Overview of Prostaglandin E2 (PGE2)-Targeting Radiolabelled Imaging Probes from Preclinical Perspective: Lessons Learned and Road Ahead

As malignancies still represent one of the major health concerns worldwide, early tumor identification is among the priorities of today’s science. Given the strong association between cyclooxygenase-2 (COX-2)/prostaglandin E2 (PGE2), PGE2 receptors (EPs), and carcinogenesis, target-specific molecules directed towards the components of the COX2/PGE2/EP axis seem to be promising imaging probes in the diagnostics of PGE2pos. neoplasms and in the design of anti-cancer drugs. Featured with outstanding inclusion forming capability, β-cyclodextrins (CDs) including randomly methylated β-CD (RAMEB) were reported to complex with PGE2. Therefore, radiolabelled β-CDs could be valuable vectors in the molecular imaging of PGE2-related tumorigenesis. In vivo preclinical small animal model systems applying positron emission tomography (PET) ensure a well-suited scenario for the assessment of PGE2-affine labelled CD derivatives. Previous translational studies dealt with the evaluation of the tumor-homing capability of Gallium-68 (68Ga) and Bismuth-205/206 (205/206Bi)-appended β-CD compounds conjugated with chelator NODAGA or DOTAGA: [68Ga]Ga-NODAGA-2-hydroxypropyl-β-cyclodextrin/HPBCD, [68Ga]Ga-NODAGA-RAMEB, [68Ga]Ga-DOTAGA-RAMEB, and [205/206Bi]Bi-DOTAGA-RAMEB in experimental tumors with different PGE2 expression. These imaging probes project the establishment of tailor-made PET diagnostics of PGE2pos. malignancies. In the present review, we provide a detailed overview of the in vivo investigations of radiolabelled PGE2-directed CDs, highlighting the importance of the integration of translational discoveries into routine clinical usage.

Membrane-attached, inducible COX-2 is the key rate-limiting enzyme in the conversion of arachidonic acid into prostaglandins [1]. Literature data indicate that COX-2 overexpression seems critical in cancer initiation, tumor progression, and tumor maintenance [2]. Further, prior studies reported COX-2 upregulation in different cancer types, such as glioblastoma, colon, breast, prostate, and urinary bladder cancer [3][4][5][6][7]. Hence, COX2 specific inhibitors (COXibs) or non-steroidal anti-inflammatory drugs (NSAIDs) that block the activity of COX2 could contribute to the mitigation of cancer-linked mortality [8]. For example, Harris et al. published that celecoxib and rofecoxib effectively reduced the risk of development of colon cancer in a study conducted with the enrolment of 326 patients Identically to COX2, there is a notable interplay between the excess production of PGE2 and tumorigenesis [15]. PGE2 acting via G-protein coupled rhodopsin-type prostanoid receptors E 1-4 (EP1, EP2, EP3, EP4) triggers various tumor-associated intracellular signaling pathways, out of which the most remarkable ones are the stimulation of the epidermal growth factor-receptor (EGFR), the phosphorylation of protein kinase C (PKC)-mediated extracellular signal-regulated kinase (ERK), and the activation of G-protein dependent beta-catenin [1,2,12,16,17]. The PGE2 regulated activation of EP receptor induces tumorrelated angiogenesis along with the enhancement of tumor cell invasion and the halting of natural immune responses [12]. Further, PGE2-generated tumor proliferation as well as apoptosis inhibition lead to tumor volume expansion [12]. Former translational studies registered moderate tumor propagation in EP2-knockout mice in comparison with the wild-type ones [18]. Recent research indicating a link between 15-PGDH down-regulation and the appearance of certain malignancies including lung and transitional bladder cancers may project the role of this degrading enzyme in carcinogenesis [19,20]. Moreover, the metastatic spread of triple negative breast cancer was promoted by MRP4 [21]. In addition, Tong and colleagues claimed that PGE2 may serve as a biomarker in different malignancies such as pancreatic, breast, oral, or renal cancer [1].
Target-specific molecules directed towards the components of the COX2/PGE2/EP axis seem to be promising imaging probes in the early diagnostics of PGE2pos. neoplastic alterations as well as in the design of anti-cancer drug candidates. Nuclear medical in vivo preclinical model systems applying target-selective radiolabelled compounds may aid to further understand the molecular pathways of COX2/PGE2 that would attribute to the identification of well-suited diagnostic and anti-tumor biomarkers.
In the present review we provide a detailed overview about PGE2-specific radiolabelled diagnostic probes and their potential therapeutic applications in PGE2 overexpressing malignancies.

PGE2 and Cyclodextrins (CDs)
Cyclodextrins (CDs) are arising as powerful diagnostic agents in the field of nuclear medicine. Built up by D-(+)-glucopyranose molecules, these truncated cone-shaped oligosaccharides have a hydrophobic inner cage and a hydrophilic outer surface [22][23][24][25]. Given their advantageous physicochemical characteristic features, CDs are highly valuable for forming inclusion complexes with a myriad of molecules including PGE2. The complex forming capability of β-CDs with PGE2 was already established by Hirayama et al. in 1984 [26]. Exploiting the beneficial biochemical characteristics of β-CDs, Hirayama and co-workers managed to obtain the stabilization of PGE2 in aqueous solution with the application of the following two modified β-CD derivatives: heptakis (2, 6-di-O-methyl)-β-CD (DM-β-CD) and heptakis (2, 3, 6-tri-O-methyl)-β-CD (TM-β-CD). Compared to TM-β-CD, DM-β-CD appeared to be more effective in the maintenance of the stability of PGE2 projecting its feasibility in the proposal of pharmaceutical dosage formulas [26].
Besides the above-mentioned CD compounds, the complexation of another β-CD derivative-randomly methylated β-cyclodextrin (RAMEB)-with PGE2 was exhaustively evaluated as well. In a recent in vitro study focusing on the interaction between RAMEB and various proinflammatory molecules such as PGE2, substance P (SP), bradykinin (BRK), and calcitonin gene-related peptide (CGRP), Sauer and co-workers observed that PGE2 displays more significant binding potential to RAMEB in comparison with the other investigated agents [27]. To indicate the best fitting β-CD guest molecule, the Molecular Operating Environment docking algorithm-based molecular docking of SP, CGRP, BRK, and PGE2 was carried out. While RAMEB-affine PGE2 (∆Gbind = −12.57 kcal/moL) was characterised by high CD binding capability, BRK was moderately connected to RAMEB (−10.54 kcal/moL). Neither of the remaining assessed molecules showed affinity towards RAMEB. Earlier studies also focused on the evaluation of the complexation between CDs and other PGE subtypes, such as prostaglandin E1 (PGE1). Inoue et al. managed to obtain the enhanced stability of PGE1 derivative Limaprost/α-CD complex (Limaprost-alfadex) in humid circumstances by adding β-CD to the existing agent [28]. Later, they investigated the way of complexation between the PGE1 compound and the CD molecules, applying 1 H-and 13 C-Nuclear Magnetic Resonance (NMR) spectroscopy [29]. Referring to their results, a kind of ternary complex formation could be accountable for the stabilization of the drug. Overall, their findings also strengthened the strong association between CDs and PGE molecules.
Based on the above remarked former discoveries, PGE2-affine CDs appear to be valuable diagnostic vectors and drug carrier vehicles in PGE2-associated pathological processes, including carcinogenesis. Moreover, radiolabelled CDs complexed with PGE2 may establish the basis of target-specific diagnostics of PGE2 expressing malignancies. Applying radiometals with therapeutic radiation for labelling purposes, the theranostic potential of PGE2-directed CDs could also be tested ( Figure 1). Prior to the definitive integration of PGE2-affine CD-based imaging probes into routine clinical use, the need for the verification of their in vivo diagnostic applicability is warranted. Preclinical small animal model systems ensure appropriate circumstances for the assessment of the stability, the pharmacokinetics, the biodistribution, and the tumor-homing effectiveness of radiolabelled CD compounds.
Besides the above-mentioned CD compounds, the complexation of another β-CD derivative-randomly methylated β-cyclodextrin (RAMEB)-with PGE2 was exhaustively evaluated as well. In a recent in vitro study focusing on the interaction between RAMEB and various proinflammatory molecules such as PGE2, substance P (SP), bradykinin (BRK), and calcitonin gene-related peptide (CGRP), Sauer and co-workers observed that PGE2 displays more significant binding potential to RAMEB in comparison with the other investigated agents [27]. To indicate the best fitting β-CD guest molecule, the Molecular Operating Environment docking algorithm-based molecular docking of SP, CGRP, BRK, and PGE2 was carried out. While RAMEB-affine PGE2 (ΔGbind= −12.57 kcal/moL) was characterised by high CD binding capability, BRK was moderately connected to RAMEB (−10.54 kcal/moL). Neither of the remaining assessed molecules showed affinity towards RAMEB. Earlier studies also focused on the evaluation of the complexation between CDs and other PGE subtypes, such as prostaglandin E1 (PGE1). Inoue et al. managed to obtain the enhanced stability of PGE1 derivative Limaprost/α-CD complex (Limaprost-alfadex) in humid circumstances by adding β-CD to the existing agent [28]. Later, they investigated the way of complexation between the PGE1 compound and the CD molecules, applying 1 H-and 13 C-Nuclear Magnetic Resonance (NMR) spectroscopy [29]. Referring to their results, a kind of ternary complex formation could be accountable for the stabilization of the drug. Overall, their findings also strengthened the strong association between CDs and PGE molecules.
Based on the above remarked former discoveries, PGE2-affine CDs appear to be valuable diagnostic vectors and drug carrier vehicles in PGE2-associated pathological processes, including carcinogenesis. Moreover, radiolabelled CDs complexed with PGE2 may establish the basis of target-specific diagnostics of PGE2 expressing malignancies. Applying radiometals with therapeutic radiation for labelling purposes, the theranostic potential of PGE2-directed CDs could also be tested ( Figure 1). Prior to the definitive integration of PGE2-affine CD-based imaging probes into routine clinical use, the need for the verification of their in vivo diagnostic applicability is warranted. Preclinical small animal model systems ensure appropriate circumstances for the assessment of the stability, the pharmacokinetics, the biodistribution, and the tumor-homing effectiveness of radiolabelled CD compounds. Figure 1. Schematic overview of the in vivo diagnostics and radioisotope therapy of PGE2-expressing and PGE receptor-positive tumors using radiolabelled CDs. Applying radioisotopes with diagnostic (γ) and therapeutic radiation (α or β), radiolabelled CDs complexed with tumor cell derived PGE2 may establish the basis of target-specific diagnostics and personalized radionuclide therapy of PGE2-expressing malignancies. CD: cyclodextrin; PGE2: prostaglandin E2. Positron emission tomography (PET) is the mainstay diagnostic device in the detection of primary tumors and pertinent metastases. PET purveys data on in vivo physiological and pathological processes with the application of various targeted radiopharmaceuticals containing positron emitting radioisotopes [30]. The non-invasive characteristics and high sensitivity of PET imaging make it well-suited for the preclinical evaluation of tumorselective PGE2-specific molecules [31,32].

PGE2pos. Preclinical Tumor Models
BxPC-3 pancreatic adenocarcinoma cell lines are firmly associated with elevated PGE2 synthesis and EP2 receptor upregulation [36]. Investigating the following pancreatic cell lines in vitro: MiaPaCa-2, BxPC-3, PANC-1, and Capan-1, Takahashi and co-workers observed the expression of EP and various growth factor receptors by reverse transcriptasepolymerase chain reaction (RT-PCR). EIA-based (Enzyme immunoassay) determination of PGE2 levels and the presence of COX1 and COX2 mRNA were also evaluated. Applying EIA, ten-fold higher PGE2 concentrations were observed in BxPC-3 culture media relative to the other investigated specimens. Moderate EP2 mRNA presence and enhanced EP4 mRNA expression were identified in BxPC-3 cells. Further, comparing all investigated cells lines, the most remarkable COX-and EP4-reliant autocrine loop was manifested in the BxPC-3 culture.
The discrete presence of COX2 enzyme and related inconsiderable PGE2 synthesis of Panc-Tu-1 pancreatic ductal adenocarcinoma (PDAC) cells was reported by Gonnermann et al. [37]. Four PDAC cells lines including PancTu-1, Pt45P1, Panc89, and Colo357 were analysed in their study. As part of the comparison analyses of the PDAC cell cultures, disproportions between the appearance of COX2 and COX2-generated PGE2 synthesis were determined. In accordance with the PGE2 levels, in PancTu-1 and in Panc89 cells both flow-cytometry-and Western Blot-based measurements revealed moderate COX2 presence in comparison with receptor upregulated Colo357 cell lines. Less than 0.5 ng/mL PGE2 was produced by PancTu-1 and Panc89 cells, while Colo357-associated synthesis exceeded 6 ng/mL. In a like manner, Szabó et al. also confirmed the low EP2 receptorial expression of PancTu-1 cells during the immunohistochemical verification of the presence of EP2 receptor in several cell lines [35].
To our best knowledge, no exact literature data is available regarding the EP2 receptor status of HT1080 cells; however, according to the results of some earlier studies, the existence of these receptors and pertinent PGE2 production may be concluded. Anti-Prostaglandin E Receptor EP2 antibody immunohistochemistry with 3,3-diaminobenzidine (DAB) chromogens was employed to detect the appearance of EP2 receptor on HT1080 cells [35]. Histological staining revealed low amount of membrane EP2 receptors on the cell surface of HT1080 cell lines. Pifithrin-α (PFT-α)-induced COX2 level enhancement experienced in HT1080 cells with p53 wild-type suggests the presence of both PGE2 and its receptors in fibrosarcoma cell lines [38]. P-53 transactivation inhibitor PTF-α triggers COX2 overexpression via mitogen-activated protein kinase (MAPK) kinase (MEK)/extracellular signal-regulated kinase (ERK)-based molecular mechanisms. In a bid to investigate PFT-αgenerated COX2 overexpression, wild-type p53 HT1080 fibrosarcoma cells were administered with 40 µmol/L PFT-α. COX2 presence was authenticated applying Western blot analyses. Compared to the control, HT1080 cells exhibited 5.3-fold higher COX2 presence after PFT-α application.
B-cell-associated immunoglobulin production is regulated by PGE2 in immunologicalallergic processes [40]. Of note, 9 times higher serum IgE concentrations and much elevated monocyte-driven PGE2 synthesis characterised the patients suffering from progressive Hodgkin's disease (HD). Previous literature data confirm the PGE2-related EP receptor positivity of B lymphocytes [41][42][43]. Further, in vivo investigation of the receptor profile of mouse B lymphocytes revealed the presence of type 2 and type 4 EP receptors on the surface of the cell membrane (EP2, EP4) [44]. Taking the B-cell origin of HD into consideration, these findings may imply the strong EP receptor expression of B cells and different B cell lines.
More recently, chemically-induced, rat-originated syngeneic Ne/De (rat mesoblastic nephroma) and He/De (rat hepatocellular carcinoma) tumors and their metastases also proved to be PGE2pos. during preclinical PET imaging studies [35].

Insight into Preclinical In Vivo Nuclear Medical Studies
Driven by prior literature findings and the radiochemical results of Hajdu and coworkers, Trencsényi et al. proposed a novel PGE2-selective radiotracer-[ 68 Ga]Ga-NODAGA-RAMEB-to assess its tumor-targeting competence as well as its organ distribution at the preclinical level, applying PET [33]. To accomplish in vivo and ex vivo experiments, PGE2 overexpressing BxPC-3 human pancreas tumors and PancTu-1 human pancreas adenocarcinoma with low receptorial expression were subcutaneously (sc.) generated in CB17 severe combined immunodeficient (SCID) mice. Based on the receptor profile of the selected tumors, they were appropriate for the conduction of this research. Former evidence implied excess production of PGE2 in BxPC-3 cells [45]. In reference to flow cytometry and Western Blot analysis results, Gonnermann et al. observed insignificant COX2 presence and PGE2 synthesis in Panc-Tu-1 cell lines [37]. In a similar vein, Trencsényi and his team verified marked EP2 receptor expression in BxPc-3 tumors, while PancTu-1 pancreatic neoplasms were depicted with moderate receptorial presence using immunohistochemical staining ( Figure 3A). With the application of MiniPET-II small animal PET scanner, postinjection of 7.3 ± 0.3 MBq [ 68 Ga]Ga-NODAGA-RAMEB through the lateral tail vein, in vivo static and dynamic PET acquisition took place under inhalation anaesthesia in both control and in tumor-bearing (BxPC-3 and PancTu-1) small animals to analyse in vivo tracer distribution. For the objective evaluation of the tracer accretion of the tumor, different organs, and tissues, Trencsényi and co-workers determined the subsequent quantitative parameters: standardized uptake value (SUV) and tumor-to-muscle ratio (T/M). In the context of the ex vivo measurements, 30, 60, and 90 min-followed by the iv. application of 7.31 ± 0.32 MBq [ 68 Ga]Ga-NODAGA-RAMEB-tumor naïve and tumorous animals were put into sleep to remove three tissue samples for radioactivity (presented in %ID/g tissue) determination. Regarding healthy mice, 90 min post-administration, quantitative SUV mean figures-corresponding to the qualitative assessment-revealed low tracer uptake in the lung (SUV mean : 0.31 ± 0.07), heart (SUV mean : 0.12 ± 0.04), intestines (SUV mean : 0.13 ± 0.03), liver (SUV mean : 0.16 ± 0.06), stomach (SUV mean : 0.12 ± 0.04), and in the brain (SUV mean : 0.12 ± 0.04), whereas, due to the renal excretion, elevated urine radioactivity was registered. The in vivo observed urinary route of elimination was in accordance with both the ex vivo data and the results of Hajdu et al. with [ 68 Ga]Ga-NODAGA-HPBCD [31,33]. Referring to the outcomes of other prior research studies, quick renal clearance was disclosed both at clinical and preclinical levels using either radio-appended or non-radiolabelled HPBCD [31,46,47]. The quantitative in vivo parameters corresponded to the ex vivo ones measured by a calibrated gamma counter (Perkin-Elmer Packard Cobra, Waltham, MA, USA). Although no disparities were recorded in case of the tracer accumulation of the urine, kidneys, intestines, fat tissue, lung, and brain 30, 60, and 90 min postinjection, notable unevenness was disclosed between the 30 min and the 90 min %ID/g figures of the blood, small intestine, stomach, muscle tissue, heart, bone, salivary glands, gall bladder, and the pancreas (p ≤ 0.01). Of note, enhanced [ 68 Ga]Ga-NODAGA-RAMEB uptake of the lungs could be explained by the prolonged presence of the tracer in the watery pulmonary areas (%ID/g: 0.28 ± 0.08, 0.21 ± 0.04, 0.14 ± 0.03, 30, 60, and 90 min postinjection, respectively). Comparing the ID%/g and SUV values of [ 68 Ga]Ga-NODAGA-RAMEB and [ 68 Ga]Ga-NODAGA-HPBCD, [ 68 Ga]Ga-NODAGA-RAMEB accumulation was more moderate in the thoracic and abdominal organs [31,33]. The assessment of the in vivo PET images of the BxPC-3 tumor-bearing mice revealed that-owing to the reduction in the background activity-the contrast of the tumors was directly proportional to the length of the incubation time (T/M SUV mean : 7.80 ± 1.64 and T/M SUV mean : 18.57 ± 2.64, 10 and 90 min after the tracer injection, respectively). Thanks to this noteworthy T/M ratio (highest at 80-90 min post-radiopharmaceutical administration), PET images with superior quality/contrast could be ensured in PGE2 upregulated BxPC-3 tumorous experimental animals. Supported by the 3-14-fold lower SUV mean , SUV max , T/M SUV mean , and T/M SUV max values of 0.04 ± 0.01, 0.08 ± 0.02, 1.33 ± 0.19 and 1.66 ± 0.22, respectively, PancTu-1 tumors with discrete receptor expression could only be characterised by faint Initiated by the findings of Trencsényi and colleagues, Csige et al. presented another PGE2-affine RAMEB compound with potential future diagnostic as well as therapeutic applications [34]. In this study the same β-CD-RAMEB-was conjugated with chelator DOTAGA and radiolabelled with both 68 Ga and Bismuth-205/206 ( 205/206 Bi; the surrogate of 213-Bismuth/ 213 Bi) [34]. Taking the perfect complexation characteristics and kinetic inertness of DOTAGA into account, it seemed ideal to be attached to the amino group of NH2-RAMEB [48,49]. The whole process of the biosynthesis as well as the in vivo distribution and pharmacokinetic profile of both 68 Ga and 205/206 Bi-appended DOTAGA-RAMEB were outlined in detail. For this purpose, in vivo PET examinations and ex vivo biodistribution experiments were brought about, applying CB17 SCID male PGE2 receptor positive BxPC-3 tumor-bearing mice of 12-weeks of age. The enhanced EP2 receptorial expression of the BxPC-3 tumors was immunohistochemically strengthened in the cell plasm and in the cell membrane as well. Approximately 12 days post tumor cell inoculation, at a tumor volume of 98 ± 5 mm 3 , followed by the iv. injection of 6.4 ± 0.3 MBq [ 68 Ga]Ga-DOTAGA-RAMEB, in vivo static and dynamic PET imaging was performed. Healthy experimental mice-as control cohort-also underwent PET acquisition. As part of quantitative PET data analysis, SUV values and T/M ratios were registered. In line with the hydrophilic nature of [ 68 Ga]Ga-DOTAGA-RAMEB (LogP:-3.5), markedly enhanced tracer uptake was pinpointed in the kidneys and in the urinary bladder of the healthy control mice. In a like manner-as remarked above-Trencsényi et al. also observed renal excretion in case of [ 68 Ga]Ga-NODAGA-RAMEB [33]. In contrast, the qualitative PET image assessment of the healthy mice revealed slight radiopharmaceutical accumulation in the thoracic (lungs, heart) and abdominal organs (intestines, liver, stomach). These results of Csige et al. correspond to those of Trencsényi and co-workers [33,34]. Consequently, we can conclude that the change in the chelator did not influence tracer uptake. In addition, this newly synthesized 68 Ga-labelled, DOTAGA-chelated PET probe made the visualization of BxPC-3 tumors with PGE2 upregulation possible with the highest T/M ratios defined at 90 min post tracer application (2.5 ± 0.2) ( Figure 3B). This way the PGE2pos. tumortargeting potential of radiolabelled RAMEB-already established by Trencsényi et al.-was further authenticated. Identically, Trencsényi et al. also noted the most elevated T/M ratios 80-90 min following iv. tracer administration [33]. This finding further strengthens that the PGE2-binding affinity of 68 Ga-labelled RAMEB is regardless of the applied chelator.  We suppose that the variability in EP2 receptor expression and PGE2 levels between the HT1080, A20, and B16-F10 tumors may explain why the concentration of the two molecular probes considerably differed. Upon PET image analyses, the extent of PGE2 positivity exceeded that registered on the histological specimens. Szabó et al. hypothesized that alterations in the receptor status during tumor development as well as the binding of the CD probes to both the EP2 receptor and the PGE2 itself lead to more outstanding receptor positivity on the PET scans.
Tumor cells developing under the left renal capsule of the experimental animals give metastasis to the parathymic lymph node (PTLN) of the thorax. Although the primary Ne/De tumors and their PTLN metastases accumulated [ 18 F]FDG and both 68 Ga-labelled agents (Figure 4), [ 18 F]FDG uptake was two times higher compared to the accumulation of the CD derivatives (p ≤ 0.01). In addition, the primary Ne/De tumors were visualized using both 68 Ga-labelled imaging probes; however, more increased SUV values were detected in case of the HPBCD derivative (SUV mean : 3. 52    Comparing the radiotracer accretion of the SRCA-induced and the sc. transplanted Ne/De tumors, more increased accumulation of the radiolabelled CDs was depicted in the case of the subrenally growing Ne/De tumors compared to the sc. developing ones. This could probably be attributed to the orthotopic subrenal biological media that favours tumor propagation and metastatic spread [50]. In accordance with this finding, Máté and co-workers experienced the same during the comparison of the presence of neoangiogenesisrelated molecules between SRCA-generated and sc. transplanted Ne/De tumor models [51]. Gamma counter-based ex vivo radioactivity measurements-in line with the in vivo datademonstrated that PGE2 enriched BxPC-3, A20, Ne/De and He/De tumors exhibited the most elevated uptakes of both 68 Ga-labelled CDs. Although, Szabó et al. remarked a notable difference between the ex vivo tracer accumulation of the SRCA-induced Ne/De neoplasms and the sc. growing Ne/De tumors. In summary, the research findings of Szabó and colleagues further confirm that the place for radiolabelled CDs-RAMEB and HPBCD-is warranted in the molecular diagnostics of PGE2 expressing tumors.

Closing Remarks
Although recent years have witnessed milestone advances in battling against cancer, there are still several uncovered fields in tumorigenesis that remain to be better elucidated. Therefore, growing interest has been placed upon the investigation of tumor formation and progression at a molecular level. A broad set of experiments has intensively evaluated the role of RAS oncogene-the member of small GTPases-in cancer development [52]. GTP-activated H/N/K-RAS stimulates a large number of down-stream signalling pathways including Raf/MAPK kinase, MEK/ERKs, and PI3K/Akt that exert regulatory effects on cell proliferation, survival, apoptosis, and differentiation [52][53][54][55]. RAS mutations with related steady oncogenic activity can be encountered in a wide range of malignancies [53,56]. Beyond RAS-mediated signalling actions, COX2 and COX2-driven PGE2 production also have central contribution to neoplastic transformation. Based on literature data, the moderation of cancer-linked COX2 expression and COX2-triggered PGE2 synthesis is strongly associated with the activity of RAS oncoprotein [57][58][59]. On the other hand, COX2-dependent PGE2 production upregulates RAS signalling [53].
Taking the outstanding role of COX2/PGE2/RAS feedback circuit in carcinogenesis into account, uncovering the entire molecular pathway may favour our understanding of the underlying mechanisms of tumor development and metastatic spread. Broadened knowledge on the functioning of the COX2/PGE2/RAS axis might be for the benefit of the proposal of novel diagnostic molecular probes as well as anti-tumor drug candidates that could lead to the ultimate goal of the establishment of tailor-made cancer treatment.