Enhancing therapeutic efficacy in triple‐negative breast cancer and melanoma: synergistic effects of modulated electro‐hyperthermia (mEHT) with NSAIDs especially COX‐2 inhibition in in vivo models

Triple‐negative breast cancer (TNBC) is a leading cause of cancer mortality and lacks modern therapy options. Modulated electro‐hyperthermia (mEHT) is an adjuvant therapy with demonstrated clinical efficacy for the treatment of various cancer types. In this study, we report that mEHT monotherapy stimulated interleukin‐1 beta (IL‐1β) and interleukin‐6 (IL‐6) expression, and consequently cyclooxygenase 2 (COX‐2), which may favor a cancer‐promoting tumor microenvironment. Thus, we combined mEHT with nonsteroid anti‐inflammatory drugs (NSAIDs): a nonselective aspirin, or the selective COX‐2 inhibitor SC236, in vivo. We demonstrate that NSAIDs synergistically increased the effect of mEHT in the 4T1 TNBC model. Moreover, the strongest tumor destruction ratio was observed in the combination SC236 + mEHT groups. Tumor damage was accompanied by a significant increase in cleaved caspase‐3, suggesting that apoptosis played an important role. IL‐1β and COX‐2 expression were significantly reduced by the combination therapies. In addition, a custom‐made nanostring panel demonstrated significant upregulation of genes participating in the formation of the extracellular matrix. Similarly, in the B16F10 melanoma model, mEHT and aspirin synergistically reduced the number of melanoma nodules in the lungs. In conclusion, mEHT combined with a selective COX‐2 inhibitor may offer a new therapeutic option in TNBC.


Introduction
Previously, we have demonstrated a strong antitumor effect of modulated electro-hyperthermia (mEHT) treatment in the 4T1 triple-negative breast cancer (TNBC) and in the B16F10 melanoma lung mouse models [1,2].
report that mEHT monotherapy in addition to the described reaction stimulates local inflammatory proteins such as interleukin-1 Beta (IL-1b), interleukin-6 (IL-6), and cyclooxygenase-2 (COX-2).The lAPR and the IL-1b, IL-6, and COX-2 induction can be considered as part of a self-defensive reaction of tumor cells due to mEHT-induced stress, although the role of the inflammation in cancer is not fully understood [4][5][6][7].In the present in vivo study, mEHT was combined with nonsteroid anti-inflammatory drugs (NSAIDs; aspirin (ASA) or SC236), which target multiple components of the acute phase reaction [8], and alone have already demonstrated promising outcomes in various tumors [9].Here, we demonstrated that NSAID treatment synergistically increased the antitumor effect of mEHT in both 4T1 TNBC and B16F10 melanoma mouse models.Besides the inhibition of tumor growth, we have observed an increased tumor destruction ratio (TDR) in combined-treated tumors, which was accompanied by an increase in cleaved caspase-3 (cC3), suggesting that apoptosis plays an important role in the antitumor effect of mEHT and NSAIDs in vivo.To understand the molecular bases of synergism, we performed mRNA studies with RT-PCR and nanostring.Molecular studies revealed a statistically significant reduction in the IL-1b and COX-2 expression by the combination therapy.Analysis of differentially expressed genes (DEG) using Gene Ontology (GO) through the Database for Annotation Visualization and Integrated Discovery (DAVID).Our findings suggest that the synergistic combination of mEHT and a selective COX-2 inhibitor may regulate the extracellular matrix turnover and cell membrane protein expression.
Triple-negative breast cancer and melanoma are aggressive, and recurring, with early metastasis compared with most other cancer types.TNBC accounts for 15% of all types of breast cancer.TNBC cells do not express the estrogen receptor, the progesterone receptor, or for the human epidermal growth factor receptor 2; consequently, there are no tumor-specific (such as targeted or hormonal) treatment options, at present [10].Contemporary treatments for melanoma include surgical resection, chemotherapy, and radiotherapy [11].Thus, adjuvant treatments such as mEHT may aid TNBC and melanoma treatment [12,13].
Modulated electro-hyperthermia represents an adjuvant or supplementary treatment in different cancer models that are recently utilized in clinical medical practice [14].mEHT contributes to tumor-cell destruction selectively by accumulation of the radiation energy within the tumor without affecting healthy neighboring tissues [15].Specific energy accumulation by the tumor cells is partially due to the altered (anaerobic) glucose metabolism of tumor cells (independent of the availability of oxygen called 'Warburg effect') [16].
Numerous epidemiological and experimental studies show that NSAIDs reduce the risk, incidence, and mortality in some cancers [17], including melanoma [18] and breast cancer [19].Both nonselective (NSAIDs) and selective COX inhibitors (selCOXIBs) have been associated with lower cancer incidence; however, selCOXIBs have demonstrated greater significance than nonselective NSAIDs [20].NSAIDs inhibit the production of prostaglandins, proinflammatory cytokines, and tumor growth factors [21,22].Furthermore, NSAIDs may affect COX-independent inflammatory pathways such as NF-jB [23], MAPK [24], mTOR [25], PDK-1/Akt [26], and Wnt/b-catenin [27], by the inhibition of activation of transcription factors.These pathways support cell proliferation and angiogenesis but suppress apoptosis, thus supporting tumor growth as well as participate in the regulation of the tumor microenvironment (TME).Therefore, COX inhibition influences tumor progression by decreasing migration [28], metastasis [29,30], angiogenesis [31], increasing apoptosis, and sensitivity to other conventional anticancer therapies such as chemotherapy, immunotherapy, or radiotherapy [32,33].Besides the novel strategy of potential anticancer therapy, NSAIDs have a few adverse effects, such as gastrointestinal and cardiovascular complications as well as increased risk of renal or hepatic injury [34].
In this hypothesis, it is suggested that NSAIDs may regulate the TME and mEHT-induced proinflammatory cytokines: IL-1b, IL-6, and COX-2.As a result, NSAIDs may enhance the mEHT-induced tumor cell death.Combining clinically available mEHT with NSAIDs is a new potential tool in oncologic therapy.

In vivo treatment of 4T1 TNBC
4T1 TNBC cells were cultured and prepared for inoculation.Using a Hamilton syringe (Hamilton Company, Reno, NV, USA), 1 9 10 6 cells/50 L PBS (phosphatebuffered saline with no magnesium and calcium, #17-516F, Lonza A. G., Basel, Switzerland) were inoculated in the fourth mammary fat pad of 6-to 8-weekold female BALB/c mice.Female BALB/c mice were ordered from the National Institute of Oncology (Hungary) and were housed under minimal disease (MD) conditions at the Animal Facility of the Basic Medical Science Center of Semmelweis University with free access to standard mouse chow and tap water ad libitum and under a 12 h dark/light cycle.Animals were anesthetized with isoflurane (Baxter International Inc., Deerfield, IL, USA), 4-5% for induction, 1.5-2% to maintain anesthesia, and compressed air (0.4-0.6 L/min) for tumor-cell injection.4T1 tumor cells were subcutaneously injected into each mouse's inguinal mammary fat pad.On the eighth day after inoculation, tumor volume was measured using ultrasound and a digital caliper, as described earlier by Danics et al. [35].Mice were randomized according to tumor volume into six different groups.On the same day, animals were given a daily dose of aspirin 100 mgÁkg À1 (Sigma-Aldrich Co., St. Louis, MO, USA) or SC236 6 mgÁkg À1 (Axon Medchem BV, Groningen, The Netherlands.) via intraperitoneal injections [29].Drugs were dissolved in 10% DMSO, 40% PEG300, 5% Tween-80, and 45% saline.Treatment with COX inhibitors, administered every day during the entire experiment, and was combined with mEHT (Fig. 1A).Mice were treated four times every 48 h with a newly constructed labEHY-200 (Oncotherm Ltd., P aty, Hungary) mEHT device, as detailed in our prior publications [1,35].Tumor volume was monitored by ultrasound and digital caliper every other day until the termination of the experiment.In two separate experiments, we investigated long-term effects of mEHT treatment.In the time-kinetic experiment (Fig. 1B), mice were treated with mEHT or Sham and tumors were harvested after 3 mEHT treatments at different time points.In the long-term follow-up experiment (Fig. 1C) tumors were harvested 96 h after two mEHT treatments.The mice were euthanized by cervical dislocation the next day after the last treatment.Tumors were excised, and they were cleaned for further processing.Half of the tumors were fixed in a 4% formaldehyde solution (Molar Chemicals Ltd., Hal asztelek, Hungary), and the other half was cut into a few small pieces and stored at À80 °C for molecular studies.Interventions and animal housing were carried out in accordance with Hungarian Law Nos.XXVIII (1998 and2002) and LXVII (2002), both of which deal with animal protection and welfare, as well as European Union guidelines.The National Scientific Ethical Committee on Animal Experimentation approved all animal treatments under the code PE/EA/50-2/2019 on 1 November 2019.

2.3.
In vivo treatment of B16F10 melanoma 1 9 10 5 B16F10 melanoma cells were injected into the tail vein of 7-to 9-week-old female C57BL/6 mice that have induced tumor nodules in the lungs.The next day after inoculation, mice were treated with mEHT alone or mEHT combined with aspirin 11.1 mmolÁL À1 concentration [35] in drinking water [36].Because of aspirin insolubility in water, first, aspirin was dissolved in DMSO 0.2% and then mixed with the drinking water using a magnetic mixer to mix it until a clear solution was achieved.Solution pH was adjusted to 7.4 physiological pH using a combination of NaOH and/or HCl and a pH meter.Animals have been treated six times using the LabEHY-200 device set up to maintain 41-42 °C inside the mice's lungs [2].The first mEHT treatment was performed 1 day after the tail vein inoculation of the B16F10 cells for 30 min, which was repeated six times in total on every third day (Fig. 2).Animals were terminated on Day 20, 48 h after the last mEHT treatment.The lung melanoma burden was assessed by counting the number of tumor nodules on the surface of the lungs [2].

RNA isolation and real-time PCR
RNA was isolated using TRIzol reagent (Molecular Research Center Inc., Cincinnati, OH, USA) according to the manufacturer's instructions.A high-capacity cDNA reverse transcription kit was used to reverse transcribe the isolated RNA (Applied Biosystems, Carlsbad, CA, USA).The amplified cDNA was used as template for the RT-PCR.SYBR Green-based RT-PCR using Sso Advanced TM Universal SYBR Ò Green Supermix and the CFX96 Touch Real-Time PCR Detection System was used to detect messenger RNA in the samples (Bio-Rad, Hercules, CA, USA).
The 18S and GAPDH genes were used as housekeeping genes (Table 1).

Nanostring
Extracted RNA from in vivo experimental samples was used for RNA detection with NanoString technology.NanoString uses unique optical barcoded RNA that hybridizes to the target RNA in the sample to enable digital counting of individual RNA molecules without possible artifacts introduced by enzymatic steps.The gene expression panel (NanoString; Redwood, CA, USA) was custom-made based on our previous publication in which mEHT-regulated genes were detected by next-generation sequencing (NGS) and verified by NanoString.The custom panel composed of 134 genes [1].One hundred nanograms of RNA was used for hybridization.After hybridization, samples were transferred to the nCounter Prep Station for data collection on the nCounter Digital Analyzer.The 4.0 NSOLVER Analysis Software (NanoString, Redwood, CA, USA) was used for data analysis.Genes with log2 fold change values greater than 1.5 or less than À1.5 were considered the most regulated for further analysis.Values obtained from three replicates of two groups, mEHT or mEHT + SC236, were used to generate the volcano plot.DEGs were conducted utilizing the GO which was accessed through the DAVID [37].GO analysis was used to identify genes that can be classified into different groups.In our study, we used the database for DAVID (https://david.ncifcrf.gov/)to perform functional annotation clustering the most regulated genes.The P-value represents the probability of chance association between genes and a specific functional category.The P-value was adjusted with the Benjamini-Hochberg (shortly: Benjamini) procedure to control for false discovery rate (FDR) by correcting for multiple comparisons.

Statistical data analysis
The GRAPHPAD PRISM software (v.6.01;GraphPad Software, Inc., La Jolla, CA, USA) was used for statistical analysis.Unpaired Mann-Whitney nonparametric tests were performed to compare the Sham and the mEHT-treated groups.Follow-up examinations were statistically evaluated with a one-way ANOVA.Differences were considered statistically significant as *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0. 0001.Data are presented as mean AE SEM.

Ethics approval
The studies did not involve human participants, human data, or human tissue.

mEHT-inhibited tumor growth was accelerated by NSAID co-treatment
During the mEHT and NSAID combination therapy experiment (Fig. 5A), tumor volumes were assessed by ultrasound and digital caliper.Tumor volumes were similar at randomization: 81.8 AE 17 mm 3 (Fig. 5J,K).Tumor volumes progressed in Sham + vehicle-treated mice from 264 AE 67 mm 3 (after two treatments) to 320 AE 59 mm 3 (after three treatments) and to 413 AE 77 mm 3 (after four mEHT treatments) (Fig. 5B-G, J-K).Monotherapy mEHT (+vehicle) was able to significantly reduce tumor size only after 3 mEHT treatments (P = 0.06) (Fig. 5D,E).Average volume in the mEHT-treated group was 258 AE 54 mm 3 compared with 320 AE 59 mm 3 in the Sham-treated group.However, the combination therapy (mEHT + ASA/SC236) was able to significantly reduce tumor volume already after 2 mEHT treatments to 156 AE 51 mm 3 (P = 0.02) and 145 AE 49 mm 3 (P = 0.02) compared with the Sham-treated group with an average volume of 264 AE 67 mm 3 (Fig. 5B,C).After four treatments, only the COX-2-specific combination (mEHT + SC236) with 166 AE 54 mm 3 average volume was significantly more effective than mEHT monotherapy-average volume 285 AE 77 mm 3 (P = 0.02).The COX-2-specific combination (mEHT + SC236) was the most effective inhibitor of tumor growth (Fig. 5J,K).This observation was supported by significant reduction in the tumor weight at the end of the study (Fig. 5H,I).Average tumor weight in the mEHT-treated group was 284 AE 88 mg, while in the mEHT + SC236, it was only 175 AE 51 mg (P = 0.04).
Most mice lost not more than 5-10% bodyweight by the third treatment; however, they recovered by the end  of the study.The body weight loss was not significantly different between the groups.Sham-treated mice (20.1 AE 1 g) lost an average 0.5 grams (19.6 AE 0.6 g) by the third treatment.The mEHT-treated group (19.4 AE 1.5 g) lost an average 1.7 g (17.7 AE 1.6 g).Animals that received the combination treatment of mEHT and ASA (20.7 AE 0.7 g) lost only 1.2 g (19.5 AE 0.6 g), while those treated with mEHT + SC236 (20.7 AE 1 g) did not lose weight (Fig. 5L).Bodyweight by the end of the study did not differ significantly from the initial bodyweight in any of the groups.

Multiplex (nanostring) analysis demonstrated that COX-2 inhibition negatively correlated with tumor-promoting factors associated with tumor-cell membrane and extracellular matrix
Samples from the mEHT + NSAIDs experiment (Fig. 7A) passed the nanostring quality control (QC) and the nanostring run was successful.Seventyfour identified DEGs are displayed on the heat map (Fig. 7B).Most regulated genes are presented on the volcano plot as well (Fig. 7C).Genes were identified (Table 3) and clustered in two groups using DAVID.Seven genes were identified as membrane proteins and six genes as secreted proteins.The P-value established with the Benjamini procedure is also given on (Fig. 7D).
• Epithelial cell adhesion molecule (EPCAM) and ephrin type-B receptor 6 (EPHB6), which were identified by DAVID as membrane proteins.

Discussion
In the present study, we demonstrated that mEHT monotherapy induced inflammatory cytokines such as IL-6 and IL-1b.Inflammatory cytokines are a sign of more aggressive cancer phenotype [38,39].By creating an inflammatory microenvironment within the tumor, IL-6 and IL-1b may promote tumor-cell proliferation by activation of transcription factors and angiogenesis by the induction of key factors such as vascular endothelial growth factor (VEGF).Mutant cells generate cytokines and attract inflammatory cells [40].COX-2 has a strong relationship with IL-6 and IL-1b [41].Our data revealed a COX-2 induction 72-and 96-h after mEHT treatment, which may be explained by the earlier upregulation of IL-6 and IL-1b at 12 h and 24 h.Inflammatory COX-2 has been demonstrated to be induced by IL-1b [42] or IL-6 [43].IL-6 has been proposed as an important tumor-promoting cytokine [44].Similarly, to our results a recent paper demonstrated relatively late COX-2 induction at 24 h by chemotherapy in 4T1 cells [45].In that study, COX-2 inhibition was essential for effective combination therapy (chemo + immune therapy), similar to our study.Furthermore, previous studies demonstrated that other anticancer therapies, such as chemo-and photodynamic therapies have also induced significant IL-1b and COX-2 expression activating the tumorfavorable microenvironment.These studies have also utilized COX-2 inhibitors to overcome this problem.A combination of photodynamic therapy with selective COX-2 inhibitors effectively inhibited inflammatory cytokine and VEGF synthesis and enhanced therapy-induced cytotoxicity and apoptosis [46].Similarly, co-administration of COX-2 inhibitors with chemotherapy or ablation of COX-2 with CRISPR effectively enhanced the treatment efficacy by reducing production of inflammatory proteins [45].
We combined mEHT with aspirin or the selCOXIB SC-236 in TNBC 4T1 and B16F10 melanoma experimental mouse models.The 4T1 TNBC growth inhibitory effect of mEHT monotherapy was quite similar to our previously published studies, where mEHT alone effectively inhibited 4T1 TNBC growth after the third treatment [1,35].However, here the combination of mEHT with NSAIDs demonstrated significant tumor shrinkage earlier: already after the second treatment, as revealed by both ultrasound and a digital caliper measurements of tumor volume.The clinical relevance of achieving tumor reduction at an earlier stage of treatment is enormous as earlier inhibition of exponential tumor growth offers better prognosis.Shrinking the tumor early may render it more susceptible to subsequent treatments such as surgery, radiation, or chemotherapy.Thus, early tumor size reduction could potentially make these treatments more effective, leading to better outcomes [47].Early cancer treatment is also emphasized by the recent NHS cancer guideline [48].After the fourth treatment only the SC236 combination group reached statistical significance when compared to mEHT.A previous study demonstrated that in CT26 colon cancer, mEHT efficacy was significantly enhanced by curcumin, with known anti-inflammatory and antioxidant properties [49].
Neither mEHT nor the NSAIDs nor any combination appeared to be toxic based on the body weight data as animals did not lose more than 5-10% of their body weight [47].Assessing treatment toxicity in preclinical models is crucial to evaluate the safety for clinical translation [50].Based on the body surface area dose conversion method [51], 100 mgÁkg À1 ASA is equivalent to a moderate human ASA dose of 300-500 mgÁday À1 [28].Long-term ASA treatment can result in renal, cardiovascular, and gastric toxicity as well as bleeding and hypersensitivity [52].SC236 used in the present study is analogous to celecoxib.Celecoxib is associated with a decreased risk of gastrointestinal bleeding, although it still has a higher incidence of cardiovascular events in comparison with traditional NSAIDs [53].Although the applied NSAID dose + mEHT combination did not appear to have any toxicity, the toxicity findings of the present mouse study must be reevaluated in large animal models and early-phase clinical studies before reaching conclusions relevant for the clinical setting.
Besides the tumor size reduction in the NSAID combined groups, the TDR of the remining small tumors was also larger.However, in mEHT-monotherapy-treated tumors, only tumor size was smaller than in the Shamtreated tumors; the TDR did not differ significantly, similarly to our previous studies [1].The lack of significant increase in TDR is largely due to the severe spontaneous destruction of Sham tumors growing to a size where spontaneous necrosis is evident [54].
In addition, the tumor damage response was accompanied by a significant increase in cC3, suggesting that apoptosis played an important role in the antitumor effect of mEHT and NSAIDs [55].As cancer cells often evade apoptosis, inducing apoptosis in cancer cells is a critical goal of cancer treatment.The intrinsic resistance of cancer cells to apoptosis is the basis of resistance to chemotherapy [56].Thus, induction or reactivation of apoptosis in cancer sensitizes cancer cells to therapy [57].Monotherapy with both mEHT or COX-2 inhibitors induced significant cC3 staining in TNBC in previous [33,35] as well as in the present study.COX-2-and mEHT-induced apoptosis may have contributed to the synergistic effects of mEHT + NSAIDs [17,58].According to our RT-PCR data, NSAIDs attenuated mEHT-induced IL-1b and COX-2 expression.COX-2 and IL-1b are regulated together [8].Besides inflammatory proteins, NSAIDs may affect COX-independent inflammatory pathways, by the inhibition of transcription factor activation of pathways, which support cell proliferation and angiogenesis but suppress apoptosis [23][24][25][26].Therefore, COX inhibition induced apoptosis sensitizes cancer to other conventional anticancer therapies such as chemotherapy, immunotherapy, radiotherapy [32,33] or mEHT.
The results of the nanostring multiplex data analysis demonstrated that the combination of mEHT with a selective COX-2 inhibitor regulated secreted and cell membrane proteins.Some of the secreted proteins are extracellular matrix (ECM) proteins (EFEMP1, NPNT, and STC), ECM regulators (ADAMTS15, CXCL11, and FGFBP1) and thus are an essential part of the TME.The identified membrane proteins serve as receptors (EPHB6, ACKR1, and EDNRB) or adhesion molecules (EPCAM) and may enhance antitumor immunity (TENM2 and CLEC-10A); thus, they may contribute to enhanced mEHT-induced cancer destruction through regulating the TME related immune response.
ADAMTS15 was upregulated in mEHT+ selCoxIb (SC236) vs mEHT monotherapy.Its biological significance is its involvement in extracellular matrix remodeling [59,60].ADAMTS15 participates in tissue organization and vascular homeostasis.ADAMTS15 is a secreted protease modifying the extracellular matrix components like proteoglycans and collagen [61].ADAMTS15's role in tumor pathophysiology is inhibition of angiogenesis and cell migration [62].To the best of our knowledge, the regulation of ADAMTS15 by COX inhibition has not been described before.ADAMTS15 may be acting as a tumor suppressor in breast cancer by modulating cell-environment interactions [63,64].ADAMTS15 expression is a favorable prognostic factor in breast cancer.Higher expression has been associated with better clinical outcomes, including longer overall survival and disease-free survival [62].As ADAMTS15 high levels are signs of good prognosis in breast cancer [65].Thus, the observed upregulation could have contributed to the observed synergistic effects.
Similarly, secreted proteins, such as EFEMP1 and CXCL11, which are known targets of selCOXIBs [8], were also upregulated in the combination group.EFEMP1 is an extracellular matrix protein, associated with elastic fiber formation and cell adhesion [66] and is considered as a tumor-suppressor gene [67,68].Its expression is diminished in breast cancer [69].In tumor biology, antiangiogenic properties (reduction in angiogenic sprouting) of EFEMP1 have been described [70].Although contradictory data suggesting EFEMP1 to be a bad prognostic indicator have been also published [71], our data support the beneficial role of EFEMP1 in breast cancer.CXCL11 is involved in the immune response including T-cell and TME regulation.CXCL11 can enhance antitumor immune cell migration and infiltration in the breast cancer tissue [72].In our study, upregulation of EFEMP1 and CXCL11 proteins could have contributed to the synergistic effect between mEHT and the selCOXib therapy.
Additionally, downregulated secreted glycoproteins were FGFBP1, STC1, and NPNT.FGFBP1's physiologic role is regulating fibroblasts and some cellular processes, such as differentiation and growth.In cancer biology, FGFBP1 may stimulate angiogenesis, PD-L1 expression, and immune inhibition [73,74].STC1 is a secreted glycoprotein.Its biological significance is the regulation of phosphate and calcium homeostasis [75].In cancer, STC1 enhances metastasis via the PI-3K/Akt/NF-kB signaling pathways [76].Similar to our results, a study demonstrated that NSAIDs (Ibuprofen) attenuated IL-1b-induced STC1 expression in chondrocytes [77].NPNT is a secreted glycoprotein into the TNBC extracellular matrix and plays a role in adhesion and migration.In cancer, NPNT may promote metastasis via its integrin-binding site, which is important for adhesion and transmigration through the endothelial cells [78].Thus, downregulation of NPNT by NSAIDs may have contributed to the reduced tumor formation in our lung melanoma model.Regulation of NPNT and FGFBP1 by COX inhibition has not been described before.As described, FGFBP1, STC1, and NPNT proteins have tumor-promoting effect thus, downregulation of these proteins could have contributed to the synergistic effect in our study.
Membrane glycoproteins such as ACKR1, EDNRB, TENM2, and CLEC-10A were highly expressed in the selCoxIb-treated group.These proteins are considered as predictor biomarkers of good prognosis of various cancers.They may negatively regulate the tumorfavorable microenvironment and support the antitumor immunity in breast cancer, although their role in tumor pathophysiology is not yet completely understood [79][80][81][82].The regulation of these proteins by COX inhibition has not been described before.In our study, the upregulation of these proteins by selCoxIb probably contributed to the enhanced effects of mEHT in treated tumors.
Neurexin 3 (NRXN3) encodes a protein involved in synaptic signaling and neural development in the nervous system [83].Overlaps between genes involved in neural development and in cancer suggest links between nervous system development and tumorigenesis [84,85].Emerging research suggests a role of NRXN3 in cancer progression, including breast cancer [86], through its involvement in cellular adhesion, migration, and invasion.
In addition, membrane proteins, EPCAM, and EPHB6 gene expression were downregulated by sel-CoxIb treatment.Their biological role has been associated with cell adhesion and signaling.Thus, they may stimulate 4T1 cell growth; therefore, EPCAM and EPHB6 are associated with poor survival [87].Silencing EPCAM significantly decreased the capacity of TNBC cell proliferation [88].EPCAM expression is related to COX-2 expression, TME regulation, and angiogenesis [89].Downregulation of EPCAM and EPHB6 glycoproteins could contribute to the positive effects of the addition of selCoxIb to mEHT in 4T1 TNBC.
The listed regulated genes are mostly considered predictor biomarkers of prognosis in various cancers.The clinical implications of these genes are often contextdependent, and continuous research is essential to comprehend their roles in diseases and how they can be utilized for clinical purposes such as diagnosis, prognosis, and targeted therapies.
Furthermore, we investigated the combination therapy of mEHT with aspirin in the B16F10 melanoma model.Tumor nodules in the lungs were targeted with a new application method developed by Thomas [2].The pulmonary nodules were counted to evaluate the effect of the combination therapy.Our findings suggest that mEHT efficacy was enhanced by aspirin.Aspirin alone did not reduce the nodule count in the lungs.A Previous studies have demonstrated beneficial effects of aspirin in the B16F10 tumor model, by the induction of cC3-mediated apoptosis [30,36].The role of cC3-mediated apoptosis is supported by our study and cC3-mediated apoptosis may play a role in the synergistic tumor growth inhibition observed by us [34,35].
Limitations of the study include the lack of human data.Although mEHT has been extensively applied in the clinical practice in several countries, its combination with selective COX2 inhibitors or nonselective nonsteroidal anti-inflammatory drugs has not been tested before.Further investigation of the role of the proteins identified by nanostring as potential mediators of the synergy between mEHT and selective COX2 inhibition are potential avenues for future research.

Conclusion
We discovered that mEHT treatment stimulated the expression of proinflammatory cytokines IL1-b, IL-6, and COX-2 in TNBC.Our animal studies investigating the mEHT and NSAID combination therapy demonstrated a synergistic inhibition of tumor growth in both 4T1 TNBC and B16F10 melanoma cancer models.RT-PCR and nanostring analysis suggested that therapeutic combination attenuated the mEHT-induced IL1-b and COX-2 stimulation.In addition, selCoxibs may modulate the extracellular matrix, and cell membrane function in the TME leading to inhibition of cancer cell proliferation.Thus, this new combination treatment protocol could be implicated in the clinical setting as a therapeutic option due to its effectiveness and availability.Using the combination of mEHT with COX inhibitors may offer a new therapeutic possibility for various cancer types.

Fig. 1 .
Fig. 1.Experimental protocols of the 4T1 model.4T1 TNBC cells were inoculated at Day zero.Mice were randomized at Day 8. mEHT treatments were performed every other day.NSAIDs were administered every day from randomization until the end of the experiment.Ultrasound and caliper tumor volume measurements were performed on every intermittent day.Tumors were harvested at the end of the study.(A) Combination therapy: 49 mEHT treatment in combination with aspirin or SC23.(B) Time-kinetic experiment: 39 with mEHT treatment + tumor harvest at different time points.(C) Long-term follow-up: 2xmEHT treatment + tumor harvest at 96 h after the last treatment.

Fig. 2 .
Fig. 2. Experimental protocol of the B16F10 melanoma experiment.Randomization was performed at Day zero, B16F10 melanoma cells were inoculated through the tail vein at Day 1, mEHT treatments were performed at Days 2, 5, 8, 11, 14, and 17.Aspirin was administered every day from Day zero until the end of the experiment.Tumors were harvested at Day 19.
Interventions and housing of the animals conformed to the Hungarian Laws No. XXVIII/1998 and LXVII/2002 about the protection and welfare of animals, and the directives of the European Union.All animal procedures were approved by the National Scientific Ethical Committee on Animal Experimentation under the No. PE/ EA/50-2/2019.

Table 1 .
Primer pair designed for RT-PCR.

Table 2 .
Antibodies and conditions used for IHC.

Table 3 .
Absolute mRNA count of secreted and membrane proteins from the nanostring data.Individual data of the mEHT and mEHT + SC236 group members and group averages.Bold indicates downregulation, and nonbold indicates upregulation.Linear mRNA data were received through the nSolver advanced analysis method.Cell line 4T1.The underlined values are the average value of the 3 samples (as the label of the column).Avg., average shows.