2 was detected. The formation of long-living species such as H 2 O 2 can be explained by the short diffusion length of OH radicals in the liquid that leads to recombination reactions. As shown in the literature, more than 60% of the liquid phase H 2 O 2 was estimated to be generated without a direct plasma effect in the experiment under observation 63 . The OH radicals produced during the plasma process due to the presence of water vapour in the gas phase diffuse through the liquid, producing hydrogen peroxide with different chemical patterns, as shown below: It is known from radiation chemistry that in the presence of high levels of molecular oxygen, H 2 O 2 could also be formed by the recombination of hydroperoxyl radicals as follows: These results allow us to assess that active N-containing species, including nitrites, are produced in PALM obtained with air, and these species, in turn, can probably promote a protective effect towards cancer cells. In fact, as reported in the literature, in addition to H 2 O 2 , nitrites are also a potential therapeutic effect against cancer. Aqueous solutions of sodium nitrite are delivered intravenously or even orally as a therapy 64 . As is well documented in the pertinent literature, nitrites can be reduced to nitric oxide in acidic environment containing low concentrations of oxygen. In cell culture medium, an acidic environment does not develop due to the presence of a buffer that keeps the pH constant in the range 7.4-7.8 regardless of the plasma treatment implemented. Due to the organic components present in cell culture medium, and the presence of cells, when the liquids are kept in contact with the cells, the reactions in the liquid phase involving nitrites can be much more complex than those observed in vivo 65 .
In both the MM and PDAC models, the investigation of the antiproliferative activity of PALM (O 2 3 min) and the determination of the intracellular accumulation of ROS after 3 h of exposure to PALM showed a marked inhibition of cell proliferation, which was inversely correlated with the intracellular ROS concentration. Although we are aware that our model is only exploratory, based on the 3 cell lines tested, the data seem to indicate that the pancreatic cancer cell line PANC-1, which had the highest basal intracellular level of ROS and the lowest accumulation after treatment, was also less responsive to PALM treatment; however, in the MM cell lines, lower basal ROS values and a higher accumulation after PALM treatment correlated with a stronger response. These data are in agreement with the data in the literature, which showed that ROS cause a reduction in cell proliferation [66][67][68] .
Upon analysis of the cell death mechanism responsible for the anticancer activity of PALM, we found that PANC-1 cells died during apoptosis, while MM cell lines mainly underwent autophagy. These data are in agreement with those previously reported, underscoring the fact that the MM cell lines are intrinsically autophagic and that this autophagy is higher in the presence of BRAF mutations, as evident in Hmel1 cells 56 . However, it is known that ROS can kill cells through either apoptosis or autophagy [69][70][71] , but, it was demonstrated for the first time that PALM might inhibit proliferation through various cell death mechanisms.
Furthermore, we investigated whether this new therapeutic approach may also induce ICD activation because the two tumour models chosen are profoundly different in their response to immunotherapy.
Immune checkpoint inhibitor treatment is the definitive gold standard for MM patients carrying the BRAF V600 mutation who no longer respond to BRAF or MEK inhibitors, or have wild-type BRAF. In PDAC, however, this therapeutic approach was not successful, especially when immune checkpoint inhibitors are used as a monotherapy, except in the case of a rare subset of tumours harbouring microsatellite instability (<2%). The poor immunogenicity of PDAC is due to the abundant stroma; this makes this tumour difficult to treat, and reduce the ability to recruit immunosuppressive cells through cancer-associated-fibroblast activation and transforming growth factor β secretion 72,73 .
Consequently, our study, which was focused on ICD, reveals that PALM can cause a type of cell death that induces activation of the innate immune system through the release of DAMPs, and thus can ultimately provoke an all-encompassing antitumour immune response in tumour pathologies that are less responsive to classical immunotherapy 43,[74][75][76] .
We demonstrated that both MM and PDAC tumour models, when exposed to PALM, showed an increase in CRT exposure on the plasma membrane and elevated ATP release, two crucial DAMPs for ICD; the intensities of the cells' responses were different. As far as the possible mechanisms responsible for the induction of ICD, as it relates to the well-known link between oxidative stress through increased intracellular ROS levels and the induction of ICD 10,77 , autophagy and apoptosis could also be involved. In our experimental conditions, cells were incubated with PALM at 1 h after its production, and in consideration of future clinical applications, it is not possible to prepare PALM and immediately administer it to the patient. Thus, even if we agree with other authors that ROS are involved in ICD, our suggestion is that its induction depends on long-lived species (H 2 O 2 , NO, etc.) rather than short-lived reactive species (•OH, etc). Autophagy promotes ICD and antitumour immune www.nature.com/scientificreports www.nature.com/scientificreports/ responses through the regulation of ATP release, while it has no impact on the surface exposure of CRT [78][79][80] . In fact, in Hmel1 cells, which showed the strongest PALM-induced autophagy response, a higher levels of ATP release were found. Regarding CRT, we found that the highest amount of membrane exposure was in the PDAC model, in which cells died due to apoptosis induction. This evidence agrees with reports in the literature that state that the cytosolic concentration of CRT increases under apoptotic stress conditions 81,82 . Furthermore, despite the differences in the PALM preparation protocols, our results are in agreement with those reported by Liedtke et al., who demonstrated that plasma jet-treated medium increased the exposure of CRT on tumour cells in a murine model of peritoneal spread of pancreatic cancer 46 .
Thus, our data suggest that, in MM and PDAC, treatment with PALM activates ICD; therefore, if this system is properly validated in larger cell line panels and in vivo models, PALM produced with O 2 , could serve as an ICD inducer that can activate the innate immune system to target these tumours.
In fact, it is known that DAMPs, including toll-like receptors (TLRs), can be recognized by receptors on macrophages and dendritic cells, and initiate a broad immune response through a well-defined mechanism, resulting in the increased presentation of cell-associated antigens to CD4+ and CD8+ T lymphocytes 43 .
Moreover, to harness the immune system to preserve survival and increase the efficacy of immunotherapy in low immunogenic cancers, the combination of PALM, as a potential ICD inducer, and immunotherapy (e.g., anti-CD47 antibody and anti-CTLA-4) may be suitable to improve therapeutic outcomes, as suggested by the literature [83][84][85][86] . Indeed, this combination might result in a bidirectional synergy between classical immunotherapy, which is efficient for patients with "T cell-hot" tumours, and PALM-dependent induction of ICD, which could facilitate an initial T cell-driven anticancer immune response potentially sustained by the presence of immune checkpoint blockers 85,87 .

Conclusion
Most plasma medicine papers dealing with cancer treatments are mainly focused on the duration of the direct plasma treatment rather than on other plasma parameters as a means of showing how to improve the efficacy of the treatment. To the best of our knowledge, our paper is the first describing the indirect treatment approach mediated by PALM treated with different gas feeds in a closed DBD system to distinguish the effects of ROS from those of RNS on tumour cells. We used only O 2 in the feed and a sealed plasma chamber to avoid contamination of external air and limit the formation of RNS in PALM. Moreover, for prospective work, our study should strengthen the knowledge base on the activity of ICD inducers and should serve as a platform for the identification of reliable biomarkers for monitoring the induction of ICD by PALM, which could be used in the selection of patients for appropriate treatments.